No file available
To read the file of this research,
you can request a copy directly from the authors.
Label-free non-invasive classification of rice seeds using optical coherence tomography assisted with deep neural network
Preprints and early-stage research may not have been peer reviewed yet.
Abstract
Identification of the seed varieties is essential in the quality control and high yield crop growth. The existing methods of varietal identification rely primarily on visual examination and DNA fingerprinting. Although the pattern of DNA fingerprinting allows precise classification of seed varieties but fraught with challenges such as low rate of polymorphism amongst closely related species, destructive method of analysis and a huge cost involved in identification of robust markers such as simple sequence repeat (SSR) and single nucleotide poly-morphisms. Here, we propose a fast, non-contact and non-invasive technique, deep learning assisted optical coherence tomography (OCT) for subsurface imaging in order to distinguish different seed varieties. The volu-metric dataset of, (a) four rice varieties (PUSA Basmati 1, PUSA 1509, PUSA 44 and IR 64) and, (b) seven morphologically similar seeds of rice landrace Pokkali was acquired using OCT technique. A feedforward deep neural network is implemented for deep feature extraction and to classify the OCT images into their relevant classes. The proposed method provides the classification accuracy of 89.6% for the dataset of total 158,421 OCT images and 82.5% in classifying the dataset of total 56,301 OCT images collected from Pokkali seeds. The current technique can accurately classify seed varieties irrespective of the morphological similarities and can be adopted for the removal of varietal duplication and assessment of the purity of the seeds.
... OCT has been found to be applicable in various agricultural contexts, encompassing tasks such as monitoring seed growth, assessing leaf sample quality, and detecting defects and diseases in seeds. In a study conducted by Joshi et al., a novel methodology was investigated for categorizing rice seeds employing OCT in tandem with a DNN, eliminating the necessity for labeling and invasive procedures [78]. This approach facilitates rapid and non-invasive identification of diverse seed varieties, offering significant advantages for quality assurance and enhancing agricultural productivity. ...
... The initial approach involves generating a feature map by excluding the last layers of the pre-trained model and subsequently inputting it into a classical ML model for classification [68], [80]. The alternative method entails integrating Fully Connected Neural Networks (FCNNs) into the pre-trained models after removing the final layers to execute the classification task [50], [51], [52], [53], [54], [55], [70], [78]. ...
Optical Coherence Tomography (OCT) emerged as a technology for the detection of retinal disease. Owing to its excellent performance and ability to provide in-vivo high-resolution images, the popularity increased dramatically across various application domains. Consequently, OCT has been widely used in other branches of medical applications, i.e., oncology and otolaryngology, industry, and agriculture. Despite its widespread use, OCT image analysis is an inherently subjective, laborious, and time-intensive task that requires expertise. Deep learning (DL) stands as the current state-of-the-art method for image analysis. Hence, several research groups have directed their efforts toward incorporating DL algorithms with OCT imaging to reduce the time as well as the subjectivity. This article comprehensively reviews the principal technological advancements in DL methods applied across various OCT applications. Additionally, it explores the latest trends in developing DL methods for OCT, highlights their limitations, and discusses future opportunities in a comprehensive manner.
... The accuracy rate reached 90.00%, 80.33%, and 85.00%, respectively, which achieved very good results. Joshi et al. [15] used deep neural networks and optical coherence tomography to achieve marker-free and lossless classification of rice seeds, achieving good classification results. Ge et al. [16] proposed a public hyperspectral image of rice seed data set and a case difficulty weighted K-nearest neighbor algorithm (IDKNN), which has achieved good results in the classification of different varieties of rice. ...
Rice is one of the most important crops for food supply, and there are multiple differences in the quality of rice in different geographic regions, which have a significant impact on subsequent yields and economic benefits. The traditional rice identification methods are time-consuming, inefficient, and delicate. This study proposes a deep learning-based method for fast and non-destructive classification of rice grown in different geographic environments. The experiment collected rice with the name of Ji-Japonica 830 from 10 different regions, and a total of 10,600 rice grains were obtained, and the fronts and backsides of the seeds were photographed with a camera in batches, and a total of 30,000 images were obtained by preprocessing the data. The proposed improved residual network architecture, High-precision Residual Network (HResNet), was used to compare the performance of the models. The results showed that HResNet obtained the highest classification accuracy result of 95.13%, which is an improvement of 7.56% accuracy with respect to the original model, and validation showed that HResNet achieves a 98.7% accuracy in the identification of rice grown in different soil classes. The experimental results show that the proposed network model can effectively recognize and classify rice grown in different soil categories. It can provide a reference for the identification of other crops and can be applied for consumer and food industry use.
... Methods like partial least squares discriminant analysis (PLS-DA), SVM, and K-nearest neighbor (KNN) have proven effective in spectral discrimination, differentiating rice seed varieties. Joshi et al. (2021) proposed a DL-assisted optical coherence tomography (OCT) technique for subsurface imaging to differentiate between various seed species. After extensive training, the network demonstrated excellent accuracy with test datasets. ...
The past few years have witnessed significant progress in emerging disease detection techniques for accurately and rapidly tracking rice diseases and predicting potential solutions. In this review we focus on image processing techniques using machine learning (ML) and deep learning (DL) models related to multi-scale rice diseases. Furthermore, we summarize applications of different detection techniques, including genomic, physiological, and biochemical approaches. In addition, we also present the state-of-the-art in contemporary optical sensing applications of pathogen–plant interaction phenotypes. This review serves as a valuable resource for researchers seeking effective solutions to address the challenges of high-throughput data and model recognition for early detection of issues affecting rice crops through ML and DL models.
... Rice is one of the products manufactured in numerous countries and consumed worldwide. Rice can be priced on different variables in the market [3]. Colour, texture, fracture rate, and shape are some of these parameters. ...
Automated rice variety detection and classification is a task that includes automatically categorizing and identifying varieties or different types of rice based on different characteristics namely grain texture, shape, colour, and size. This process is essential for quality assessment, agricultural management, and research purposes. Deep learning (DL) is a subfield of machine leaching (ML) that focuses on training an artificial neural network (ANN) with multiple layers to learn hierarchical representations of data. Convolutional Neural Network (CNN) was widely applied in image-based tasks such as rice variety detection, as they could efficiently capture visual features and patterns. In this study, we propose an Automated Rice Variety Detection and Classification using Quantum Inspired Moth Flame Optimizer with Deep Learning (ARVDC-QIMFODL) technique. The presented ARVDC-QIMFODL technique focuses on the automated identification and classification of distinct kinds of rice varieties. To accomplish this, the ARVDC-QIMFODL technique uses the Median modified wiener filter (MMWF) technique for the noise removal process. Followed by, the feature extraction process takes place by an improved ShuffleNet model. For rice variety detection and classification, the long short-term memory (LSTM) approach was applied. At last, the QIMFO algorithm-based hyperparameter selection process is performed to optimize the detection results of the LSTM system. The simulation outcome of the ARVDC-QIMFODL method is tested on a rice image dataset. An extensive set of experiments showed the remarkable efficiency of the ARVDC-QIMFODL system over other models.
... In the study [16], a hyperspectral imaging model is presented for the classification of 10 different high-quality rice varieties using a deep learning network. In the study [17], a label-free classification model is proposed for different rice varieties using a deep neural network assisted imaging technique. In the study [18], a deep learning-based method is used to classify both grain-shaped and milled images of five varieties of Spanish rice. ...
Deep learning is a machine learning approach that has been widely used in many different fields in recent years. It is used in agriculture for various purposes, such as product classification and diagnosis of agricultural diseases. In this study, we propose a deep-learning model for the classification of rice species. Rice is an agricultural product that is widely consumed in Turkey as well as in the world. In our study, a rice data set that contains 7 morphological features obtained by using 3810 rice grains belonging to two species is used. Our model consists of three hidden layers and two dropouts (3H2D) added to these layers to prevent overfitting in classification. The success of the model is compared with Logistic Regression (LR), Multilayer Perceptron (MLP), Support Vector Machine (SVM), Decision Tree (DT), Random Forest (RF), Naïve Bayes (NB), K-Nearest Neighbors (KNN), Ada Boost (AB), Bagging (BG), and Voting (VT) classifiers. The success rates of these methods are as follows: 93.02%, 92.86%, 92.83%, 92.49%, 92.39%, 91.71%, 88.58%, 92.34%, 91.68%, and 90.52% respectively. The success rate of the proposed method is 94.09%. According to the results obtained, the proposed method is more successful than all of these machine learning methods.
... A Machine vision in agriculture applications related to rice quality inspection and grading has been reviewed and summarized [3]- [5]. There are various works related to the field of rice quality inspection. ...
Seed inspection is crucial for plant nurseries and farmers as it ensures seed quality when growing seedlings. It is traditionally accomplished by expert inspectors filtering samples manually, but there are some challenges, such as cost, accuracy, and large numbers. Speed and accuracy were the main conditions for increasing agricultural productivity. Machine learning is a sub-science of Artificial Intelligence that can be applied in research on the classification of rice seed quality. The pipeline of a machine learning system is dataset collection, training, validation, and testing. Model making begins with taking data on the characteristics of rice seeds based on physical parameters in the form of seed shape and color. The dataset used is two thousand images divided into two categories, namely superior seeds and non-superior seeds. Training and Validation was conducted using the Convolutional Neural Network (CNN) algorithm with the concept of cross-validation on Google Collaboratory notebooks. The ratio split of train data and validation data in modeling from a dataset is 80:20. The result of the model formed is a model with the development of a Deep Convolutional Neural Network (Deep CNN) that can classify the digital image data of rice seeds from the results of data calls uploaded into the system. The results of the experiment conducted on 30 test data can be analyzed so that the system can classify superior and non-superior seeds with a precision value of 93% and a recall of 95%.
... Recently, Yoon and Lee (Yoon & Lee, 2019) adopted the same technique to identify fungus-infected tomato seeds. Similarly, Joshi et al. (Joshi, et al., 2021) presented the use of OCT system for subsurface imaging of rice seeds to distinguish different seed varieties. In addition, OCT techniques have also been applied for studying leaf senescence (Anna et al., 2019) and root growth in soil (Larimer et al., 2020). ...
Seed germination and uniform plant stand in the field are the most critical crop growth stages determining the final yield. Pea (Pisum sativum L.) seeds production is often hampered due to the seed dormancy caused by the hard seed coat. Such effect is mainly attributed to poor or uneven germination and unsynchronised seedling emergence. Understanding the time course of water intake and several critical germination indicators can reveal many features of seed germination such as rate and uniformity. This paper used optical coherence tomography (OCT), a noninvasive and cross‐sectional imaging technique, to monitor the inner structural changes throughout the germination process. A sequence of cross‐sectional OCT images of pea (P. sativum L.) seeds, together with additional microscopic optical images, was recorded continuously and in situ for over 40 h. OCT and microscopic images revealed the changes in the internal structure and the external shape of the pea seeds during germination, respectively. It was found that the cross‐sectional OCT images helped to identify the critical indicators distinguishing the different phases of germination pea seeds. Therefore, the presented OCT approach offers a fast and nondestructive way to precisely measure the structural indicators in different germination phases.
... Cultivating rice varieties with resistance to the disease is the most cost-effective way to prevent rice blast [3], which plays an important role in ensuring grain security. Identification of rice seeds has always been an important issue in crop breeding [4]. Traditional rice seed detection methods include morphological identification, microscopic identification, simple sequence repeat (SSR) molecular markers and the field experiment method [5]. ...
Rice blast is a serious threat to rice yield. Breeding disease-resistant varieties is one of the most economical and effective ways to prevent damage from rice blast. The traditional identification of resistant rice seeds has some shortcoming, such as long possession time, high cost and complex operation. The purpose of this study was to develop an optimal prediction model for determining resistant rice seeds using Ranman spectroscopy. First, the support vector machine (SVM), BP neural network (BP) and probabilistic neural network (PNN) models were initially established on the original spectral data. Second, due to the recognition accuracy of the Raw-SVM model, the running time was fast. The support vector machine model was selected for optimization, and four improved support vector machine models (ABC-SVM (artificial bee colony algorithm, ABC), IABC-SVM (improving the artificial bee colony algorithm, IABC), GSA-SVM (gravity search algorithm, GSA) and GWO-SVM (gray wolf algorithm, GWO)) were used to identify resistant rice seeds. The difference in modeling accuracy and running time between the improved support vector machine model established in feature wavelengths and full wavelengths (200-3202 cm-1) was compared. Finally, five spectral preproccessing algorithms, Savitzky-Golay 1-Der (SGD), Savitzky-Golay Smoothing (SGS), baseline (Base), multivariate scatter correction (MSC) and standard normal variable (SNV), were used to preprocess the original spectra. The random forest algorithm (RF) was used to extract the characteristic wavelengths. After different spectral preproccessing algorithms and the RF feature extraction, the improved support vector machine models were established. The results show that the recognition accuracy of the optimal IABC-SVM model based on the original data was 71%. Among the five spectral preproccessing algorithms, the SNV algorithm's accuracy was the best. The accuracy of the test set in the IABC-SVM model was 100%, and the running time was 13 s. After SNV algorithms and the RF feature extraction, the classification accuracy of the IABC-SVM model did not decrease, and the running time was shortened to 9 s. This demonstrates the feasibility and effectiveness of IABC in SVM parameter optimization, with higher prediction accuracy and better stability. Therefore, the improved support vector machine model based on Ranman spectroscopy can be applied to the fast and non-destructive identification of resistant rice seeds.
Circular leaf spot (CLS) disease poses a significant threat to persimmon cultivation, leading to substantial harvest reductions. Existing visual and destructive inspection methods suffer from subjectivity, limited accuracy, and considerable time consumption. This study presents an automated pre-identification method of the disease through a deep learning (DL) based pipeline integrated with optical coherence tomography (OCT), thereby addressing the highlighted issues with the existing methods. The investigation yielded promising outcomes by employing transfer learning with pre-trained DL models, specifically DenseNet-121 and VGG-16. The DenseNet-121 model excels in differentiating among three stages of CLS disease (healthy (H), apparently healthy (or healthy-infected (HI)), and infected (I)). The model achieved precision values of 0.7823 for class-H, 0.9005 for class-HI, and 0.7027 for class-I, supported by recall values of 0.8953 for class-HI and 0.8387 for class-I. Moreover, the performance of CLS detection was enhanced by a supplemental quality inspection model utilizing VGG-16, which attained an accuracy of 98.99% in discriminating between low-detail and high-detail images. Moreover, this study employed a combination of LAMP and A-scan for the dataset labeling process, significantly enhancing the accuracy of the models. Overall, this study underscores the potential of DL techniques integrated with OCT to enhance disease identification processes in agricultural settings, particularly in persimmon cultivation, by offering efficient and objective pre-identification of CLS and enabling early intervention and management strategies.
Significance
Full-field optical coherence microscopy (FF-OCM) is a prevalent technique for backscattering and phase imaging with epi-detection. Traditional methods have two limitations: suboptimal utilization of functional information about the sample and complicated optical design with several moving parts for phase contrast.
Aim
We report an OCM setup capable of generating dynamic intensity, phase, and pseudo-spectroscopic contrast with single-shot full-field video-rate imaging called bichromatic tetraphasic (BiTe) full-field OCM with no moving parts.
Approach
BiTe OCM resourcefully uses the phase-shifting properties of anti-reflection (AR) coatings outside the rated bandwidths to create four unique phase shifts, which are detected with two emission filters for spectroscopic contrast.
Results
BiTe OCM overcomes the disadvantages of previous FF-OCM setup techniques by capturing both the intensity and phase profiles without any artifacts or speckle noise for imaging scattering samples in three-dimensional (3D). BiTe OCM also utilizes the raw data effectively to generate three complementary contrasts: intensity, phase, and color. We demonstrate BiTe OCM to observe cellular dynamics, image live, and moving micro-animals in 3D, capture the spectroscopic hemodynamics of scattering tissues along with dynamic intensity and phase profiles, and image the microstructure of fall foliage with two different colors.
Conclusions
BiTe OCM can maximize the information efficiency of FF-OCM while maintaining overall simplicity in design for quantitative, dynamic, and spectroscopic characterization of biological samples.
Biometric seed analysis can be used to characterize and differentiate forest species. However, forest species are generally studied using manual methods such as measurements with a digital caliper, which provides a limited amount of information on plant morphological characteristics, whereas agronomic species are analyzed using expensive and often inaccessible equipment. Thus, the objective of the present study was to demonstrate that seed image analysis and processing tools can help characterize and differentiate Brazilian forest species. For this purpose, the seeds of 155 forest species belonging to 42 families were photographed and analyzed to extract data on their morphometric descriptors using a new methodological approach. A total of 18 characteristics were assessed, namely eight dimensions, four shape characteristics, and six color characteristics. A set of approximately 1.827 million data was extracted from 101,521 seed images. Digital image processing efficiently characterized the studied seeds and the obtained characteristics allowed us to differentiate between species, including those belonging to the same botanical family and genus. Therefore, seed image analysis using the proposed methodology can be used to characterize, differentiate, and automatedly identify forest species in Brazil.
Keywords
Biometry; Forest seeds; Image processing; Seed analysis
Non-destructive measurements of internal morphological structures in plant materials such as seeds are of high interest in agricultural research. The estimation of pericarp thickness is important to understand the grain quality and storage stability of seeds and can play a crucial role in improving crop yield. In this study, we demonstrate the applicability of fiber-based Bessel beam Fourier domain (FD) optical coherence microscopy (OCM) with a nearly constant high lateral resolution maintained at over ~400 µm for direct non-invasive measurement of the pericarp thickness of two different sorghum genotypes. Whereas measurements based on axial profiles need additional knowledge of the pericarp refractive index, en-face views allow for direct distance measurements. We directly determine pericarp thickness from lateral sections with a 3 µm resolution by taking the width of the signal corresponding to the pericarp at the 1/e threshold. These measurements enable differentiation of the two genotypes with 100% accuracy. We find that trading image resolution for acquisition speed and view size reduces the classification accuracy. Average pericarp thicknesses of 74 µm (thick phenotype) and 43 µm (thin phenotype) are obtained from high-resolution lateral sections, and are in good agreement with previously reported measurements of the same genotypes. Extracting the morphological features of plant seeds using Bessel beam FD-OCM is expected to provide valuable information to the food processing industry and plant breeding programs.
Rice variety identification is important for genetic breeding classification and crop yield estimation. Traditional identification methods are time-consuming and inaccurate. This paper proposes a method for rice variety identification based on the hyperspectral characteristics of leaves. Hyperspectral data of rice leaves were collected using a geophysical spectrometer imaging system. To reduce the redundance among the hyperspectral data and save the identification cost, locality preserving projections (LPP) is first applied to extract low-dimensional representative features from the leaf hyperspectral data. Then, support vector machine (SVM) is combined for conducting the identification of rice varieties. The experimental results show that the identification rate of 10 varieties of early rice was found to be 91.67% and the identification rate of 10 varieties of late rice was 97.33%.
Acid Mine Drainage (AMD) is one of the world's worst wastewater pollutions, and it causes a massive threat to crop. In this study, as a way to speedily monitor the effects of AMD on rice and non-destructively, we have used a technique called biospeckle Optical Coherence Tomography (bOCT)and measured after exposing to 40 ml/L and 80 ml/L simulated AMD for 24, 48 and 72 hours and compared the results with conventional physical such as root and shoot lengths and physiological parameters such as enzymes. Comparison revealed that the effects due to the AMD could be seen within 48hrs by bOCT with a decrease in a parameter called bisopeckle contrast which reflects the internal activity of the seeds in comparison to the physical (shoot and root lengths) and physiological parameters such as seed vigor, antioxidative system, iron uptake that appear only after 7 to 10 days. Despite the fast measurement time bOCT, the results agree with conventional indicators with both showing the same tendency. Due to the nature of the non-invasiveness of bOCT and also the possibility of speedy measurements, we expect bOCT has the potential to become an invaluable tool in agriculture not only for investigating the effects of AMD but also pave ways for speedily evaluating the influence of AMD.
Malaysia's rice production, consumption and area harvested are relatively small compared to its neighboring countries. Farmers lose an estimated average of 37% of their rice crop to pests and diseases every year due to poor pests and disease monitoring for the paddy rice. The research aims to find the optimal range of hyperparameters to assess their performance. The application of deep learning in detection of diseases and pests in rice plants can help farmers in applying timely treatment on the plants. Therefore, we can reduce the economic losses substantially. Most research conducted previously required large number datasets to get the best accuracy. In addition to that, most deep learning approaches have outperformed the other machine learning approaches in performing the classification tasks. However, deep learning approaches have many hyperparameters that require optimization. Thus, in this research, we will determine the optimal value of all hyperparameters used in the classification tasks. In this work, we also investigating the effects of varying the values of all the hyperparameters used in the CNN architecture by comparing the accuracy performance.KeywordsPaddy rice fieldConvolutional neural networkDeep learning
Fungal infection is a pre-harvest and post-harvest crisis for farmers of peanuts. In environments with temperatures around 28 °C to 30 °C or relative humidity of approximately 90%, mold-contaminated peanuts have a considerable likelihood to be infected with Aflatoxins. Aflatoxins are known to be highly carcinogenic, posing danger to humans and livestock. In this work, we proposed a new approach for detection of mold-contaminated peanuts at an early stage. The approach employs the optical coherence tomography (OCT) imaging technique and an error-correcting output code (ECOC) based Support Vector Machine (SVM) trained on features extracted using a pre-trained Deep Convolutional Neural Network (DCNN). To this end, mold-contaminated and uncontaminated peanuts were scanned to create a data set of OCT images used for training and evaluation of the ECOC-SVM model. Results showed that the proposed approach is capable of detecting mold-contaminated peanuts with respective accuracies of approximately 85% and 96% after incubation periods of 48 and 96 h.
As a non-destructive method, spectral-domain optical coherence tomography (OCT) has already been adapted to study seed samples. SD-OCT provides a set of capabilities in terms of millimetre-scale penetration depth, high-speed acquisition, and resolution typically in the range of ten microns. It has been therefore gaining popularity in this specific field. However, only a limited number of seed cultivars have been investigated with OCT so far, and there has been little discussions about the applicability of the OCT to other seed varieties. In this work, we report the measurement of 14 types of seed using OCT system. It was found that OCT can provide high-resolution cross-sectional images of 12 out of 14 seed types. This work represents the first systematic study towards a seed catalog which provides the essential information about the applicability of OCT for studying a variety.
Sperm cell motility and morphology observed under the bright field microscopy are the only criteria for selecting a particular sperm cell during Intracytoplasmic Sperm Injection (ICSI) procedure of Assisted Reproductive Technology (ART). Several factors such as oxidative stress, cryopreservation, heat, smoking and alcohol consumption, are negatively associated with the quality of sperm cell and fertilization potential due to the changing of subcellular structures and functions which are overlooked. However, bright field imaging contrast is insufficient to distinguish tiniest morphological cell features that might influence the fertilizing ability of sperm cell. We developed a partially spatially coherent digital holographic microscope (PSC-DHM) for quantitative phase imaging (QPI) in order to distinguish normal sperm cells from sperm cells under different stress conditions such as cryopreservation, exposure to hydrogen peroxide and ethanol. Phase maps of total 10,163 sperm cells (2,400 control cells, 2,750 spermatozoa after cryopreservation, 2,515 and 2,498 cells under hydrogen peroxide and ethanol respectively) are reconstructed using the data acquired from the PSC-DHM system. Total of seven feedforward deep neural networks (DNN) are employed for the classification of the phase maps for normal and stress affected sperm cells. When validated against the test dataset, the DNN provided an average sensitivity, specificity and accuracy of 85.5%, 94.7% and 85.6%, respectively. The current QPI + DNN framework is applicable for further improving ICSI procedure and the diagnostic efficiency for the classification of semen quality in regard to their fertilization potential and other biomedical applications in general.
Optical coherence tomography (OCT) is being increasingly adopted as a label-free and non-invasive technique for biomedical applications such as cancer and ocular disease diagnosis. Diagnostic information for these tissues is manifest in textural and geometric features of the OCT images, which are used by human expertise to interpret and triage. However, it suffers delays due to the long process of the conventional diagnostic procedure and shortage of human expertise. Here, a custom deep learning architecture, LightOCT, is proposed for the classification of OCT images into diagnostically relevant classes. LightOCT is a convolutional neural network with only two convolutional layers and a fully connected layer, but it is shown to provide excellent training and test results for diverse OCT image datasets. We show that LightOCT provides 98.9% accuracy in classifying 44 normal and 44 malignant (invasive ductal carcinoma) breast tissue volumetric OCT images. Also, >96% accuracy in classifying public datasets of ocular OCT images as normal, age-related macular degeneration and diabetic macular edema. Additionally, we show ∼96% test accuracy for classifying retinal images as belonging to choroidal neovascularization, diabetic macular edema, drusen, and normal samples on a large public dataset of more than 100,000 images. The performance of the architecture is compared with transfer learning based deep neural networks. Through this, we show that LightOCT can provide significant diagnostic support for a variety of OCT images with sufficient training and minimal hyper-parameter tuning. The trained LightOCT networks for the three-classification problem will be released online to support transfer learning on other datasets.
Hyperspectral imaging has attracted great attention as a non-destructive and fast method for seed quality and safety assessment in recent years. The capability of this technique for classification and grading, viability and vigor detection, damage (defect and fungus) detection, cleanness detection and seed composition determination is illustrated by presentation of applications in quality and safety determination of seed in this review. The summary of hyperspectral imaging technology for seed quality and safety inspection for each category is also presented, including the analyzed spectral range, sample varieties, sample status, sample numbers, features (spectral features, image features, feature extraction methods), signal mode and data analysis strategies. The successful application of hyperspectral imaging in seed quality and safety inspection proves that many routine seed inspection tasks can be facilitated with hyperspectral imaging.
Non-invasive investigation of rice leaf specimens to characterize the morphological formation and particular structural information that is beneficial for agricultural perspective was demonstrated using a low coherence interferometric method called swept source optical coherence tomography (SS-OCT). The acquired results non-invasively revealed morphological properties of rice leaf, such as bulliform cells; aerenchyma, parenchyma, and collenchyma layer; and vascular bundle. Beside aforementioned morphologic characteristics, several leaf characteristics associated with cytological mechanisms of leaf rolling (leaf inclination) were examined for the pre-identification of inevitable necrosis and atrophy of leaf tissues by evaluating acute angle information, such as angular characteristics of the external bi-directional angles between the lower epidermis layer and lower mid-vein, and internal angle of lower mid-vein. To further assist the pre-identification, acquired cross-sections were employed to enumerate the small veins of each leaf specimen. Since mutants enlarge leaf angles due to increased cell division in the adaxial epidermis, healthy and abnormal leaf specimens were morphologically and quantitatively compared. Therefore, the results of the method can be used in agriculture, and SS-OCT shows potential as a rigorous investigation method for selecting mutant infected rice leaf specimens rapidly and non-destructively compared to destructive and time consuming gold-standard methods with a lack of precision.
Abstract Farmers around the world have recently experienced significant crop losses due to severe heat and drought. Such extreme weather events and the need to feed a rapidly growing population have raised concerns for global food security. While plant breeding has been very successful and has delivered today’s highly productive crop varieties, the rate of genetic improvement must double to meet the projected future demands. Here we discuss basic principles and features of crop breeding and how modern technologies could efficiently be explored to boost crop improvement in the face of increasingly challenging production conditions.
Leaf senescence provides a unique window to explore the age-dependent programmed degradation at organ label in plants. Here, spectral domain optical coherence tomography (SD-OCT) has been used to study in vivo senescing leaf microstructural changes in the deciduous plant Acer serrulatum Hayata. Hayata leaves show autumn phenology and change color from green to yellow and finally red. SD-OCT image analysis shows distinctive features among different layers of the leaves; merging of upper epidermis and palisade layers form thicker layers in red leaves compared to green leaves. Moreover, A-scan analysis showed a significant (p < 0.001) decrease in the attenuation coefficient (for wavelength range: 1100–1550 nm) from green to red leaves. In addition, the B-scan analysis also showed significant changes in 14 texture parameters extracted from second-order spatial gray level dependence matrix (SGLDM). Among these parameters, a set of three features (energy, skewness, and sum variance), capable of quantitatively distinguishing difference in the microstructures of three different colored leaves, has been identified. Furthermore, classification based on k-nearest neighbors algorithm (k-NN) was found to yield 98% sensitivity, 99% specificity, and 95.5% accuracy. Following the proposed technique, a portable noninvasive tool for quality control in crop management can be anticipated.
Genebanks are valuable resources for crop improvement through the acquisition, ex-situ conservation and sharing of unique germplasm among plant breeders and geneticists. With over seven million existing accessions and increasing storage demands and costs, genebanks need efficient characterization and curation to make them more accessible and usable and to reduce operating costs, so that the crop improvement community can most effectively leverage this vast resource of untapped novel genetic diversity. However, the sharing and inconsistent documentation of germplasm often results in unintentionally duplicated collections with poor characterization and many identical accessions that can be hard or impossible to identify without passport information and unmatched accession identifiers. Here we demonstrate the use of genotypic information from these accessions using a cost-effective next generation sequencing platform to find and remove duplications. We identify and characterize over 50% duplicated accessions both within and across genebank collections of Aegilops tauschii, an important wild relative of wheat and source of genetic diversity for wheat improvement. We present a pipeline to identify and remove identical accessions within and among genebanks and curate globally unique accessions. We also show how this approach can also be applied to future collection efforts to avoid the accumulation of identical material. When coordinated across global genebanks, this approach will ultimately allow for cost effective and efficient management of germplasm and better stewarding of these valuable resources.
In breast cancer, 20%–30% of cases require a second surgery because of incomplete excision of malignant tissues. Therefore, to avoid the risk of recurrence, accurate detection of the cancer margin by the clinician or surgeons needs some assistance. In this paper, an automated volumetric analysis of normal and breast cancer tissue is done by a machine learning algorithm to separate them into two classes. The proposed method is based on a support-vector-machine-based classifier by dissociating 10 features extracted from the A-line, texture, and phase map by the swept-source optical coherence tomographic intensity and phase images. A set of 88 freshly excised breast tissue [44 normal and 44 cancers (invasive ductal carcinoma tissues)] samples from 22 patients was used in our study. The algorithm successfully classifies the cancerous tissue with sensitivity, specificity, and accuracy of 91.56%, 93.86%, and 92.71% respectively. The present computational technique is fast, simple, and sensitive, and extracts features from the whole volume of the tissue, which does not require any special tissue preparation nor an expert to analyze the breast cancer as required in histopathology. Diagnosis of breast cancer by extracting quantitative features from optical coherence tomographic images could be a potentially powerful method for cancer detection and would be a valuable tool for a fine-needle-guided biopsy.
Abstract The tremendous advances achieved in the biophotonics technologies have intensified the necessity for non-invasive modalities that can characterize diverse biological materials with increased sensitivity and resolution. Optical coherence tomography (OCT) is one of the techniques that has been applied for biological applications in medicine and agriculture to identify structural properties. Herein, we report the successful incorporation of OCT for the identification of morphological changes that occur as a result of the bitter rot disease, through continuous detection of structural changes. Detailed inner morphological structural changes occurring in fruit specimens were precisely analyzed as a function of the disease incubation period using OCT. The conducted histological correlation and quantitative three-dimensional evaluations provide a robust platform for further discoveries related to plant materials. The results highlight the initial identification of bitter rot progression on apple specimens owing to the non-invasive inspection capability of OCT. Therefore, we expect that the proposed method will enable immediate sensitivity improvements in the inspection of plant diseases for postharvest utility.
The volume and complexity of diagnostic imaging is increasing at a pace faster than the availability of human expertise to interpret it. Artificial intelligence has shown great promise in classifying two-dimensional photographs of some common diseases and typically relies on databases of millions of annotated images. Until now, the challenge of reaching the performance of expert clinicians in a real-world clinical pathway with three-dimensional diagnostic scans has remained unsolved. Here, we apply a novel deep learning architecture to a clinically heterogeneous set of three-dimensional optical coherence tomography scans from patients referred to a major eye hospital. We demonstrate performance in making a referral recommendation that reaches or exceeds that of experts on a range of sight-threatening retinal diseases after training on only 14,884 scans. Moreover, we demonstrate that the tissue segmentations produced by our architecture act as a device-independent representation; referral accuracy is maintained when using tissue segmentations from a different type of device. Our work removes previous barriers to wider clinical use without prohibitive training data requirements across multiple pathologies in a real-world setting.
The non-destructive classification of plant materials using optical inspection techniques has been gaining much recent attention in the field of agriculture research. Among them, a near-infrared (NIR) imaging method called optical coherence tomography (OCT) has become a well-known agricultural inspection tool since the last decade. Here we investigated the non-destructive identification capability of OCT to classify diversely stained (with various staining agents) Capsicum annuum seed specimens of different cultivars. A swept source (SS-OCT) system with a spectral band of 1310 nm was used to image unstained control C. annuum seeds along with diversely stained Capsicum seeds, belonging to different cultivar varieties, such as C. annuum cv. PR Ppareum, C. annuum cv. PR Yeol, and C. annuum cv. Asia Jeombo. The obtained cross-sectional images were further analyzed for the changes in the intensity of back-scattered light (resulting due to dye pigment material and internal morphological variations) using a depth scan profiling technique to identify the difference among each seed category. The graphically acquired depth scan profiling results revealed that the control specimens exhibit less back-scattered light intensity in depth scan profiles when compared to the stained seed specimens. Furthermore, a significant back-scattered light intensity difference among each different cultivar group can be identified as well. Thus, the potential capability of OCT based depth scan profiling technique for non-destructive classification of diversely stained C. annum seed specimens of different cultivars can be sufficiently confirmed through the proposed scheme. Hence, when compared to conventional seed sorting techniques, OCT can offer multipurpose advantages by performing sorting of seeds in respective to the dye staining and provides internal structural images non-destructively.
The feasibility of using hyperspectral imaging with convolutional neural network (CNN) to identify rice seed varieties was studied. Hyperspectral images of 4 rice seed varieties at two different spectral ranges (380–1030 nm and 874–1734 nm) were acquired. The spectral data at the ranges of 441–948 nm (Spectral range 1) and 975–1646 nm (Spectral range 2) were extracted. K nearest neighbors (KNN), support vector machine (SVM) and CNN models were built using different number of training samples (100, 200, 300, 400, 500, 600, 700, 800, 900, 1000, 1500, 2000, 2500 and 3000). KNN, SVM and CNN models in the Spectral range 2 performed slightly better than those in the Spectral range 1. The model performances improved with the increase in the number of training samples. The improvements were not significant when the number of training samples was large. CNN model performed better than the corresponding KNN and SVM models in most cases, which indicated the effectiveness of using CNN to analyze spectral data. The results of this study showed that CNN could be adopted in spectral data analysis with promising results. More varieties of rice need to be studied in future research to extend the use of CNNs in spectral data analysis.
Computer software—the Rice Seed Germination Evaluation System (RSGES)—was developed which can evaluate a rice seed image for germination prediction by using digital image processing and an artificial neural networks technique. The digital images are taken with a normal digital camera or mobile phone camera, which is very easy for farmers to process. RSGES consists of six main processing modules: 1) image acquisition, 2) image preprocessing, 3) feature extraction, 4) germination evaluation, 5) results presentation and 6) germination verification. The experiment was conducted on seed of the Thai rice species CP-111 in Bangkok and Chiang Mai, Thailand. RSGES extracted 18 features: 3 color features, 7 morphological features and 8 textural features. The system applied artificial neural network techniques to perform germination prediction. The system precision rate was 7.66% false accepted and 5.42% false rejected, with a processing speed of 8.31 s per image.
An important application of intravascular optical coherence tomography (IVOCT) for atherosclerotic tissue analysis is using it to estimate attenuation and backscatter coefficients. This work aims at exploring the potential of the attenuation coefficient, a proposed backscatter term, and image intensities in distinguishing different atherosclerotic tissue types with a robust implementation of depth-resolved (DR) approach. Therefore, the DR model is introduced to estimate the attenuation coefficient and further extended to estimate the backscatter-related term in IVOCT images, such that values can be estimated per pixel without predefining any delineation for the estimation. In order to exclude noisy regions with a weak signal, an automated algorithm is implemented to determine the cut-off border in IVOCT images. The attenuation coefficient, backscatter term, and the image intensity are further analyzed in regions of interest, which have been delineated referring to their pathology counterparts. Local statistical values were reported and their distributions were further compared with a two-sample t-test to evaluate the potential for distinguishing six types of tissues. Results show that the IVOCT intensity, DR attenuation coefficient, and backscatter term extracted with the reported implementation are complementary to each other on characterizing six tissue types: mixed, calcification, fibrous, lipid-rich, macrophages, and necrotic core.
1.Introduction
We present an algorithm for identifying retinal pathologies given retinal optical coherence tomography (OCT) images. Our approach fine-tunes a pre-trained convolutional neural network (CNN), GoogLeNet, to improve its prediction capability (compared to random initialization training) and identifies salient responses during prediction to understand learned filter characteristics. We considered a data set containing subjects with diabetic macular edema, or dry age-related macular degeneration, or no pathology. The fine-tuned CNN could effectively identify pathologies in comparison to classical learning. Our algorithm aims to demonstrate that models trained on non-medical images can be fine-tuned for classifying OCT images with limited training data.
Almost three decades ago Alec Jeffreys published his seminal Nature papers on the use of minisatellite probes for DNA fingerprinting of humans (Jeffreys and colleagues Nature 1985, 314:67-73 and Nature 1985, 316:76-79). The new technology was soon adopted for many other organisms including plants, and when Hilde Nybom, Kurt Weising and Alec Jeffreys first met at the very First International Conference on DNA Fingerprinting in Berne, Switzerland, in 1990, everybody was enthusiastic about the novel method that allowed us for the first time to discriminate between humans, animals, plants and fungi on the individual level using DNA markers. A newsletter coined "Fingerprint News" was launched, T-shirts were sold, and the proceedings of the Berne conference filled a first book on "DNA fingerprinting: approaches and applications". Four more conferences were about to follow, one on each continent, and Alec Jeffreys of course was invited to all of them. Since these early days, methodologies have undergone a rapid evolution and diversification. A multitude of techniques have been developed, optimized, and eventually abandoned when novel and more efficient and/or more reliable methods appeared. Despite some overlap between the lifetimes of the different technologies, three phases can be defined that coincide with major technological advances. Whereas the first phase of DNA fingerprinting ("the past") was dominated by restriction fragment analysis in conjunction with Southern blot hybridization, the advent of the PCR in the late 1980s gave way to the development of PCR-based single- or multi-locus profiling techniques in the second phase. Given that many routine applications of plant DNA fingerprinting still rely on PCR-based markers, we here refer to these methods as "DNA fingerprinting in the present", and include numerous examples in the present review. The beginning of the third phase actually dates back to 2005, when several novel, highly parallel DNA sequencing strategies were developed that increased the throughput over current Sanger sequencing technology 1000-fold and more. High-speed DNA sequencing was soon also exploited for DNA fingerprinting in plants, either in terms of facilitated marker development, or directly in the sense of "genotyping-by-sequencing". Whereas these novel approaches are applied at an ever increasing rate also in non-model species, they are still far from routine, and we therefore treat them here as "DNA fingerprinting in the future".
In this study, we distinguished Cucumber green mottle mosaic virus (CGMMV) infected seeds from healthy seeds using optical coherence tomography (OCT). Two dimensional OCT images and stereo micrographs revealed that the infected seeds had narrow gap between the seed coat and endosperm that were not present in the healthy seeds. Three dimensional OCT images confirmed that the narrow gaps were present in the inner structure of overall the infected seeds. A-scan analysis was also performed to calculate the distance from the seed coat to the endosperm. The results revealed a difference in the width of the gap of about 20 μm between healthy and infected cucumber seeds. Taken together, these results suggest that OCT could be applied as an effective non-destructive method for CGMMV infection of cucumber seeds.
The machinery and operations when improperly designed may generate rice kernel cracking and breakage and consequently a low marketing price. The objective of this work was to determine the influence of the rice processing operations on physical and mechanical properties of different rice varieties. Three varieties of rice, rough, brown and milled, were used in this work. The bulk densities of all varieties increased with processing up to 51% and there were differences among the varieties; the rice grain specific gravity was influenced neither by the processing nor by the varieties. The processing influenced the porosity of the bulk rice grains; the external static and dynamic friction coefficients were reduced. The higher friction coefficient values were observed on wood surface and the lowest on steel surface; the compression force needed to promote the rice kernel collapse was affected significantly by the processing.
Optical coherence tomography (OCT) is a novel, high-resolution diagnostic tool that is capable of imaging the arterial wall and plaques. The differentiation between different types of atherosclerotic plaque is based on qualitative differences in gray levels and structural appearance. We hypothesize that a quantitative data analysis of the OCT signal allows measurement of light attenuation by the local tissue components, which can facilitate quantitative spatial discrimination between plaque constituents. High-resolution OCT images (at 800 nm) of human atherosclerotic arterial segments obtained at autopsy were histologically validated. Using a new, simple analysis algorithm, which incorporates the confocal properties of the OCT system, the light attenuation coefficients for these constituents were determined: for diffuse intimal thickening (5.5±1.2 mm-1) and lipid-rich regions (3.2±1.1 mm-1), the attenuation differed significantly from media (9.9±1.8 mm-1), calcifications (11.1±4.9 mm-1) and thrombi (11.2±2.3 mm-1) (p<0.01). These proof of principle studies show that simple quantitative analysis of the OCT signals allows spatial determination of the intrinsic optical attenuation coefficient of atherosclerotic tissue components within regions of interest. Combining morphological imaging by OCT with the observed differences in optical attenuation coefficients of the various regions may enhance discrimination between various plaque types.
The rapid and non-destructive discriminant analysis of rice seeds has great significance for large-scale agriculture. Using near-infrared (NIR)diffuse-reflectance spectroscopy with partial least squares-discriminant analysis (PLS-DA), a variety identification method of multi-grain rice seeds was developed. The equidistant combination method was adopted for large-range wavelength screening. A step-by-step phase-out method was proposed to eliminate interference wavelengths and improve the predicted effect. The optimal wavelength model was a combination of 54 wavelengths within 808–974 nm of the short-NIR region. One type of pure rice variety (Y Liangyou 900)was used for identification (negative). Positive samples included the other four pure varieties and contamination of Y Liangyou 900 by the above four varieties. The recognition-accuracy rates for positive, negative and total validation samples reached 93.1%, 95.1%, and 94.3%, respectively. In the long-NIR region, the local optimal wavelength model was a combination of 49 wavelengths within 1188–1650 nm, and the recognition-accuracy rates for positive, negative and total validation samples were 90.3%, 94.1%, and 92.5%, respectively. Results confirmed the feasibility of NIR spectroscopy for variety identification of multi-grain rice seeds. The proposed two discrete-wavelength models located in the short- and long-NIR regions can provide valuable reference to a dedicated spectrometer.
Moisture content is an important feature of fruits and vegetables. As 80% of apple content is water, so decreasing the moisture content will degrade the quality of apples (Golden Delicious). The computational and texture features of the apples were extracted from optical coherence tomography (OCT) images. A support vector machine with a Gaussian kernel model was used to perform automated classification. To evaluate the quality of wax coated apples during storage in vivo, our proposed method opens up the possibility of fully automated quantitative analysis based on the morphological features of apples. Our results demonstrate that the analysis of the computational and texture features of OCT images may be a good non-destructive method for the assessment of the quality of apples.
Optical coherence tomography (OCT) is the optical analog of ultrasound imaging and is emerging as a powerful imaging technique that enables non-invasive, in vivo, high resolution, cross-sectional imaging in biological tissue. A new generation OCT technology has now been developed, representing a quantum leap in resolution and speed, achieving in vivo optical biopsy, i.e. the visualization of tissue architectural morphology in situ and in real time. Functional extensions of OCT technology enable non-invasive, depth resolved functional assessment and imaging of tissue. These new techniques should not only improve image contrast, but should also enable the differentiation of pathologies via metabolic properties or functional state. The book introduces OCT technology and applications not only from an optical and technological viewpoint, but also from biomedical and clinical perspectives. The chapters are written by leading international research groups, in a style comprehensible to a broad audience. It will be of interest not only to physicists, scientists and engineers, but also to biomedical and clinical researchers from different medical specialties.
The purpose of this article was to explore a new feature extraction method for classifying paddy seeds using a feature extraction algorithm to achieve the Horizontal-Vertical and Front-Rear angles. The method used fusion of angle features for classification, which were then compared to features such as seed color, shape, and texture. Experiments show that the proposed features work better in classifying paddy seeds in comparison with some of the standard features, and that the proposed features have an excellent discriminating property for seeds. The discriminating power of these features was assessed using the neural network architectures for the unique identification of seeds of four Paddy (Rice) grains: viz. Karjat-6(K6), Karjat-2(K2), Ratnagiri-4(R4) and Ratnagiri-24(R24). The classification accuracies of Color-Shape-Texture obtained was 95.2% while the proposed method gave an accuracy of 97.6%.
Maintaining prime fruit quality is the key to success in the fresh fruit business. Quality defects such as bruises in apples adversely affect their market value. Line-scan x-ray imaging has shown potential for detecting these quality defects. Quality assessment of apples with computer vision techniques is possible; however, two basic issues must be addressed before an automatic sorting system can be developed: (1) which image features best correlate with the fruit quality, and (2) which classifier should be used for optimal classification. These issues are discussed in this article. Red delicious (RD) and golden delicious (GD) apples were line-scanned for bruise damage. Spatial and transform features were evaluated for their discriminating contributions to fruit classification based on bruise defects. Stepwise discriminant analysis was used for selecting the salient features. Spatial edge features detected using Robert's edge detector, combined with the selected discrete cosine transform (DCT) coefficients proved to be good indicators of old (one month) bruises. Separate artificial neural network (ANN) classifiers were developed for old (one month) and new (24 hour) bruises. When an ANN classifier was used to sort apples based on old bruises, it achieved an accuracy of 90% for RD apples and 83% (93% after threshold adjustment) for GD apples. For new bruises, the accuracy was approximately 60% for both RD and GD apples. New bruises were not adequately separated using this methodology.
In this paper, digital image analysis is applied for non-destructive classification of rice and sticky rice seeds that are mixed together. It is a difficult task because of the similar surface color of the seeds. This paper presents an automatic classification method based on RGB color features. Hardware of image capturing is designed using back light source in order to maximize the contrast between the rice seeds and their background. RGB histogram is then calculated. The rule of classification between rice seed and sticky rice seed are created. Almost 97% of rice seeds are identified correctly. The correct classification rates for two rice varieties are: rice seeds 'Jasmine' 96.34% and sticky rice seeds 100%.
First experimental results on OCT imaging of internal structure of plant tissues and in situ OCT monitoring of plant tissue regeneration at different water supply are reported. Experiments for evaluating OCT capabilities were performed on Tradescantia. The investigation of seeds swelling was performed on wheat seeds (Triticum L.), barley seeds (Hordeum L.), long-fibred flax seeds (Linum usitatissimum L.) and cucumber seeds (Cucumis sativus L.). These OCT images correlate with standard microscopy data from the same tissue regions. Seeds were exposed to a low-intensity physical factor-the pulsed gradient magnetic field (GMF) with pulse duration 0.1 s and maximum amplitude 5 mT (4 successive pulses during 0.4 s). OCT and OCM enable effective monitoring of fast reactions in plants and seeds at different water supply.
Four varieties of rice grain were imaged with OCT technique. The structures of waxy rice endosperm were analyzed with B-scan image and the attenuation coefficients of four rice grains were calculated with averaged A-scan signal. Attenuation coefficient was also used to identify different varieties of rice grains.
Variety identification is an indispensable tool to assure grain purity and quality. Based on machine vision and pattern recognition, five China corn varieties were identified according to their external features. Images of non-touching corn kernels were acquired using a flat scanner. A total of 17 geometric features, 13 shape and 28 color features were extracted from color images of corn kernels. Two optimal feature sets were generated by stepwise discriminant analysis, and used as inputs to classifiers. A two-stage classifier combining distance discriminant and a back propagation neural network (BPNN) was built for identification. On the first stage, corn kernels were divided into three types: white, yellow and mixed corn by distance discriminant analysis. And then different varieties in the same type were identified by an improved BPNN classifier. The classification accuracies of BAINUO 6, NONGDA 86, NONGDA 108, GAOYOU 115, and NONGDA 4967 were 100, 94, 92, 88 and 100%, respectively.
The incidence of physiological and/or pathological defects in many fresh produce types is still unacceptably high and accounts for a large proportion of waste. With increasing interest in food security their remains strong demand in developing reliable and cost effective technologies for non-destructive screening of internal defects and rots, these being deemed unacceptable by consumers. It is well recognized that the internal defects and structure of turbid scattering media can be effectively visualized by using optical coherence tomography (OCT). In the present study, the high spatial resolution and advantages of OCT have been demonstrated for imaging the skins and outer laminae (concentric tissue layers) of intact whole onion bulbs with a view to non-invasively visualizing potential incidence/severity of internal defects. (© 2010 by Astro Ltd., Published exclusively by WILEY-VCH Verlag GmbH & Co. KGaA)
Diseases in plants cause major production and economic losses in agricultural industry worldwide. Monitoring of health and detection of diseases in plants and trees is critical for sustainable agriculture. To the best of our knowledge, there is no sensor commercially available for real-time assessment of health conditions in trees. Currently, scouting is most widely used mechanism for monitoring stress in trees, which is an expensive, labor-intensive, and time-consuming process. Molecular techniques such as polymerase chain reaction are used for the identification of plant diseases that require detailed sampling and processing procedure. Early information on crop health and disease detection can facilitate the control of diseases through proper management strategies such as vector control through pesticide applications, fungicide applications, and disease-specific chemical applications; and can improve productivity.The present review recognizes the need for developing a rapid, cost-effective, and reliable health-monitoring sensor that would facilitate advancements in agriculture. It describes the currently used technologies that can be used for developing a ground-based sensor system to assist in monitoring health and diseases in plants under field conditions. These technologies include spectroscopic and imaging-based, and volatile profiling-based plant disease detection methods. The paper compares the benefits and limitations of these potential methods.
Magnetic resonance imaging was used to acquire images of the internal structure of mandarins for non-destructive seed identification. Two different types of fast MRI sequences were investigated: a gradient echo and a spiral–radial, with 484 ms acquisition time for the former compared to 240 ms for the latter. The radial–spiral option allows over-sampling of the central area of the k-space maintaining the contrast within the MRI images and so the feasibility of seed segmentation. Three segmentation techniques were applied for image post-processing: region-based, one-dimension histogram variance, and two-dimension histogram variance, among which the latter procedure has been demonstrated to give the most promising results. Image features including perimeter, compactness, maximum distance to the gravity centre, and aspect ratio were employed in a linear discriminant function, by which seed identification of mandarins could be achieved with 100% accuracy using radial–spiral sequence and 98.7% accuracy with gradient echo images.
Global food demand is increasing rapidly, as are the environmental impacts of agricultural expansion. Here, we project global demand for crop production in 2050 and evaluate the environmental impacts of alternative ways that this demand might be met. We find that per capita demand for crops, when measured as caloric or protein content of all crops combined, has been a similarly increasing function of per capita real income since 1960. This relationship forecasts a 100-110% increase in global crop demand from 2005 to 2050. Quantitative assessments show that the environmental impacts of meeting this demand depend on how global agriculture expands. If current trends of greater agricultural intensification in richer nations and greater land clearing (extensification) in poorer nations were to continue, ~1 billion ha of land would be cleared globally by 2050, with CO(2)-C equivalent greenhouse gas emissions reaching ~3 Gt y(-1) and N use ~250 Mt y(-1) by then. In contrast, if 2050 crop demand was met by moderate intensification focused on existing croplands of underyielding nations, adaptation and transfer of high-yielding technologies to these croplands, and global technological improvements, our analyses forecast land clearing of only ~0.2 billion ha, greenhouse gas emissions of ~1 Gt y(-1), and global N use of ~225 Mt y(-1). Efficient management practices could substantially lower nitrogen use. Attainment of high yields on existing croplands of underyielding nations is of great importance if global crop demand is to be met with minimal environmental impacts.
Unravelling the factors determining the allocation of carbon to various plant organs is one of the great challenges of modern plant biology. Studying allocation under close to natural conditions requires non-invasive methods, which are now becoming available for measuring plants on a par with those developed for humans. By combining magnetic resonance imaging (MRI) and positron emission tomography (PET), we investigated three contrasting root/shoot systems growing in sand or soil, with respect to their structures, transport routes and the translocation dynamics of recently fixed photoassimilates labelled with the short-lived radioactive carbon isotope (11)C. Storage organs of sugar beet (Beta vulgaris) and radish plants (Raphanus sativus) were assessed using MRI, providing images of the internal structures of the organs with high spatial resolution, and while species-specific transport sectoralities, properties of assimilate allocation and unloading characteristics were measured using PET. Growth and carbon allocation within complex root systems were monitored in maize plants (Zea mays), and the results may be used to identify factors affecting root growth in natural substrates or in competition with roots of other plants. MRI-PET co-registration opens the door for non-invasive analysis of plant structures and transport processes that may change in response to genomic, developmental or environmental challenges. It is our aim to make the methods applicable for quantitative analyses of plant traits in phenotyping as well as in understanding the dynamics of key processes that are essential to plant performance.
Identification of fungus-infected tomato seeds based on fullfield optical coherence tomography
- T Bharti
- B Yoon
- Lee
Bharti, T. Yoon, B. Lee, Identification of fungus-infected tomato seeds based on fullfield optical coherence tomography, Curr. Opt. Photon. 3 (2019) 571-576.
Inception-v4, Inception-ResNet and the impact of residual connections on learning
- C Szegedy
- S Ioffe
- V Vanhoucke
C. Szegedy, S. Ioffe, V. Vanhoucke, A.A. Alemi, Inception-v4, Inception-ResNet and
the impact of residual connections on learning, in: Proceedings of the Thirty-First
AAAI Conference on Artificial Intelligence (AAAI-17), 4278-7284.
Mapping tissue optical attenuation to identify cancer using optical coherence tomography
- R A Mclaughlin
- L Scolaro
- P Robbins
- C Saunders
- S L Jacques
- D D Sampson
R.A. McLaughlin, L. Scolaro, P. Robbins, C. Saunders, S.L. Jacques, D.D. Sampson,
Mapping tissue optical attenuation to identify cancer using optical coherence
tomography, Med. Image Comput. Assist. Interv. 12 (2009) 657-664.