Figure - available from: Frontiers in Plant Science
This content is subject to copyright.
Example of a residual network with 18 parameter layers, with dotted shortcuts increasing the dimensions.
Source publication
Introduction
Blackheart is one of the most common physiological diseases in potatoes during storage. In the initial stage, black spots only occur in tissues near the potato core and cannot be detected from an outward appearance. If not identified and removed in time, the disease will seriously undermine the quality and sale of theentire batch of po...
Similar publications
The present study investigates the relationship between the near-infrared (NIR) spectral characteristics of cross-linked polyethylene (XLPE) insulation materials and their macroscopic properties, with the aim of establishing a reference framework for non-destructive material aging analysis. Accelerated thermal aging tests were conducted on samples...
Citations
... In previous research, edge extraction analysis methods based on image enhancement have been widely used for anomaly detection and quality inspection in industrial assembly lines due to their low computational requirements [5][6][7][8][9][10]. These methods were among the earliest machine vision techniques applied to potato sprout eye identification. ...
Seed potatoes without sprouts usually need to be manually selected in mechanized production, which has been the bottleneck of efficiency. A fast and efficient object recognition algorithm is required for the additional removal process to identify unqualified seed potatoes. In this paper, a lightweight deep learning algorithm, YOLOv8_EBG, is proposed to both improve the detection performance and reduce the model parameters. The ECA attention mechanism was introduced in the backbone and neck of the model to more accurately extract and fuse sprouting features. To further reduce the model parameters, Ghost convolution and C3ghost were introduced to replace the normal convolution and C2f blocks in vanilla YOLOv8n. In addition, a bi-directional feature pyramid network is integrated in the neck part for multi-scale feature fusion to enhance the detection accuracy. The experimental results from an isolated test dataset show that the proposed algorithm performs better in detecting sprouts under natural light conditions, achieving an mAP0.5 of 95.7% and 91.9% AP for bud recognition. Compared to the YOLOv8n model, the improved model showed a 6.5% increase in mAP0.5, a 12.9% increase in AP0.5 for bud recognition, and a 5.6% decrease in the number of parameters. Additionally, the improved algorithm was applied and tested on mechanized sorting equipment, and the accuracy of seed potato detection was as high as 92.5%, which was sufficient to identify and select sprouted potatoes, an indispensable step since only sprouted potatoes can be used as seed potatoes. The results of the study can provide technical support for subsequent potato planting intelligence.
... Different models were compared. The results showed a prediction accuracy of 0.971 for the test set of black-cored and healthy potatoes when the original spectrum was combined with the improved ResNet model [56]. Binnkowski used near-infrared spectroscopy combined with the partial least-squares method and an inverse neural network to establish a discrimination model for potato late blight and black tibia disease. ...
... Experimental setup: plant material, inoculation protocol, symptom development, and spectral measurement[54]; b example images of potatoes with different blackheart grades[56] Page 10 of 13 Ren et al. Chem. ...
NIR spectroscopy-based detection technology is an analytical methodology that utilises the absorption, reflection, and transmission properties of near-infrared light when interacting with a variety of substances. The technique facilitates the assessment of the composition and characteristics of the materials being analysed. Notably, NIR spectroscopy is characterised by its nondestructive nature, rapid execution, high sensitivity, ease of operation, and efficiency in analysis. In recent years, this technology has been widely applied and expanded in many fields, such as food analysis, biology, and medicine. Root crops, including but not limited to potatoes, cassava, yams, and sweet potatoes, are vital nutritional components of human diets and also serve as critical raw materials in food processing and industrial applications. The significance of these crops is underscored by their impact on consumer health and the economic viability of enterprises, thereby highlighting the importance of effective detection methods for these crops. NIR spectroscopy detection technology is capable of conducting thorough evaluations of both the internal qualities (e.g., starch, protein, sugars, and soluble solids) and the external qualities (e.g., appearance, morphology, pest infestations, and diseases) of root crops. In comparison with alternative spectroscopic techniques, NIR spectroscopy offers a more straightforward approach for the detection and analysis of root crop samples, whilst preserving the integrity of the samples. This emphasises the significant potential of NIR spectroscopy for real-time online monitoring of root crops. The present paper provides a concise overview of the principles underlying NIR spectroscopy detection technology and synthesises research findings regarding its application in the quality assessment of root crops. It emphasises recent advancements in the field, particularly concerning sample pretreatment, spectral collection and processing, and model development. The discussion further encompasses the advantages and limitations of NIR spectroscopy technology, along with the primary challenges encountered in its practical applications and prospects for future development.
Graphical Abstract
... Each internal defect in potatoes affects the Vis-SWNIR spectra distinctly. Blackheart, characterized by black cortical tissue, significantly increases absorbency, particularly affecting wavelengths in the 650-750 nm range [25,36]. Internal browning, which alters the internal structure of tubers, increases the transmittance [37]. ...
Potatoes are a staple food crop consumed worldwide, with their significance extending from household kitchens to large-scale food processing industries. Their market value and quality are often compromised by various internal defects such as pythium, bruising, internal browning, hollow heart, gangrene, blackheart, internal sprouting, and dry rot. This study aimed to classify internal-based defects and investigate the quantification of internal defective areas in potatoes using visible and short-wavelength near-infrared spectroscopy. The acquisition of the spectral data of potato tubers was performed using a spectrometer with a wavelength range of 400–1100 nm. The classification of internal-based defects was performed using partial least squares discriminant analysis (PLS-DA), while the quantification of the internal defective area was based on partial least squares regression (PLSR). The PLS-DA double cross-validation accuracy for the distinction between non-defective and all internally defective potatoes was 90.78%. The double cross-validation classification accuracy achieved for pythium, bruising, and non-defective categories was 91.03%. The internal defective area model based on PLSR achieved a correlation coefficient of double cross-validation of 0.91 and a root mean square error of double cross-validation of 0.85 cm². This study makes a valuable contribution to advancing nondestructive techniques for evaluating internal defects in potatoes.