Fig 2 - uploaded by Loris Nanni
Content may be subject to copyright.
Source publication
Plankton is the most fundamental component of ocean ecosystems, due to its function at many levels of the oceans food chain. The variations of its distribution are useful indicators for oceanic or climatic events; therefore, the study of plankton distribution is crucial to protect marine ecosystems. Currently, much research is concentrated on the a...
Context in source publication
Context 1
... of the current problem. The idea is that when CNNs are trained on images, the first convolution layers either resemble Gabor filters or color blobs that tend to be generalizable, thus the knowledge and skills learned in previous tasks can be applied to a novel task. In the experiments, we test and combine the following different CNN architectures (Fig. 2) "fine-tuned" on the current ...
Similar publications
A universal scaling relationship exists between organism abundance and body size1,2. Within ocean habitats this relationship deviates from that generally observed in terrestrial systems2-4, where marine macro-fauna display steeper size-abundance scaling than expected. This is indicative of a fundamental shift in food-web organization, yet a conclus...
Citations
... Dalam hal klasifikasi plankton secara otomatis, terdapat tiga tantangan utama yang perlu dihadapi, yaitu citra plankton yang seringkali tidak jelas karena resolusi rendah dan objek yang sulit diidentifikasi bahkan oleh ahli dibidangnya, ukuran dataset yang terlalu kecil jika dibandingkan dengan masalah klasifikasi gambar lainnya sehingga membuat training model menjadi lebih sulit, serta terjadi ketidakseimbangan pada dataset di antara kelas, dan pergeseran data dapat terjadi pada saat training dan testing (Lumini & Nanni, 2019b). ...
The oceans, covering approximately 71% of the Earth's surface, are teeming with life, including plankton, which are microscopic organisms forming the base of the marine food chain. Phytoplankton and zooplankton, the two main categories of plankton, play a vital role in maintaining the balance of marine ecosystems. In early studies, plankton identification relied heavily on manual methods, which were costly and impractical for large-scale use. However, automatic plankton classification also faces several challenges, such as unclear plankton images due to low resolution, small dataset sizes, and data imbalance across some classes. Current plankton research is divided into two approaches: feature descriptors and deep learning. While these methods are related in terms of function and history, they are treated as separate approaches. Therefore, a hybrid approach is used, with Convolutional Neural Networks for feature extraction and Extreme Learning Machines for classification. Additionally, SMOTE is applied to address class imbalance, and Flask is chosen as the web framework for model implementation to ensure easy accessibility. The testing of the CNN-ELM model showed that the best model achieved an accuracy of 98.89% with a training-to-testing data ratio of 80:20, using 16 CNN filters and 1000 ELM hidden nodes. However, this model faced difficulties in classifying the Nitzschia, Pleurosigma, and Thalassiosira classes. This research aims to improve understanding and decision-making in the field of marine science.
... Dengan kemajuan teknologi, metode berbasis pencitraan telah muncul sebagai solusi yang lebih efisien [4]. Namun, klasifikasi otomatis plankton menghadapi tantangan seperti kualitas gambar yang rendah, ukuran dataset yang kecil, dan ketidakseimbangan kelas antar spesies [5]. ...
Penelitian ini bertujuan untuk meningkatkan akurasi dalam klasifikasi plankton secara otomatis dengan pendekatan hibrida CNN-ELM. Menggunakan Convolutional Neural Network (CNN) untuk ekstraksi fitur dan Extreme Learning Machine (ELM) sebagai pengklasifikasi, model ini dirancang untuk mengatasi tantangan citra plankton yang buram, dataset kecil, dan ketidakseimbangan kelas. SMOTE digunakan untuk menangani ketidakseimbangan data. Dari implementasi SMOTE dengan metode interpolasi, permasalahan ketidakseimbangan kelas berhasil diatasi dengan menjadikan jumlah data latih sama rata untuk setiap kelas. Dari hasil pengujian, konfigurasi dengan 32 filter dan 2000 hidden node serta 64 filter dan 2000 hidden node memberikan performa terbaik dengan akurasi 97,78%. Sebaliknya, model dengan 64 filter dan 4000 hidden node menunjukkan performa terendah dengan akurasi 82,78% yang diakibatkan overfitting. Analisis confusion matrix mengungkapkan kinerja tinggi pada beberapa kelas plankton, namun masih kesalahan klasifikasi sering terjadi pada kelas seperti Alexandrium, Noctiluca, dan Nitzschia. Temuan ini menunjukkan bahwa konfigurasi dengan filter dan node yang lebih kompleks tidak selalu menghasilkan kinerja lebih baik. Penelitian ini diharapkan dapat mendukung pengambilan keputusan di bidang kelautan.
... The authors compare the various fusion models used by them with FUS Hand [53], Baseline [54], Gaussian SVM [54], and "multiple kernel learning" (MKL) (with three kernels) [54]. Among their fusion models, the FUS 2R+FUS 1R model performs the best, and its architecture is explained in Table XI. ...
In recent years, there has been an enormous interest in using deep learning to classify underwater images to identify various objects like fishes, plankton, coral reefs, seagrass, submarines, and gestures of sea-divers. This classification is essential for measuring the water bodies' health and quality and protecting the endangered species. Further, it has applications in oceanography, marine economy and defense, environment protection, and underwater exploration and human-robot col-laborative tasks. This paper presents a survey of deep learning techniques for performing the underwater image classification. We underscore the similarities and differences of several methods. We believe that underwater image classification is one of the killer application that would test the ultimate success of deep learning techniques. Towards realizing that goal, this survey seeks to inform researchers about state-of-the-art on deep learning on underwater images and also motivate them to push its frontiers forward. Index Terms-Deep neural networks, artificial intelligence, autonomous underwater vehicle, transfer learning.
... The authors compare the various fusion models used by them with FUS Hand [19], Baseline [38], Gaussian SVM [38], and "multiple kernel learning" (MKL) (with three kernels) [38]. Among their fusion models, the FUS 2R+FUS 1R model performs the best, and its architecture is explained in Table XI. ...
In recent years, there has been an enormous interest in using deep learning to classify underwater images to identify various objects like fishes, plankton, coral reefs, seagrass, submarines, and gestures of sea-divers. This classification is essential for measuring the water bodies' health and quality and protecting the endangered species. Further, it has applications in oceanography, marine economy and defense, environment protection, and underwater exploration and human-robot collaborative tasks. This paper presents a survey of deep learning techniques for performing the underwater image classification. We underscore the similarities and differences of several methods. We believe that underwater image classification is one of the killer application that would test the ultimate success of deep learning techniques. Towards realizing that goal, this survey seeks to inform researchers about state-of-the-art on deep learning on underwater images and also motivate them to push its frontiers forward.
... Few underwater imaging devices have been designed and tested for marine zooplankton studies for many years which erased the need of manual sampling by plankton nets up to major extent [8]. But the abrasive quality and size of underwater images dataset carries a challenging task in analyzing and classification due to unclear morphological traits and training of automatic classification model [9] [10]. However, focusing the critical importance of zooplankton, it is still in the interest of researchers to develop more advanced, robust and automatic system for its imaging and classification [11]. ...
Zooplankton is enormously diverse and fundamental group of microorganisms that exists in almost every freshwater body, determining its ecology and play a vital role in food chain. Considering the significance of zooplankton, the study of freshwater zooplankton is very essential which intensely relies on the classification of images. However, the routine manual analysis and classification is laborious, time consuming and expensive, and poses a significant challenge to experts. Thus, for recent decade much research is focused on the development of underwater imaging technologies and intelligent classification system of zooplankton. This work presents devotion to observation of freshwater zooplankton by designed underwater microscope and modeling the system for automatic classification among four different taxa. Unlike most of the existing zooplankton image classification systems, this model is trained on a comparatively small dataset collected from freshwater by designed underwater microscope. Transfer learning of pre-trained AlexNet Convolutional Neural Network (CNN) model proved to be a potential approach in the system design. Among four networks trained over two datasets, the best overall classification accuracy of up to 93.1%, comparable to other existing systems was achieved on test dataset (92.5% for Calanoid and Cyclopoid (Female), 90% for Cyclopoid (Male) and 97.5% for Daphnia). Graphical User Interface (GUI) of the model constructed on MATLAB, makes it easy for the users to collect images for building database, train network and to classify images of different taxa. Moreover, the designed system is adaptable to the addition of more classes in the future.
... Krizhevsky et al. [12] introduce AlexNet and it achieves the best performance on ImageNet Large Scale Visual Recognition Challenge in 2012 AlexNet consists of eight convolution layers, max-pooling, and three are fully connected layers, as shown in Figure 3. Softmax is used as activation for the fully connected layers. This architectural image is redrawn from [13]. ...
Deep learning technology has a better result when trained using an abundant amount of data. However, collecting such data is expensive and time consuming. On the other hand, limited data often be the inevitable condition. To increase the number of data, data augmentation is usually implemented. By using it, the original data are transformed, by rotating, shifting, or both, to generate new data artificially. In this paper, generative adversarial networks (GAN) and deep convolutional GAN (DCGAN) are used for data augmentation. Both approaches are applied for diseases detection. The performance of the tea diseases detection on the augmented data is evaluated using various deep convolutional neural network (DCNN) including AlexNet, DenseNet, ResNet, and Xception. The experimental results indicate that the highest GAN accuracy is obtained by DenseNet architecture, which is 88.84%, baselines accuracy on the same architecture is 86.30%. The results of DCGAN accuracy on the use of the same architecture show a similar trend, which is 88.86%.
... Dai et al. (2016a) present a similar approach in the design of a zooplankton classifier. Other authors combine CNNs with different machine learning techniques, including active learning (Bochinski et al., 2018), hybrid systems (Dai et al., 2016b), parallel networks , imbalance learning (Lee et al., 2016) or different forms of information fusion (Cui et al., 2018;Lumini and Nanni, 2019). It should be noted that the improvements on plankton classification described in these papers may not be directly transferable to quantification systems, as classification and quantification are two different tasks. ...
The study of marine plankton data is vital to monitor the health of the world’s oceans. In recent decades, automatic plankton recognition systems have proved useful to address the vast amount of data collected by specially engineered in situ digital imaging systems. At the beginning, these systems were developed and put into operation using traditional automatic classification techniques, which were fed with hand-designed local image descriptors (such as Fourier features), obtaining quite successful results. In the past few years, there have been many advances in the computer vision community with the rebirth of neural networks. In this paper, we leverage how descriptors computed using convolutional neural networks trained with out-of-domain data are useful to replace hand-designed descriptors in the task of estimating the prevalence of each plankton class in a water sample. To achieve this goal, we have designed a broad set of experiments that show how effective these deep features are when working in combination with state-of-the-art quantification algorithms.
... -A modern tendency of plankton research in the world ocean is the ecosystem approach [1]. Models of plankton ecosystems refer to complex systems with a great number of degrees of freedom and complex nonlinear relationships. ...
... To measure individual (differential) species characteristics, the spatiotemporal scale of individual plankton species should be realized. The plankton species can be classified according to different criteria connected with metabolism (autotrophic and heterotrophic), sizes (macro-, meso-, micro-, nano-, and picoplankton), life cycle (meroplankton and holoplankton), or taxonomy [1]. These data are informative for investigation of biodiversity and dynamics of water ecosystems. ...
Planktonic organisms including phyto-, zoo-, and mixoplankton are key components of aquatic ecosystems and respond quickly to changes in the environment, therefore their monitoring is vital to follow and understand these changes. Advances in imaging technology have enabled novel possibilities to study plankton populations, but the manual classification of images is time consuming and expert-based, making such an approach unsuitable for large-scale application and urging for automatic solutions for the analysis, especially recognizing the plankton species from images. Despite the extensive research done on automatic plankton recognition, the latest cutting-edge methods have not been widely adopted for operational use. In this paper, a comprehensive survey on existing solutions for automatic plankton recognition is presented. First, we identify the most notable challenges that make the development of plankton recognition systems difficult and restrict the deployment of these systems for operational use. Then, we provide a detailed description of solutions found in plankton recognition literature. Finally, we propose a workflow to identify the specific challenges in new datasets and the recommended approaches to address them. Many important challenges remain unsolved including the following: (1) the domain shift between the datasets hindering the development of an imaging instrument independent plankton recognition system, (2) the difficulty to identify and process the images of previously unseen classes and non-plankton particles, and (3) the uncertainty in expert annotations that affects the training of the machine learning models. To build harmonized instrument and location agnostic methods for operational purposes these challenges should be addressed in future research.
Planktonic organisms are key components of aquatic ecosystems and respond quickly to changes in the environment, therefore their monitoring is vital to understand the changes in the environment. Yet, monitoring plankton at appropriate scales still remains a challenge, limiting our understanding of functioning of aquatic systems and their response to changes. Modern plankton imaging instruments can be utilized to sample at high frequencies, enabling novel possibilities to study plankton populations. However, manual analysis of the data is costly, time consuming and expert based, making such approach unsuitable for large-scale application and urging for automatic solutions. The key problem related to the utilization of plankton datasets through image analysis is plankton recognition. Despite the large amount of research done, automatic methods have not been widely adopted for operational use. In this paper, a comprehensive survey on existing solutions for automatic plankton recognition is presented. First, we identify the most notable challenges that that make the development of plankton recognition systems difficult. Then, we provide a detailed description of solutions for these challenges proposed in plankton recognition literature. Finally, we propose a workflow to identify the specific challenges in new datasets and the recommended approaches to address them. For many of the challenges, applicable solutions exist. However, important challenges remain unsolved: 1) the domain shift between the datasets hindering the development of a general plankton recognition system that would work across different imaging instruments, 2) the difficulty to identify and process the images of previously unseen classes, and 3) the uncertainty in expert annotations that affects the training of the machine learning models for recognition. These challenges should be addressed in the future research.
