Asish Bera
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Asish Bera has a total of 36 co-authors. Here are the top co-authors by number of shared publications:
- Debotosh Bhattacharjee
Department of Computer Science and Engineering, Jadavpur University, Kolkata, India
19 shared publications Debotosh Bhattacharjee
9 shared publicationsMita Nasipuri
Department of Computer Science and Engineering Jadavpur University Kolkata WB India
8 shared publications- Mita Nasipuri
Department of Computer Science and Engineering, Jadavpur University, Kolkata, India
8 shared publications - Ondrej Krejcar
Department of Cybernetics and Biomedical Engineering, Faculty of Electrical Engineering and Computer Science (FEEC), VSBāTechnical University of Ostrava, Ostrava, Czechia
8 shared publications Ondrej Krejcar
3 shared publications
Latest Publications
- Open Access
- 16 May 2026
- Published byĀ Springer NatureĀ inĀ Parasites & Vectors
- Vol.Ā 19 Ā (1)
A hybrid machine learning approach for automated malaria diagnosis from thin blood smear images
Background: Malaria remains a significant global health challenge, requiring diagnostic approaches that are rapid, cost-effective, and accurate. The present study proposes a hybrid deep learning framework for malaria diagnosis using thin blood smear cell images obtained from the National Institute of Malaria Research (NIMR), a leading research institution of the Indian Council of Medical Research (ICMR). Methods: A total of 15,938 thin blood smear microscopic images were preprocessed to enhance the visibility of cellular structures and suppress background noise. Segmentation was performed using the Otsu thresholding method, which automatically determines the optimal threshold to maximize cell contrast and improve feature extraction. The processed images were then used to train the proposed Hybrid Inception-v3 convolutional neural network (CNN), specifically designed for automated cell classification. To establish the robustness of the proposed CNN, comparative and statistical analyses were performed against conventional diagnostic techniques and other state-of-the-art machine learning models. Results: The proposed hybrid CNN model achieved a training accuracy of 98.9%, sensitivity of 97.3%, specificity of 99.9%, along with high F 1 score and area under the curve (AUC) values. These results were obtained using the ICMR-NIMR microscopic image dataset, demonstrating the robustness and generalizability of our approach. The model consistently outperformed other evaluated methods, including state-of-the-art machine learning classifiers, and demonstrated performance comparable to or better than conventional microscopy-based examinations. Conclusions: The hybrid deep learning framework demonstrated robust performance for malaria diagnosis using microscopic images, suggesting potential to reduce reliance on expert microscopists. These findings highlight its potential to improve diagnostic precision and strengthen malaria control efforts, especially in resource-limited settings.
- 16 May 2026
- Published byĀ Springer NatureĀ inĀ Parasites & Vectors
- Vol.Ā 19 Ā (1)
Open AccessA hybrid machine learning approach for automated malaria diagnosis from thin blood smear images
Background: Malaria remains a significant global health challenge, requiring diagnostic approaches that are rapid, cost-effective, and accurate. The present study proposes a hybrid deep learning framework for malaria diagnosis using thin blood smear cell images obtained from the National Institute of Malaria Research (NIMR), a leading research institution of the Indian Council of Medical Research (ICMR). Methods: A total of 15,938 thin blood smear microscopic images were preprocessed to enhance the visibility of cellular structures and suppress background noise. Segmentation was performed using the Otsu thresholding method, which automatically determines the optimal threshold to maximize cell contrast and improve feature extraction. The processed images were then used to train the proposed Hybrid Inception-v3 convolutional neural network (CNN), specifically designed for automated cell classification. To establish the robustness of the proposed CNN, comparative and statistical analyses were performed against conventional diagnostic techniques and other state-of-the-art machine learning models. Results: The proposed hybrid CNN model achieved a training accuracy of 98.9%, sensitivity of 97.3%, specificity of 99.9%, along with high F 1 score and area under the curve (AUC) values. These results were obtained using the ICMR-NIMR microscopic image dataset, demonstrating the robustness and generalizability of our approach. The model consistently outperformed other evaluated methods, including state-of-the-art machine learning classifiers, and demonstrated performance comparable to or better than conventional microscopy-based examinations. Conclusions: The hybrid deep learning framework demonstrated robust performance for malaria diagnosis using microscopic images, suggesting potential to reduce reliance on expert microscopists. These findings highlight its potential to improve diagnostic precision and strengthen malaria control efforts, especially in resource-limited settings.
- Open Access
- 26 August 2025
- Published byĀ Institute of Electrical and Electronics Engineers (IEEE)Ā inĀ IEEE Access
- Vol.Ā 13, 151137-151152
Advanced Encryption Using Generative Adversarial Network for Enhancing Security of Non-Fungible Tokens (NFTs)
Non-fungible tokens (NFTs) represent a unique form of digital asset stored on a blockchain, encompassing a wide array of assets from digital art to real-world commodities. While NFTs offer robust security through decentralization and smart contract enforcement, they are not impervious to cyber threats. Traditionally, NFTs are transferred over blockchain platforms via smart contracts. The primary objective of this study is to introduce a cryptographic framework that integrates an advanced encryption algorithm layer to fortify NFT image transfers, reinforcing content protection in an evolving digital ecosystem. Various pixel-based encryption algorithms have been implemented using image processing techniques and compared considering their levels of encryption and execution times. Based on the comparison, an advanced encryption algorithm has been devised with an added level of encryption and fast execution time. The proposed encryption algorithm leverages alpha composition with a randomly generated image obtained using a Generative Adversarial Network (GAN). The GAN-generated image is infused with Gaussian noise to deter decryption by unauthorized algorithms, thereby enhancing resilience against cryptanalysis. The encrypted image metadata would be transferred through both on-chain (i.e., blockchain) and off-chain methods to facilitate efficiency and security during NFT transfers. The image decryption protocol mandates the new owner/receiver of NFT to possess the correct private identification, transaction key, and off-chain data, ensuring exclusive access while automatically revoking ownership from the previous holder upon successful transfer. This approach not only secures the transfer process but also minimizes the risk of data leaks, as intermediaries should not have access to the complete metadata. This cryptographic approach not only safeguards digital assets but also aligns with the futuristic vision of secure transactions in the metaverse and Web 3.0. The experimental analysis based on the security level and encryption time based on multiple encryption algorithms justifies the benefits of the proposed method.
- 26 August 2025
- Published byĀ Institute of Electrical and Electronics Engineers (IEEE)Ā inĀ IEEE Access
- Vol.Ā 13, 151137-151152
Open AccessAdvanced Encryption Using Generative Adversarial Network for Enhancing Security of Non-Fungible Tokens (NFTs)
Non-fungible tokens (NFTs) represent a unique form of digital asset stored on a blockchain, encompassing a wide array of assets from digital art to real-world commodities. While NFTs offer robust security through decentralization and smart contract enforcement, they are not impervious to cyber threats. Traditionally, NFTs are transferred over blockchain platforms via smart contracts. The primary objective of this study is to introduce a cryptographic framework that integrates an advanced encryption algorithm layer to fortify NFT image transfers, reinforcing content protection in an evolving digital ecosystem. Various pixel-based encryption algorithms have been implemented using image processing techniques and compared considering their levels of encryption and execution times. Based on the comparison, an advanced encryption algorithm has been devised with an added level of encryption and fast execution time. The proposed encryption algorithm leverages alpha composition with a randomly generated image obtained using a Generative Adversarial Network (GAN). The GAN-generated image is infused with Gaussian noise to deter decryption by unauthorized algorithms, thereby enhancing resilience against cryptanalysis. The encrypted image metadata would be transferred through both on-chain (i.e., blockchain) and off-chain methods to facilitate efficiency and security during NFT transfers. The image decryption protocol mandates the new owner/receiver of NFT to possess the correct private identification, transaction key, and off-chain data, ensuring exclusive access while automatically revoking ownership from the previous holder upon successful transfer. This approach not only secures the transfer process but also minimizes the risk of data leaks, as intermediaries should not have access to the complete metadata. This cryptographic approach not only safeguards digital assets but also aligns with the futuristic vision of secure transactions in the metaverse and Web 3.0. The experimental analysis based on the security level and encryption time based on multiple encryption algorithms justifies the benefits of the proposed method.
- 19 July 2025
- Published byĀ SAGE PublicationsĀ inĀ Proceedings of the Institution of Mechanical Engineers, Part P: Journal of Sports Engineering and Technology
- Vol.Ā 240 Ā (4) , 1129-1141
Women sport actions dataset for visual classification using small-scale training data
Sports action classification representing complex body postures and player-object interactions, is an emerging area in image-based sports analysis. Some works have contributed to automated sports action recognition using machine learning techniques over the past decades. However, sufficient image datasets representing womenās sports actions with enough intra- and inter-class variations are not available to the researchers. To overcome this limitation, this work presents a new dataset named WomenSports for womenās sports classification using small-scale training data. This dataset includes a variety of sports activities, covering wide variations in movements, environments, and interactions among players. In addition, this study proposes a convolutional neural network (CNN) for deep feature extraction. A channel attention scheme upon local contextual regions is applied to refine and enhance feature representation. The experiments are carried out on three different sports datasets and one dance dataset for generalizing the proposed algorithm, and the performances on these datasets are noteworthy. The deep learning method achieves 89.15% top-1 classification accuracy using ResNet-50 on the proposed WomenSports dataset, which is publicly available for research at Mendeley Data.
- 19 July 2025
- Published byĀ SAGE PublicationsĀ inĀ Proceedings of the Institution of Mechanical Engineers, Part P: Journal of Sports Engineering and Technology
- Vol.Ā 240 Ā (4) , 1129-1141
Women sport actions dataset for visual classification using small-scale training data
Sports action classification representing complex body postures and player-object interactions, is an emerging area in image-based sports analysis. Some works have contributed to automated sports action recognition using machine learning techniques over the past decades. However, sufficient image datasets representing womenās sports actions with enough intra- and inter-class variations are not available to the researchers. To overcome this limitation, this work presents a new dataset named WomenSports for womenās sports classification using small-scale training data. This dataset includes a variety of sports activities, covering wide variations in movements, environments, and interactions among players. In addition, this study proposes a convolutional neural network (CNN) for deep feature extraction. A channel attention scheme upon local contextual regions is applied to refine and enhance feature representation. The experiments are carried out on three different sports datasets and one dance dataset for generalizing the proposed algorithm, and the performances on these datasets are noteworthy. The deep learning method achieves 89.15% top-1 classification accuracy using ResNet-50 on the proposed WomenSports dataset, which is publicly available for research at Mendeley Data.
- 16 October 2024
- Published byĀ Institute of Electrical and Electronics Engineers (IEEE)Ā inĀ IEEE Transactions on AgriFood Electronics
- Vol.Ā 3 Ā (1) , 121-133
RAFA-Net: Region Attention Network for Food Items and Agricultural Stress Recognition
Deep convolutional neural networks (CNNs) have facilitated remarkable success in recognizing various food items and agricultural stress. A decent performance boost has been witnessed in solving the agro-food challenges by mining and analyzing region-based partial feature descriptors. Also, computationally expensive ensemble learning schemes fusing multiple CNNs have been studied in earlier works. This work proposes a region attention scheme for modeling long-range dependencies by building a correlation among different regions within an input image. The attention method enhances feature representation by learning the usefulness of context information from complementary regions. Spatial pyramidal pooling and average pooling pairs aggregate partial descriptors into a holistic representation. Both pooling methods establish spatial and channelwise relationships without incurring extra parameters. A context gating scheme is applied to refine the descriptiveness of weighted attentional features, which is relevant for classification. The proposed region attention network for food items and agricultural stress recognition method, dubbed RAFA-Net, has been experimented on three public food datasets, and has achieved state-of-the-art performances with distinct margins. The highest top-1 accuracy of RAFA-Net is 91.69%, 91.56%, and 96.97% on the UECFood-100, UECFood-256, and MAFood-121 datasets, respectively. In addition, better accuracies have been achieved on two benchmark agricultural stress datasets. The best top-1 accuracies on the Insect Pest (IP-102) and PlantDoc-27 plant disease datasets are 92.36%, and 85.54%, respectively; implying RAFA-Net's generalization capability.
- 16 October 2024
- Published byĀ Institute of Electrical and Electronics Engineers (IEEE)Ā inĀ IEEE Transactions on AgriFood Electronics
- Vol.Ā 3 Ā (1) , 121-133
RAFA-Net: Region Attention Network for Food Items and Agricultural Stress Recognition
Deep convolutional neural networks (CNNs) have facilitated remarkable success in recognizing various food items and agricultural stress. A decent performance boost has been witnessed in solving the agro-food challenges by mining and analyzing region-based partial feature descriptors. Also, computationally expensive ensemble learning schemes fusing multiple CNNs have been studied in earlier works. This work proposes a region attention scheme for modeling long-range dependencies by building a correlation among different regions within an input image. The attention method enhances feature representation by learning the usefulness of context information from complementary regions. Spatial pyramidal pooling and average pooling pairs aggregate partial descriptors into a holistic representation. Both pooling methods establish spatial and channelwise relationships without incurring extra parameters. A context gating scheme is applied to refine the descriptiveness of weighted attentional features, which is relevant for classification. The proposed region attention network for food items and agricultural stress recognition method, dubbed RAFA-Net, has been experimented on three public food datasets, and has achieved state-of-the-art performances with distinct margins. The highest top-1 accuracy of RAFA-Net is 91.69%, 91.56%, and 96.97% on the UECFood-100, UECFood-256, and MAFood-121 datasets, respectively. In addition, better accuracies have been achieved on two benchmark agricultural stress datasets. The best top-1 accuracies on the Insect Pest (IP-102) and PlantDoc-27 plant disease datasets are 92.36%, and 85.54%, respectively; implying RAFA-Net's generalization capability.
- Open Access
PND-Net: Plant Nutrition Deficiency and Disease Classification using Graph Convolutional Network
Crop yield production could be enhanced for agricultural growth if various plant nutrition deficiencies, and diseases are identified and detected at early stages. The deep learning methods have proven its superior performances in the automated detection of plant diseases and nutrition deficiencies from visual symptoms in leaves. This article proposes a new deep learning method for plant nutrition deficiencies and disease classification using a graph convolutional network (GNN), added upon a base convolutional neural network (CNN). Sometimes, a global feature descriptor might fail to capture the vital region of a diseased leaf, which causes inaccurate classification of disease. To address this issue, regional feature learning is crucial for a holistic feature aggregation. In this work, region-based feature summarization at multi-scales is explored using spatial pyramidal pooling for discriminative feature representation. A GCN is developed to capacitate learning of finer details for classifying plant diseases and insufficiency of nutrients. The proposed method, called Plant Nutrition Deficiency and Disease Network (PND-Net), is evaluated on two public datasets for nutrition deficiency, and two for disease classification using four CNNs. The best classification performances are: (a) 90.00% Banana and 90.54% Coffee nutrition deficiency; and (b) 96.18% Potato diseases and 84.30% on PlantDoc datasets using Xception backbone. Furthermore, additional experiments have been carried out for generalization, and the proposed method has achieved state-of-the-art performances on two public datasets, namely the Breast Cancer Histopathology Image Classification (BreakHis 40X: 95.50%, and BreakHis 100X: 96.79% accuracy) and Single cells in Pap smear images for cervical cancer classification (SIPaKMeD: 99.18% accuracy). Also, PND-Net achieves improved performances using five-fold cross validation.
Open AccessPND-Net: Plant Nutrition Deficiency and Disease Classification using Graph Convolutional Network
Crop yield production could be enhanced for agricultural growth if various plant nutrition deficiencies, and diseases are identified and detected at early stages. The deep learning methods have proven its superior performances in the automated detection of plant diseases and nutrition deficiencies from visual symptoms in leaves. This article proposes a new deep learning method for plant nutrition deficiencies and disease classification using a graph convolutional network (GNN), added upon a base convolutional neural network (CNN). Sometimes, a global feature descriptor might fail to capture the vital region of a diseased leaf, which causes inaccurate classification of disease. To address this issue, regional feature learning is crucial for a holistic feature aggregation. In this work, region-based feature summarization at multi-scales is explored using spatial pyramidal pooling for discriminative feature representation. A GCN is developed to capacitate learning of finer details for classifying plant diseases and insufficiency of nutrients. The proposed method, called Plant Nutrition Deficiency and Disease Network (PND-Net), is evaluated on two public datasets for nutrition deficiency, and two for disease classification using four CNNs. The best classification performances are: (a) 90.00% Banana and 90.54% Coffee nutrition deficiency; and (b) 96.18% Potato diseases and 84.30% on PlantDoc datasets using Xception backbone. Furthermore, additional experiments have been carried out for generalization, and the proposed method has achieved state-of-the-art performances on two public datasets, namely the Breast Cancer Histopathology Image Classification (BreakHis 40X: 95.50%, and BreakHis 100X: 96.79% accuracy) and Single cells in Pap smear images for cervical cancer classification (SIPaKMeD: 99.18% accuracy). Also, PND-Net achieves improved performances using five-fold cross validation.
- Open Access
RAFA-Net: Region Attention Network For Food Items And Agricultural Stress Recognition
Deep Convolutional Neural Networks (CNNs) have facilitated remarkable success in recognizing various food items and agricultural stress. A decent performance boost has been witnessed in solving the agro-food challenges by mining and analyzing of region-based partial feature descriptors. Also, computationally expensive ensemble learning schemes using multiple CNNs have been studied in earlier works. This work proposes a region attention scheme for modelling long-range dependencies by building a correlation among different regions within an input image. The attention method enhances feature representation by learning the usefulness of context information from complementary regions. Spatial pyramidal pooling and average pooling pair aggregate partial descriptors into a holistic representation. Both pooling methods establish spatial and channel-wise relationships without incurring extra parameters. A context gating scheme is applied to refine the descriptiveness of weighted attentional features, which is relevant for classification. The proposed Region Attention network for Food items and Agricultural stress recognition method, dubbed RAFA-Net, has been experimented on three public food datasets, and has achieved state-of-the-art performances with distinct margins. The highest top-1 accuracies of RAFA-Net are 91.69%, 91.56%, and 96.97% on the UECFood-100, UECFood-256, and MAFood-121 datasets, respectively. In addition, better accuracies have been achieved on two benchmark agricultural stress datasets. The best top-1 accuracies on the Insect Pest (IP-102) and PlantDoc-27 plant disease datasets are 92.36%, and 85.54%, respectively; implying RAFA-Net's generalization capability.
Open AccessRAFA-Net: Region Attention Network For Food Items And Agricultural Stress Recognition
Deep Convolutional Neural Networks (CNNs) have facilitated remarkable success in recognizing various food items and agricultural stress. A decent performance boost has been witnessed in solving the agro-food challenges by mining and analyzing of region-based partial feature descriptors. Also, computationally expensive ensemble learning schemes using multiple CNNs have been studied in earlier works. This work proposes a region attention scheme for modelling long-range dependencies by building a correlation among different regions within an input image. The attention method enhances feature representation by learning the usefulness of context information from complementary regions. Spatial pyramidal pooling and average pooling pair aggregate partial descriptors into a holistic representation. Both pooling methods establish spatial and channel-wise relationships without incurring extra parameters. A context gating scheme is applied to refine the descriptiveness of weighted attentional features, which is relevant for classification. The proposed Region Attention network for Food items and Agricultural stress recognition method, dubbed RAFA-Net, has been experimented on three public food datasets, and has achieved state-of-the-art performances with distinct margins. The highest top-1 accuracies of RAFA-Net are 91.69%, 91.56%, and 96.97% on the UECFood-100, UECFood-256, and MAFood-121 datasets, respectively. In addition, better accuracies have been achieved on two benchmark agricultural stress datasets. The best top-1 accuracies on the Insect Pest (IP-102) and PlantDoc-27 plant disease datasets are 92.36%, and 85.54%, respectively; implying RAFA-Net's generalization capability.
- Open Access
Two-Stage Human Verification using HandCAPTCHA and Anti-Spoofed Finger Biometrics with Feature Selection
This paper presents a human verification scheme in two independent stages to overcome the vulnerabilities of attacks and to enhance security. At the first stage, a hand image-based CAPTCHA (HandCAPTCHA) is tested to avert automated bot-attacks on the subsequent biometric stage. In the next stage, finger biometric verification of a legitimate user is performed with presentation attack detection (PAD) using the real hand images of the person who has passed a random HandCAPTCHA challenge. The electronic screen-based PAD is tested using image quality metrics. After this spoofing detection, geometric features are extracted from the four fingers (excluding the thumb) of real users. A modified forward-backward (M-FoBa) algorithm is devised to select relevant features for biometric authentication. The experiments are performed on the Bogazici University (BU) and the IIT-Delhi (IITD) hand databases using the k-nearest neighbor and random forest classifiers. The average accuracy of the correct HandCAPTCHA solution is 98.5%, and the false accept rate of a bot is 1.23%. The PAD is tested on 255 subjects of BU, and the best average error is 0%. The finger biometric identification accuracy of 98% and an equal error rate (EER) of 6.5% have been achieved for 500 subjects of the BU. For 200 subjects of the IITD, 99.5% identification accuracy, and 5.18% EER are obtained.
Open AccessTwo-Stage Human Verification using HandCAPTCHA and Anti-Spoofed Finger Biometrics with Feature Selection
This paper presents a human verification scheme in two independent stages to overcome the vulnerabilities of attacks and to enhance security. At the first stage, a hand image-based CAPTCHA (HandCAPTCHA) is tested to avert automated bot-attacks on the subsequent biometric stage. In the next stage, finger biometric verification of a legitimate user is performed with presentation attack detection (PAD) using the real hand images of the person who has passed a random HandCAPTCHA challenge. The electronic screen-based PAD is tested using image quality metrics. After this spoofing detection, geometric features are extracted from the four fingers (excluding the thumb) of real users. A modified forward-backward (M-FoBa) algorithm is devised to select relevant features for biometric authentication. The experiments are performed on the Bogazici University (BU) and the IIT-Delhi (IITD) hand databases using the k-nearest neighbor and random forest classifiers. The average accuracy of the correct HandCAPTCHA solution is 98.5%, and the false accept rate of a bot is 1.23%. The PAD is tested on 255 subjects of BU, and the best average error is 0%. The finger biometric identification accuracy of 98% and an equal error rate (EER) of 6.5% have been achieved for 500 subjects of the BU. For 200 subjects of the IITD, 99.5% identification accuracy, and 5.18% EER are obtained.
- Open Access
Fusion Based Hand Geometry Recognition Using Dempster-Shafer Theory
This paper presents a new technique for person recognition based on the fusion of hand geometric features of both the hands without any pose restrictions. All the features are extracted from normalized left and right hand images. Fusion is applied at feature level and also at decision level. Two probability based algorithms are proposed for classification. The first algorithm computes the maximum probability for nearest three neighbors. The second algorithm determines the maximum probability of the number of matched features with respect to a thresholding on distances. Based on these two highest probabilities initial decisions are made. The final decision is considered according to the highest probability as calculated by the Dempster-Shafer theory of evidence. Depending on the various combinations of the initial decisions, three schemes are experimented with 201 subjects for identification and verification. The correct identification rate found to be 99.5%, and the False Acceptance Rate (FAR) of 0.625% has been found during verification.
Open AccessFusion Based Hand Geometry Recognition Using Dempster-Shafer Theory
This paper presents a new technique for person recognition based on the fusion of hand geometric features of both the hands without any pose restrictions. All the features are extracted from normalized left and right hand images. Fusion is applied at feature level and also at decision level. Two probability based algorithms are proposed for classification. The first algorithm computes the maximum probability for nearest three neighbors. The second algorithm determines the maximum probability of the number of matched features with respect to a thresholding on distances. Based on these two highest probabilities initial decisions are made. The final decision is considered according to the highest probability as calculated by the Dempster-Shafer theory of evidence. Depending on the various combinations of the initial decisions, three schemes are experimented with 201 subjects for identification and verification. The correct identification rate found to be 99.5%, and the False Acceptance Rate (FAR) of 0.625% has been found during verification.
- 8 October 2024
- Published byĀ Springer Nature
- p. 257-270
A Graph Convolutional Network for Visual Categorization
- 8 October 2024
- Published byĀ Springer Nature
- p. 257-270
A Graph Convolutional Network for Visual Categorization