Faculty Publications

Permanent URI for this communityhttps://idr.nitk.ac.in/handle/123456789/18736

Publications by NITK Faculty

Browse

Subject: search.filters.subject.Deep convolutional neural networks

Search Results

Now showing 1 - 2 of 2
  • No Thumbnail Available
    Item
    Automated Detection of Maize Leaf Diseases in Agricultural Cyber-Physical Systems
    (Institute of Electrical and Electronics Engineers Inc., 2022) Verma, A.; Bhowmik, B.R.
    Agricultural cyber-physical systems (ACPS) are an ever-increasing sector that affects the quality and quantity of agricultural products as the population increases rapidly. Maize, also known as 'corn,' is one of the world's old food crops, consumed every part of Bharat with 1.4 billion masses across the globe. But a disease, whether on seeds, leaves, or other parts of a crop plant, poses a significant risk to food security. For example, a Maize leaf experiences three diseases-blight, common rust, and gray leaf spot. Early detection and correct identification of these diseases can help restrict the spread of infection and ensure crop quality for long-Term health. This paper proposes a deep convolutional neural network (DCNN) framework for Maize leaves named "MDCNN"that detects these diseases. The proposed MDCNN model undergoes training and is tuned to detect four prevalent classes of the conditions. The proposed model exercises a voluminous dataset of the diseases. Experimental results demonstrate that the proposed model achieves a training and test accuracy up to 95.51% and 99.54%, respectively. Furthermore, it outperforms many existing methods and delivers a superior disease control solution for Maize leaf diseases. © 2022 IEEE.
  • No Thumbnail Available
    Item
    Multi-class Classification of Wireless Capsule Endoscopy Images By Using the Fusion of Pre-trained Networks and Fusion Residual Block
    (Springer Science and Business Media Deutschland GmbH, 2025) Lavanya, K.; Aparna., P.
    In order to identify diverse endoscopic images of gastrointestinal (GI) problems, this study presented a multi-fused residual convolutional neural network (MFuRe-CNN) with auxiliary fusion layers (AuxFL) and a fusion residual block (FuRB) with alpha dropouts (ADO). The implemented MFuRe-CNN dealt with five instances sourced from reputed databases such as the KVASIR database, the ETIS-Larib Polyp Database, and the Red Lesion Endoscopic Database, which included Esophagitis, Ulcerative colitis, Polyps, Healthy colon and Bleeding images. The proposed model was created by fusing three cutting-edge models fused into a single-feature extraction pipeline with its layers that are truncated and partially frozen. This model which despite using a small portion of the computing power of most current cutting-edge models, helped spread robust features and enhanced diagnostic performance. In addition, compared to those without the aforementioned components, the MFuRe-CNN in conjunction with AuxFLs, DOs and FuRB has significantly reduced overfitting and performance saturation. With merely 4.8 million parameters, this model achieved test accuracy of 93.79 percent, outperforming the majority of DCNNs that are trained in the traditional manner. Thus, the suggested MFuRe-CNN could improve GI tract diagnosis more cheaply than ensembles and outperform other traditional pre-trained as well as fine-tuned Deep convolutional neural networks. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.