Title: Leveraging constitutive artificial neural networks for plant leaf disease detection

Authors: Kaavya Kanagaraj; Madhumitha Kulandaivel; Francis H. Shajin; Salini Prabhakaran

Addresses: Department of Computational Intelligence, SRM Institute of Science and Technology, Kattankulathur, Chennai, Tamil Nadu, India ' Department of Computer Science, Anna University, Tamilnadu, India ' Xpertmindz Innovative Solutions Private Limited, Kuzhithurai, Tamil Nadu, India ' Department of Computer Science and Engineering, Puducherry Technological University, Puducherry, India

Abstract: The appearance of new diseases in plant leaf is a significant threat to global food security and agricultural production. Therefore, a plant leaf disease detection and constitutive artificial neural network (PLDD-CANN) is proposed in this paper to provide developments in deep learning. After segmenting the image using the adaptive convex clustering (ACC) technique, the features are extracted using the fast Fourier and continuous wavelet (FFCWT) transformations. Constitutive artificial neural network (CANN) is considered to classify the input image as normal or virus, such as yellow leaf curl virus, septoria leaf spot, two-spotted spider mite, bacterial spot, target spot, leaf mould, mosaic virus, early blight, late blight. PLDD-CANN attains 26.75%, 25.83%, 27.46% better accuracy analysed with existing models, like improved CNN strategy for tomato plant leaf infection detection (CNN-PLDD), tomato leaf ailments finding for agro-base industries (FRCNN-PLDD).

Keywords: plant leaf infection; plant village dataset; image processing; adaptive convex clustering; ACC; constitutive artificial neural network; CANN.

DOI: 10.1504/IJAHUC.2025.144399

International Journal of Ad Hoc and Ubiquitous Computing, 2025 Vol.48 No.3, pp.117 - 129

Received: 23 Nov 2023
Accepted: 31 May 2024
Published online: 11 Feb 2025 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article