Purpose: The proposed study will develop an AI-assisted automated crop disease detection system for tomato, cotton, and wheat plants based on CNN, SVM, VGG19, and ResNet models, combined with an intelligent treatment recommendation system that will help farmers address diseases in a timely and effective manner.
Design / Methodology / Approach: One of the proposed frameworks uses data pre-processing, augmentation, and comparative training of CNN, SVM, VGG19, and ResNet models, with hyperparameters, dropout regularisation, and a smart prescription module that provides organic, chemical, and agronomic treatment recommendations based on Kaggle public databases.
Research Limitation: The system is restricted to three crop types on class-imbalanced data and has yet to be tested in practical, real-time mobile or field settings, with ResNet showing unstable classification and requiring further improvement.
Findings: CNN had the highest accuracy of 92.38%, followed by VGG19 at 91.40%, SVM at 84.19%, and ResNet at 80.00%. Data augmentation increased overall generalisation by about 5, and hyperparameter optimisation is significant to the overall performance of the model.
Practical Implication: The framework offers a scalable, end-to-end disease detection and treatment advisor framework which can be deployed through mobile applications and edge AI devices to facilitate real-time, offline crop diagnostics for resource-constrained farmers in agricultural settings.
Social Implication: By enabling text-to-speech accessibility and early disease detection with AI, the system will assist low-literacy smallholder farmers, enhance food security, and encourage sustainable farming practices through the targeted and limited use of chemicals.
Originality/Value: The study introduces a unique multi-model benchmarking infrastructure, which includes both automated disease classification and a prescriptive treatment engine and an inclusive text-to-speech interface, showing that shallow CNN networks can be more accurate than deeper ones and creating a pipeline of precision agriculture, spending more responsibly and inclusively on a course-to-course basis.
C. Tripathi, V. Taksande, U. Patel et al.· African Journal Of Applied R...· 0 citations
Yoga represents an age-old practice that is beneficial for both psychological and physical health. The yoga promotes self-learning and improper poses can seriously harm muscles and ligaments. The accurate recognition of yoga poses from images remains challenging due to high intra-class similarity, background variations, and redundant feature generated by deep learning models. This paper proposed a model that combined the MobileNet with Hunter-Prey Optimization, and Long Short-Term Memory to detect the efficient and accurate yoga poses. In the proposed model, Mobile Net is used as a lightweight backbone to extract high-level unique and spatial features from yoga images. The extracted feature vectors are then optimized using the HPO algorithm that selects the most relevant and discriminative features and remove redundant feature. The optimized feature is fed into an LSTM model. The LSTM model operates on static image inputs as a gated feature refinement and classification rather than modelling temporal pose transitions. The performance of model is measured on a 6-class subset of the Yoga-82 dataset that allows the controlled experimental parameters. The final results shows that the proposed model obtained an accuracy score of 97.97% performed well as baseline models such as MobileNet and MobileNet + LSTM. The HPO-based feature selection reduces the feature dimensionality up to 69.53% to improved computational efficiency and reduced inference time. The result shows when combined the feature optimization with lightweight MobileNet model then enhances classification performance and maintain the efficiency of model. The proposed model is applicable for real-time yoga pose recognition with future work focused on extending the model to full-scale datasets and video-based temporal modelling.
A. Paul, L. Damahe· Journal of Intelligent Decis...· 0 citations
An integrated multi-layer hybrid framework for accurate, real-time posture assessment in healthcare and rehabilitation contexts is proposed, although all solutions trade off accuracy, computational cost, and practical generalizability.
A. Paul, L. Damahe· African Journal Of Applied R...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
