Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{198183,
        author = {Brahmbhatt Jaimish and Ruchika Dungarani},
        title = {DeepDiagnosis: The Transition from Manual Features to Hierarchical Learning in Plant Disease Prediction},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {11},
        pages = {8195-8201},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=198183},
        abstract = {Plant diseases present a major threat to worldwide agriculture, causing significant losses in crop yield and quality, alongside negative economic consequences. Accurate and prompt disease identification is crucial for successful crop management. However, traditional diagnosis depends on manual expert assessment, which is often subjective, time- consuming, and difficult to implement across large areas. The development of Machine Learning (ML) and Deep Learning (DL) technologies has paved the way for automated plant disease prediction systems, offering rapid, dependable, and scalable analysis. Early approaches utilized classical ML methods like Support Vector Machines (SVM), Random Forests (RF), and k-Nearest Neighbors (k-NN). These methods typically depended on manually engineered features, such as descriptors of texture, color histograms, and structural characteristics. More recently, the rise of Deep Learning (DL) has established Convolutional Neural Networks (CNNs) as the leading technique. CNNs excel because they can automatically learn complex, hierarchical feature representations directly from raw image data. State- of-the-art results on common datasets, such as Plant Village, have been achieved using prominent CNN architectures— including VGG, ResNet, Inception, and EfficientNet—often leveraging transfer learning. Current research is focused on enhancing the stability of these systems under diverse real- world conditions by incorporating advanced techniques like attention mechanisms, image segmentation models, and ensemble learning.},
        keywords = {Plant Disease Prediction, Deep Learning, Convolutional Neural Networks (CNN), IoT Based Crop Monitoring, Hyperspectral Imaging, Environmental Feature Integration},
        month = {April},
        }