Early detection of plant diseases is critical for improving crop yield, reducing economic losses, and ensuring sustainable agriculture. Recent advancements in artificial intelligence (AI), particularly machine learning (ML) and deep learning (DL), have significantly enhanced the accuracy and efficiency of plant disease identification systems. Studies such as Sujatha et al. [1] demonstrate the effectiveness of DL models like VGG-16 and Efficient DenseNet, achieving accuracies up to 97.2%, while hybrid approaches combining Inception v3 with SVM further improve classification performance. Vision Transformer-based frameworks such as PLA-ViT proposed by Murugavalli et al. [2] address limitations of traditional convolutional neural networks (CNNs) by capturing global and local dependencies, enabling superior disease localization and classification. Additionally, lightweight architectures like MobileNetV2 [3] and attention-based CNN models [5] facilitate deployment on resource-constrained devices, making AI-driven solutions accessible to farmers in developing regions. Furthermore, emerging technologies such as hyperspectral imaging [6], UAV-based remote sensing [13], and ensemble learning methods [9], [14] contribute to earlier and more precise disease detection under real-world conditions. While datasets like PlantVillage have enabled high model accuracy, studies highlight challenges in generalization to field environments [4]. Techniques including transfer learning [12], federated learning [17], and data augmentation using GANs [25] have been proposed to address issues of data scarcity, privacy, and variability. Overall, the integration of AI with IoT and edge computing technologies provides a scalable and efficient framework for early plant disease detection, supporting precision agriculture and enabling timely intervention strategies.

The advancement of Artificial Intelligence (AI) has significantly transformed the field of plant disease detection, enabling accurate, efficient, and automated identification of diseases from plant leaf images. This study reviewed various Machine Learning (ML) and Deep Learning (DL) approaches, highlighting the superior performance of deep learning models such as CNNs, EfficientNet, and Vision Transformers in disease classification tasks [1], [12], [2]. These models have demonstrated high accuracy and robustness, especially when combined with techniques such as transfer learning, attention mechanisms, and ensemble learning. The integration of object detection and segmentation methods has further enhanced the ability to localize and quantify disease-affected regions, making these systems more practical for real-world applications [8], [23]. Despite these advancements, several challenges remain in deploying AI-based plant disease detection systems in real- world agricultural environments. Issues such as poor generalization of models trained on controlled datasets, limited availability of diverse datasets, and high computational requirements continue to hinder practical implementation [4], [25]. Additionally, early detection of diseases at pre-symptomatic stages remains a critical challenge, although emerging technologies such as hyperspectral imaging and UAV- based monitoring show promising results [11], [13]. Addressing these challenges requires the development of robust, lightweight, and scalable models that can operate efficiently under varying environmental conditions. The future of early plant disease detection lies in the integration of AI with emerging technologies such as IoT, edge computing, and federated learning, which can enable real-time, secure, and decentralized disease monitoring systems [17], [19]. Furthermore, advancements in data- efficient learning techniques and the availability of large-scale real-world datasets will play a crucial role in improving model performance and generalization [15]. By addressing current limitations and leveraging technological innovations, AI-based plant disease detection systems have the potential to revolutionize precision agriculture, enhance crop productivity, and contribute to global food security.

[1]Sujatha et al., “Advancing plant leaf disease detection integrating machine learning and deep learning,” Scientific Reports, vol. 15, 2025. [2]Murugavalli et al., “Plant leaf disease detection using vision transformers for precision agriculture,” Scientific Reports, vol. 15, 2025. [3]Mhembere et al., “Low-cost smartphone-based plant disease diagnosis for Zimbabwean farmers,” International Journal of Computer Applications, vol. 187, no. 32, 2025. [4]Richter et al., “Assessing the performance of domain-specific models for plant disease detection: a comprehensive benchmark of transfer learning on open datasets,” Scientific Reports, vol. 15, 2025. [5]Karthikeyan et al., “Enhanced plant disease classification with attention- based squeeze-and-excitation CNN (CNN-SEEIB),” Frontiers in Artificial Intelligence, vol. 8, 2025. [6]Seralathan et al., “MLVI-CNN: a hyperspectral stress detection framework with machine-learning-based vegetation indices,” Frontiers in Plant Science, vol. 16, 2025. [7]Jafar et al., “Revolutionizing agriculture with artificial intelligence: plant disease detection and management,” Frontiers in Plant Science, vol. 15, 2024. [8]Aldakheel et al., “Detection and identification of plant leaf diseases using YOLOv4,” 2024. [9]Shafik et al., “Using transfer-learning- based plant disease classification with ensemble early fusion and lead voting methods,” BMC Plant Biology, vol. 24, 2024. [10]Babatunde et al., “A novel smartphone application for early detection of habanero plant diseases using a modified VGG16 deep transfer learning model,” Scientific Reports, vol. 14, p. 1423, 2024. [11]Zhang et al., “Hyperspectral imaging analysis for early detection of tomato bacterial leaf spot disease,” Scientific Reports, vol. 14, p. 27666, 2024. [12]Johri et al., “Advanced deep transfer learning techniques for efficient cotton plant disease detection,” 2024. [13]Zhu et al., “Intelligent agriculture: deep learning in UAV-based remote sensing for crop disease monitoring,” 2024. [14]He et al., “A novel ensemble learning method for crop leaf disease recognition (ELCDR),” 2024. [15]Yang et al., “From laboratory to field: cross-domain few-shot learning for plant disease detection,” 2024. [16]Gong et al., “An analysis of plant disease identification based on deep learning: comparative evaluation of YOLOv3 and Faster R-CNN in field environments,” 2023. [17]Kabala et al., “Image-based crop disease detection with federated learning,” 2023. [18]Kabala et al., “Secure and decentralized plant disease detection via federated learning,” 2023. [19]Hnatiuc et al., “Intelligent grapevine disease detection using IoT sensor networks,” 2023. [20]Borhani et al., “A deep-learning- based approach for automated plant disease classification using Vision Transformer,” 2022. [21]Pandian et al., “An improved deep residual convolutional neural network (ResNet197) for plant leaf disease detection,” 2022. [22]Khan et al., “End-to-end semantic leaf segmentation framework for plant disease identification,” 2022. [23]Li et al., “Detection of potato foliage diseases using instance, classification, and semantic segmentation models,” 2022. [24]Atila et al., “Plant leaf disease classification using EfficientNet deep learning model,” Bi et al., “Improving image-based plant disease classification with generative adversarial network and label smoothing under limited training data,” Frontiers in Plant Science, vol. 11, 2020

© 2025 International Journal of Engineering and Techniques (IJET).

Digital Object Identifier (DOI)

IJET assigns a unique DOI to every accepted article via Zenodo for persistent linking and citation. Your article’s DOI: https://doi.org/{{doi}}

Open DOI About Zenodo DOI

Close