Leaf Diseases and Weed Detection using Image Processing: A Comprehensive Survey
Abstract
The comprehensive survey investigates all the image processing systems and deep learning architectures from the conventional machine vision system to the latest generation of CNN architectures like ResNet, EfficientNet, VGG, MobileNet, DenseNet and YOLO for the detection of various plant leaf diseases and weed identification. This survey focuses on the following: (i) image acquisition platform (ground based, UAV, mobile); (ii) image processing and segmentation pipeline; (iii) image feature extraction (hand-crafted and learned); (iv) classification and object detection framework; (v) benchmark datasets (PlantVillage, DeepWeeds, and CropDeep); and (vi) performance metrics. an overview of 24 major studies from 2016 to 2026, explicitly reporting metric type and validation protocol is provided in a comparative taxonomy table. Also discussed are issues such as intra-class variability, class imbalance, lack of annotated field data, and deployment on resource constrained edge devices. Other areas with potential for future work are transformer based models, multi-modal sensor fusion, federated learning, and explainable AI (XAI). This survey is intended to give researchers and agricultural engineers information towards efficient, deployable, precision agriculture solutions.