An Efficient Attention Mechanism Framework for Plant Leaf Disease Detection

A nation's development depends heavily on agriculture since food is a necessary source of nourishment for human health. With almost 65% of its people working directly in agriculture or allied fields, India is a rapidly expanding economic power. With their substantial effects on crop yield, quality, and overall profitability, plant diseases present serious problems for farmers and the agricultural sector. Plant diseases may originate from a number of causes, including environmental factors like temperature, humidity, and soil conditions, as well as pathogens including bacteria, viruses, nematodes, and fungi. Powdery mildew, rust, blight, and leaf spot are examples of common plant diseases. Using the newest technologies and international best practices, effective crop enhancement and protection measures must be put in place to address these issues. This entails, among other sustainable solutions, techniques like biotechnology, chemical treatments, seed treatment, agronomy, and fumigation. Deep neural networks, such Convolutional Neural Networks (CNNs), are currently used to effectively and precisely diagnose plant diseases in order to address the problem. CNNs have an advantage over other techniques since they can process images and extract features on their own without assistance from humans. The suggested network seeks to increase the accuracy of identifying minute-scale diseases in plant leaves by combining channel and spatial attention.