CROP DISEASE DETECTION USING IMAGE CLASSIFICATION

However, the agricultural industries have to deal with many issues caused by plant disease, poor productivity, lack of expert knowledge, loss of soil nutrients and many others. There are numerous issues in identifying plant disease as early as possible, due to the fact that the process of cultivation is on a large scale, where manual monitoring is impossible, and requires much time and expertise, which is unattainable in rural areas. Early identification of leaf disease can save a lot of money otherwise incurred, and can even increase crop yields. Hence, agriculture in India desires to increase yield without causing any damage to the environment. Therefore, the purpose of this project is to create an automatic leaf identification system based on image processing algorithms, to help farmers identify disease in plants to enhance their productivity and yield, thereby saving a lot of money and time. For this project, various technologies like image processing, machine learning, and deep learning are used for the identification and analysis of images. Python programming language has been used along with various libraries such as OpenCV and NumPy for image processing operations. Machine learning techniques such as Convolutional Neural Networks (CNN) are used for classification. TensorFlow/Keras is used for model training and evaluation. It is well understood that agriculture continues to play a key role in the maintenance of economic stability and food security in many developing nations, where a major part of their population relies on agriculture as a source of income. With the rapid increase in population, climatic fluctuations, and various biotic stressors, the need for sustainable agriculture practices has become more pressing than ever before. Among all the other issues, diseases affecting crops rank high on the list of important considerations. Crop disease detection is crucial not only for food security but also for reducing unnecessary pesticide use.