Leaf Disease Detection for Plant Using Image Processing

Abstract

 Agriculture gave birth to civilization. Being an agricultural country, India’s economy is largely based upon crop production. Agriculture is the backbone of every economy. In a country like India which has ever-increasing demands for food due to the rising population, advances in the agriculture sector are required to meet the need.  The agriculture sector needs an enormous up-gradation to survive the changing conditions of the Indian economy. For optimum yield, the crop should be healthy; therefore some highly technical method is required for periodic monitoring of the crop. Crop disease is one of the main factors which indirectly influence the significant reduction of both the quality and quantity of agricultural products. Several sorts of pesticides are available to control diseases and increase production. But finding the foremost current disease, appropriate and effective pesticide to control the infected disease is difficult and expert advice is required which is time-consuming and expensive. The presence of disease on the plant is especially reflected by symptoms on leaves. So, there is a requirement of an automatic, accurate, and fewer expensive Machine Vision System for the detection of diseases from the image and to suggest a correct pesticide as a solution. Detection of disease through some automatic technique is really useful because an oversized work of watching in huge farms of crops is reduced by it and at the terribly early stage itself it detects the symptoms of diseases means that after they appear on plant leaves. This system presents a neural network algorithm for image segmentation technique used for automatic detection still because the classification of plants and survey on completely different diseases classification techniques that may be used for plant leaf disease detection. Image segmentation, which is a really important aspect for malady detection in plant disease, is completed by a victimization genetic algorithm. Deep learning is a hot research topic in pattern recognition and machine learning at present; it can successfully solve these problems in vegetable pathology. In this study, we propose a new sorghum leaf disease detection method based on convolutional neural networks (CNNs) techniques. To improve the detection accuracy of sorghum leaf diseases and reduce the number of network parameters, the Alex Net model based on deep learning is proposed for leaf disease detection.