Leaf Disease Detection Using CNN-Deep Learning
Ms. Kumud Sachdeva
Dept. of CSE Chandigarh University kumud.cse@cumail.com
Tamanna CSE 513/A
Chandigarh University 22bcs17239@cuchd.in
Kamaljit Kaur CSE 513/B
Chandigarh University 22bcs17256@cuchd.in
Sangam Majoka CSE 513/A
Chandigarh University
22bcs17252@cuchd.in
Abstract
To be able to Agriculture, the economy, and the effort to combat global warming all depend on plants. Consequently, it is crucial to take care of plants. Plants can contract a variety of bacterial, fungal, and viral diseases, just like people can. Early Disease The use of pets and detection are essential for improving agricultural output and quality. Due to a decline in the quality of the agricultural produce, diseased plants can cause large financial losses for individual farmers. In a nation like India, where a substantial section of the population engages in agriculture, it is imperative to recognise the disease in its earliest stages. With quicker and more precise plant disease prediction, losses might be reduced. Early disease detection and treatment are essential to preventing the destruction of the entire plant. Significant developments and improvements in deep learning have opened up the possibility of improving the effectiveness and precision of object identification and recognition systems. The main objectives of this study are the detection of plant diseases and the reduction of economic losses.Deep learning has been proposed as a method for image recognition. As the three main architectures of the neural network, we have examined the Region-based Fully CNN (R-CNN), Single shot Multibook Detector (SSD), and Faster Region-based Convolution Neural Network (Faster R- CNN). The research's suggested approach is capable of handling complex scenarios and is efficient at identifying various sickness types. The validation results outline the path ahead for an AI-based Deep Learning solution to this Complex Problem and show the viability of the Convolution Neural Network with an accuracy of 94.6%.
Keywords
Leaf disease detection, Convolutional Neural Network (CNN), Deep learning, Image processing.