Cotton Disease Detection using Deep Learning
Kadem Shravan Kumar1, Gollapudi Ramesh Chandra2, Deepak Sukheja3
1Kadem Shravan Kumar*,Department of Computer Science and Engineering, VNR Vignana Jyothi Institute of Engineering and Technology Hyderabad, India.
2Dr. Gollapudi Ramesh Chandra, Department of Computer Science and Engineering, VNR Vignana Jyothi Institute of Engineering and Technology Hyderabad, India.
3Dr. Deepak Sukheja, Department of Computer Science and Engineering, VNR Vignana Jyothi Institute of Engineering and Technology Hyderabad, India.
Manuscript received on January 13, 2020. | Revised Manuscript received on January 20, 2020. | Manuscript published on February 10, 2020. | PP: 152-156 | Volume-9 Issue-4, February 2020. | Retrieval Number: D1391029420/2020©BEIESP | DOI: 10.35940/ijitee.D1391.029420
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Cotton is one of the most important crop in India . A large number of people depends on cotton crop either by its cultivation or processing. It is used for making threads, Extract the oil from cotton seeds, Most of the diseases in cotton caused to leaves like Bacterial blight, Alternaria leaf spot, Cercospora leaf spot, red spot, they all are caused because some nutrition deficiency like magnesium deficiency, sometimes it is very difficult to farmer to check whether it is normal leaf or magnesium deficiency leaf, if farmer misclassifies magnesium deficiency leaf and non-diseased leaf it may lead to less yield and huge loss. To achieve more yield we need to automate the cotton leave disease detection. Machine learning is one of the emerging technology in recent times By using Machine learning concepts we can train the system to detect whether a given plant is diseased or not by giving the input as a cotton leaf. In this paper, We used the Sequential model in the Convolution neural network architecture which is widely used for image classification. With the Sequential model, we got an accuracy of 87% with very less dataset by using the Image Dataset Generator which increases the dataset size by augmentation
Keywords: Convolutional Neural Network, Image dataset generator, Machine learning, Sequential model.
Scope of the Article: Machine learning