Recognition of Diseases of Maize crop using Deep Learning Models
KRISHI: Publication and Data Inventory Repository
View Archive InfoField | Value | |
Title |
Recognition of Diseases of Maize crop using Deep Learning Models
Not Available |
|
Creator |
Md. Ashraful Haque
Sudeep Marwaha Chandan Kumar Deb Sapna Nigam Alka Arora |
|
Subject |
Deep Learning
Convolutional Neural Networks Disease recognition Maize crop |
|
Description |
Not Available
Disease attack on crops is one of the most serious threats to the global food supply chain. A proper, comprehensive and systematic solution is required for the early recognition of diseases and to reduce the overall crop loss. In this regard, deep learning techniques (especially convolutional neural networks (CNNs/ConvNets)) are being successfully applied for automatically recognizing the diseases of crops using digital images. This study proposes a novel 15-layer deep convolutional neural network (CNN) model for recognizing the diseases of maize crop. Around 3,852 images of maize crop were collected from the PlantVillage data-repository. This dataset contains leaf images of three diseases viz. Grey Leaf Spot (GLS), Common Rust (CR) and Northern Corn Leaf Blight (NCLB) as well as the healthy ones. The proposed model showed significant results for recognizing the unseen diseased images of the maize crop. We also employed a few popular pre-trained networks in the transfer learning approach for training on the maize dataset. We presented the comparative performance analysis between the proposed model and the pre-trained models in the result section of the manuscript. The experimental findings reported that our proposed model showed 3.2% higher prediction performance with 3x lesser trainable parameters than the best-performing pre-trained network (i.e. DenseNet121). The overall performance analysis reported that the proposed CNN model is very effective in identifying the images of maize diseases and also performs quite better than the popular pre-trained models. Not Available |
|
Date |
2022-12-14T04:53:20Z
2022-12-14T04:53:20Z 2022-11-23 |
|
Type |
Research Paper
|
|
Identifier |
Haque, M. A., Marwaha, S., Deb, C. K., Nigam, S., & Arora, A. (2022). Recognition of diseases of maize crop using deep learning models. Neural Computing and Applications, 1-15.
http://krishi.icar.gov.in/jspui/handle/123456789/75203 |
|
Language |
English
|
|
Relation |
Not Available;
|
|
Publisher |
Springer
|
|