- https://www.mathworks.com/help/releases/R2022a/deeplearning/gs/create-simple-deep-learning-classification-network.html
- https://www.mathworks.com/help/releases/R2022a/deeplearning/ug/create-simple-deep-learning-network-for-classification.html
How to implement MT-CNN for any disease detection and Classification from the images of disease affected areas?
5 views (last 30 days)
Show older comments
I am working on disease detection by image processing and currently working with MT-CNN. I have tried several methods to detect disease using MT_CNN but all went into vain. Now I really don't know how to implement it for plant disease detection from images of specific crops. I would be very much grateful if anyone could help.
0 Comments
Answers (1)
Drishti
on 4 Oct 2024
Hi Muksit,
The Multi-Task Cascaded Convolutional Networks (MT-CNN) are typically used for facial detection and recognition. Similarly, to detect and classify plant diseases from images, you can develop a custom network that utilizes the same functionality.
The MT-CNN model is composed of three stages: 'P-Net', 'R-Net', and 'O-Net'. You can design custom layers to define each of these stages for your specific application.
Refer to the provided implementation for better understanding:
% Define P-Net (12x12 input size)
layersPNet = [
imageInputLayer([12 12 3], 'Normalization', 'none', 'Name', 'input')
convolution2dLayer(3, 10, 'Padding', [1 1], 'Stride', [1 1], 'Name', 'conv1')
reluLayer('Name', 'relu1')
maxPooling2dLayer(2, 'Stride', 2, 'Name', 'pool1')
fullyConnectedLayer(2, 'Name', 'fc2')
softmaxLayer('Name', 'softmax')
classificationLayer('Name', 'output')];
All candidates of ‘P-Net’ are fed to ‘R-Net’ and following pattern ‘O-Net’ receives input from ‘R-Net’. Based on the implementation described above, you can set up the 'R-Net' and 'O-Net' networks, and then specify the training options.
For more information, you can refer to the MATLAB Documentation on creating an image classification network.
I hope this helps.
0 Comments
See Also
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!