Cassava Leaf Disease Classification using DL

Aim of the project:

The project focuses on classification of different diseases of cassava leaves using various Deep Learning Algorithms.

Libraries and Frameworks used:

1. Pandas
2. Numpy
3. Matplotlib
4. Seaborn
5. Tensorflow
6. Keras
7. sklearn
8. glob
9. OpenCV

Deep Learning Algorithms used:

1. MobileNet
2. ResNet
3. DenseNet
4. InceptionNet
5. EfficientNet

Accuracy and training time comparison of all the Deep Learning Algorithms

	Accuracy
MobileNet	59%
ResNet	60%
DenseNet	68%
InceptionNet	61%
EfficientNet	54%

Representation of diseases of cassava leaves

Counts of disease cases

Pie chart for the count of cases

Orignal image vs grayscale image

Original vs Resized image(224*224 pixels)

Accuracy and plots of all models

InceptionNetV2

DenseNet

ResNet50

EfficientNet

MobileNet

Conclusion

DenseNet model performs better comparative to other models used on the above dataset.