Multiclass-image-classification-using-deep-learning-in-python

Date:Apr23 -2025

This deep learning project to classify 6 types of wild animals using CNN.

• Classes: 1.Cheetah 2.Fox 3.Hyena 4.Lion 5.Tiger 6.Wolf

• Data colletion: Wild animals image dataset collected from kaggle.Image_size(224,224) and each classes had 100 images.

• Features: Custom convolutional neural network with 5 layers(1-input,3-hidden and 1-output layers),also using data augmentation to improve generalization.

• Trainning: model compile with CNN and fit to train a model.

• Tools: language Python , Tensorflow and keras, Jupyter notebook.

• Save: Model saved in h5(Hierarchical file)

• Result: Model got at 85% accuracy of training and validation 65% (Improvement need)

• Improvements: Model train well but validation to be low perform ,need improvement of images size,datasets and will use earlystopping for prevent overfitting.

• Deployment: Streamlit UI for user upload the image to classification.