https://github.com/hchandeepa/image_classification_using_cnn

Image Classification using CIFAR10_dataset.
https://github.com/hchandeepa/image_classification_using_cnn

ann cnn cnn-classification image-classification jupyter-notebook python

Last synced: 5 days ago
JSON representation

Image Classification using CIFAR10_dataset.

• Host: GitHub
• URL: https://github.com/hchandeepa/image_classification_using_cnn
• Owner: HChandeepa
• Created: 2024-08-15T07:39:25.000Z (3 months ago)
• Default Branch: main
• Last Pushed: 2024-08-16T11:30:53.000Z (3 months ago)
• Last Synced: 2024-09-13T16:19:16.062Z (2 months ago)
• Topics: ann, cnn, cnn-classification, image-classification, jupyter-notebook, python
• Language: Jupyter Notebook
• Homepage:
• Size: 613 KB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# CIFAR-10 Image Classification with TensorFlow

##### This project demonstrated the process of classifying images using the CIFAR-10 dataset with TensorFlow. I explored the differences in performance between a simple Artificial Neural Network (ANN) and a Convolutional Neural Network (CNN).

## Project Overview

##### The CIFAR-10 dataset consisted of 60,000 32x32 color images across 10 classes, with 6,000 images per class. The classes represented airplanes, cars, birds, cats, deer, dogs, frogs, horses, ships, and trucks. My goal was to build a model that could accurately classify these images into their respective categories.

## Steps Involved

## Data Preprocessing:

##### I loaded the CIFAR-10 dataset and normalized the image data to improve model performance.

## ANN Model Training:

##### I built a simple Artificial Neural Network to serve as a baseline, trained the ANN model, and evaluated its performance on the test data. I observed the accuracy and identified the limitations of using an ANN for image classification.

## CNN Model Training:

##### I designed a Convolutional Neural Network tailored for image classification, trained the CNN model on the same dataset, and evaluated the CNN's performance, comparing it with the ANN.

## Performance Comparison:

##### I compared the accuracy, loss, and general performance of the ANN and CNN models and understood why CNNs are preferred for image classification tasks.

## Conclusion

##### Through this project, I gained a better understanding of why Convolutional Neural Networks outperform Artificial Neural Networks in image classification tasks. The step-by-step approach helped me appreciate the advantages of using CNNs, especially when working with visual data.