Clothes Pattern Recognition

Project Overview

This project focuses on cloth pattern recognition using deep learning approaches. The goal is to build a model that can classify different clothing patterns, such as stripes, polka dots, plaid, floral, and more. By leveraging deep learning techniques, particularly Convolutional Neural Networks (CNNs), we aim to achieve high accuracy in detecting and categorizing various cloth patterns. This application can be integrated into fashion recommendation systems, inventory management, or even fabric production processes.

Setup

Python 3.10.10 or more

Install Packages

pip install -r requirements.txt

CNN Model Train cnn model

python cnn.py

Kaggle Notebooks There are two notebooks available to run the code and check output.

1. fabric-cnn.ipynb
2. fabric-effiecientnet.ipynb

optional model training

ResNet Model Train cnn model

python resnet.py

GCNN Model Check final outcomes

python gcnn.py

Model Architecture

The deep learning model is based on Convolutional Neural Networks (CNNs), which are highly effective for image classification tasks. The architecture includes:

• Convolutional layers for feature extraction
• Max-pooling layers for dimensionality reduction
• Fully connected layers for classification
• Softmax activation function for multi-class classification

Future Improvements

• Enhance dataset: Use a more diverse dataset with more patterns and variations in lighting, orientation, and fabric types.
• Fine-tune the model: Apply transfer learning with a pre-trained model such as ResNet, VGG, or Inception for better performance.
• Real-time detection: Implement real-time pattern recognition using a camera feed.
• Mobile integration: Convert the model to run on mobile devices using TensorFlow Lite or ONNX. Also compatible with real time devices like android and iOS.