Paper Key : IRJ************283
Author: Resmi R Menon,Kevin Manjila,Nazim Mohammed,Yadukrishnan M S
Date Published: 06 Nov 2024
Abstract
Dress pattern recognition is a critical task in fashion technology and digital retail that involves identifying and categorizing various clothing patterns automatically. Convolutional Neural Networks (CNNs) have emerged as a powerful tool for image recognition tasks due to their ability to learn hierarchical patterns and features from visual data. This paper presents a novel approach to dress pattern recognition using a CNN algorithm. The proposed system utilizes a deep learning framework to classify different dress patterns such as stripes, polka dots, floral, and checks. The CNN model is trained on a diverse dataset of dress images collected from various online fashion platforms, ensuring a wide range of pattern types and styles.Key components of the approach include preprocessing techniques for image normalization and augmentation to enhance the model's robustness against variations in lighting, scale, and orientation. The network architecture is optimized through hyperparameter tuning and the use of advanced techniques like dropout and batch normalization to prevent overfitting and improve generalization. Experimental results demonstrate that the CNN-based model achieves high accuracy in dress pattern recognition, outperforming traditional machine learning methods. The study concludes that the integration of CNNs in fashion analytics can significantly enhance the efficiency of digit u al cataloging and personalized shopping experiences. Future work will explore the application of transfer learning and further refinement of network architectures to improve performance across even more complex pattern types.