Person re-identification (Re-ID) is a critical computer vision job that entails identifying individuals across various images and camera viewpoints. It is very applicable to intelligent transportation systems, security, and surveillance. This study of ours uses cutting-edge color conversion methods and deep learning to give a thorough method for person Re-ID. Our approach improves matching accuracy and feature extraction under a range of lighting situations and camera setups by employing convolutional neural networks (CNNs) and cutting-edge color scheme conversions. We use a solid dataset to assess our model’s performance and show that it effectively obtains high re-identification accuracy. The outcomes demonstrate that our method works better than conventional approaches, especially in difficult situations with large color changes and occlusions. Along with discussing how various color spaces affect Re-ID performance, this study also identifies future research areas in this quickly developing topic. Our research advances person-Re-ID technology and provides improved functionality for practical applications. We have evaluated the results and performance of our proposed approach on publicly available datasets i.e., CUHK_01, CUHK_03, Market-1501, and VIPeR demonstrating satisfactory improvements in rank-based accuracy.