Color pattern recognition with the multi-channel non-zero-order joint transform correlator based on the HSV color space

Color is one of the most powerful and important types of visual information. A new approach for color pattern recognition based on the non-zero-order joint transform correlator with the HSV color space is introduced. We report on progress to reduce cross correlation energy more while keeping the correlation peaks as specified as usual. Numerical results show the effectiveness of the proposed model to further enhance the correlation output. We evaluate the performance of the system by applying multiple levels of additive Gaussian noise, as well as in a realistic color background. (C) 2004 Elsevier B.V. All rights reserved.