介绍了四种常用的缘提取方法各自的特点，将图像边缘提取与联合变换相关相结合，减少了图像中的低频成份，使联合变换相关器更多地比较了图像的细节变化，增加了相关图像的信息量，使相关亮点的能量更加集中，有效地提高联合变换相关器的识别能力和可靠性。最后，基于三个相关参数，对于四种边缘提取技术在不同图像相关识别中的应用实验进行了定量分析和计算，表明了边缘提取与联合变换相关识别结合的有效性和可行性。
The characteristics of four image edge extraction methods are introduced. The combination of the image edge extraction and joint transform correlation can reduce the low frequency components of images, and the more details of images are compared by the joint transform correlation system. The method can add the image information contents and focus on more energy of correlation peaks, so the Recognition ability and reliability of the joint transform correlation system are increased effectively. Finally, the analysis and application of four image edge extraction methods in the joint transform correlation recognition of different images are made and computed based on three correlation parameters, and the results show the combination of the image edge extraction and joint transform correlation is effective and feasible.
Tao, Y., Miao, H. and Shuang, D.U. (2013) Low Contrast Target Recognition in Cluttered Background with Optical Correlation. Semiconductor Optoelectronics, 34, 890-894.
[6]
Bal, A. and Alam, M.S. (2004) Dynamic Target Tracking with Fringe-Adjusted Joint Transform Correlation and Template Matching. Applied Optics, 43, 4874-4881. https://doi.org/10.1364/AO.43.004874
