光照鲁棒的抗模糊新组合不变矩图像匹配方法  

An Image Matching Method of New Illumination-Robust and Anti-Blur Combined Moment Invariants

作　　者：刘欢[1,2] 郝矿荣[1,3] 丁永生[1,3] 毛敏[1] 

机构地区：[1]东华大学信息科学与技术学院,上海201620 [2]井冈山大学电子与信息工程学院,江西吉安343009 [3]数字化纺织服装技术教育部工程研究中心,上海201620

出　　处：《传感技术学报》2013年第9期1258-1264,共7页Chinese Journal of Sensors and Actuators

基　　金：国家自然科学基金重点项目(61134009);长江学者和创新团队发展计划项目(IRT1220);上海市优秀学术带头人计划项目(11XD1400100);上海领军人才专项资金项目(11JC1400200);上海市科学技术委员会重点基础研究项目(13JC1407500)

摘　　要：基于特征的传统图像匹配方法对环境光照变化和模糊噪声适应性较差,针对该问题提出了一种新组合不变矩。在Hu不变矩的基础上,重新组合构造出3个新组合不变矩并将其应用到图像匹配中。实验结果表明,这种新组合不变矩具有很好的光照鲁棒性和模糊不变性,能够有效地解决受光照变化和模糊噪声共同影响造成的匹配率下降问题;在保证高精度的前提下,匹配效率大大提高。该不变矩作为图像特征的描述形式进行图像匹配是有效可行的,具有良好的参考价值。Novel combined moment invariants are put forward to deal with the problem of the poor adaptability to the illumination variation and the blur noise in the cases of the conventional image feature matching. The new combined moment invariants are re-built on the basis of Hu moments and are applied in image matching. The experimental results have demonstrated that,in virtue of its illumination robustness and blur invariance, the proposed approach can effectively overcome the decline of the correct matching rate influenced by both the illumination change and the fuzzy noise. Ensuring the high precision,the matching efficiency is greatly improved. Utilizing the new combined moment in-variants as image descriptors to complete image matching is feasible and validate. It offers a nice reference value.

关 键 词：计算机视觉 图像匹配 组合不变矩 光照鲁棒性 模糊不变性 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]