The Research of Ear Identification Based On Improved Algorithm of Moment Invariant

Abstract

In this paper, the processing of ear identification is introduced in brief, and after making an improvement of moment invariants algorithm, six feature vectors are distilled by the method of high-order moment invariants. By using two kinds of algorithm respectively, ear image is processed to identification. In ear identification, we can learn that the high-order moment invariants in the improved algorithm have the feature of invariance in translation, rotation and scale change. Finally, with the application of a BP artificial neutral network to identify 60 ear images, we get a good experiment result of 91.8% accuracy.