A new ear recognition method based on differential geometry

作者：Ma Chi^*; Tian Ying

来源：Journal of Information and Computational Science, 2013, 10(18): 6049-6056.

DOI：10.12733/jics20102570

• Binary template matching
• Broad application
• Differential geometry
• Ear recognition
• Image preprocessing
• Intensity images
• Principal curvature
• Subject recognition

The human ear is unique and stability and it has broad application prospects in the field of pattern recognition. In this paper, a novel ear recognition method based on differential geometry is proposed. Firstly, we get the normalization ear image through the image preprocessing. Secondly, according to the character of ear recognition, a method based on differential geometry was used to extract ear anatomy feature points. Finally, the best match is found from a database by using binary template matching. The proposed method can be performed automatically for subject recognition by using 2D ear intensity image. And it can efficiently overcome the influence of illumination and rotation change for ear recognition. Experimental results show that proposed method is simple, robust and better in accuracy and efficiency than previous methods.

• 出版日期2013
• 单位北京科技大学