A New Grey-Scale Template Image Matching Algorithm Using the Cross-Sectional Histogram Correlation Method

A new correlation technique for Grey-Scale Template Image Matching, the Cross-Sectional Histogram Correlation (CHC) Method, is proposed. In comparison to the traditional, widely-used Normalized Grey-Scale Correlation (NGC) Method, the CHC Method is computationally inexpensive and more robust against image noise. When implemented in tandem with the Image Pyramid Structure, CHC can theoretically be executed in the order of 10 ms or shorter on modest computing machines. Since the functionality of the CHC method is identical to the NGC method, the former method can be easily substituted for the latter method without major changes to current softwares running on NGC.

• Authors: Ryo Takei
• Keywords: Normalized Grey-Scale Correlation, Pattern Recognition, Sectional Histogram Correlation