Abstract
Normalized Cross Correlation Operating (NCCO) between two binary images, template image and target image, is independent of their gray levels. Further more, providing the template is a binary image, the calculation results of NCCO should remain unrelated to the gray level of the template image. This paper proved the inference above and presented some methods to improve the pattern matching algorithms. These methods can be applied to the microscopic vision field, where the majority of images are low in their gray levels, approximating to three-value images, or even binary. Experimental results show that the characteristics of gray level unrelated to the NCCO can give guidance to designing template images, reducing time consumption of calculations, as well as improving positioning algorithm.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
