Histogram Comparison Theory
I'm currently using Histogram Comparison, BhattaCharyya method to perform image processing in order to figure out the percentage of difference/ similarity out of two images. However, a weird result came out. From Figure (a) and Figure (b), the result is 0.0753672 and 0.103018 respectively. What I curious about is that Figure (a) shows more mismatch than Figure(b) but why value for Figure (a) is lower?! May I know in conceptual base, the reason and theory behind?
Figure (a)
Figure (b)