Histogram matching

An example of histogram matching

In image processing, histogram matching or histogram specification is the transformation of an image so that its histogram matches a specified histogram. The well-known histogram equalization method is a special case in which the specified histogram is uniformly distributed.

It is possible to use histogram matching to balance detector responses as a relative detector calibration technique. It can be used to normalize two images, when the images were acquired at the same local illumination (such as shadows) over the same location, but by different sensors, atmospheric conditions or global illumination.

Algorithm

Given two images, the reference and the adjusted images, we compute their histograms. Following, we calculate the cumulative distribution functions of the two images' histograms – $F_{1}()\,$ for the reference image and $F_{2}()\,$ for the target image. Then for each gray level $G_{1}\in [0,255]$, we find the gray level $G_{2}\,$ for which $F_{1}(G_{1})=F_{2}(G_{2})\,$, and this is the result of histogram matching function: $M(G_{1})=G_{2}\,$. Finally, we apply the function $M()$ on each pixel of the reference image.

Multiple histogram matching

The histogram matching algorithm can be extended to find a monotonic mapping between two sets of histograms. Given two sets of histograms $P=\{p_{i}\}_{i=1}^{k}$ and $Q=\{q_{i}\}_{i=1}^{k}$, the optimal monotonic color mapping $M$ is calculated to minimize the distance between the two sets simultaneously, namely $\operatorname {arg} \min _{M}\ \sum _{k}d(M(p_{k}),q_{k})$ where $d(\cdot ,\cdot )$ is a distance metric between two histograms. The optimal solution is calculated using dynamic programming.

See also

• Histogram equalization
• Image histogram
• Color mapping