| Abstract: | Color difference may exist between two views of stereoscopic images acquired with stereomicroscope,which makes trouble for the following image processing or observation.A color correction method based on the nearest cumulative histogram matching is proposed.Histogram-based contrast(HC)method is proposed to define saliency value of each pixel,then auto Grabcut segmentation method is used to segment the salient region so as to obtain a region of interest(ROI).After that,normalized histograms and cumulative histograms for ROI and region of background(ROB) are calculated.The mapping functions of the corresponding regions are derived from reference image to distorted image through the nearest cumulative histogram matching method,so that color correction can be finally achieved.Experimental results show that benefitting from the separate treatment to ROI and ROB,the proposed color correction method could avoid error propagation between the two different regions,which achieves good color correction result in comparison with other correction methods. |
