In this paper, we propose the method extracts regions with excessive disparities in 3D stereoscopic
images using block based disparity calculation and automatically binarization. The proposed method
calculates the disparity-map using not pixel based method but block based method. The disparity-map’s
resolution is decreased in proportion to block size. In the proposed method, first, the disparity-map is
calculated based on BMA (block matching algorithm). The normalized cross-correlation is used as the cost
function. The excessive disparity regions are extracted using the automatically decided threshold with Otsu
algorithm in the disparity-map. The sum of the disparities in the object regions which is normalized with
image size is used as parameter in fatigue evaluation model. In the simulation results, we can confirm that
the proposed method's results are very similar to the pixel-based method.