Method of extracting saliency target through gray statistical data depth information

The invention discloses a method of extracting a saliency target through gray statistical data depth information. Through improving an FT algorithm, a saliency map basically meeting CBP (Center Bias Prior) features of human vision attention. Through mining the depth information of gray probability distribution characteristics of the saliency map, a saliency distribution law in the original image is found, and a saliency target area based on the saliency distribution is extracted by using technologies such as curve fitting, gray statistics, curve monotonic analysis and super pixel segmentation. Finally, in combination with a graph manifold ranking technology, a saliency map close to a test set standard is realized. compared with the majority saliency target detection method, the detection algorithm provided by the invention has a quicker execution speed and lower complexity, and a higher detection precision can be ensured. 种通过灰度统计数据深度信息提取显著目标的方法，通过改进FT算法获得基本符合人类视觉注意的CBP特征的显著图。通过挖掘该显著图的灰度概率分布特性的深度信息找到原图像中显著性分布的