This paper describes that in the field of video image surveillance,due to uneven illumination,the images often have areas with low illumination areas,over-lit areas,where a large number of details cannot be observed.Improving the contrast of these areas and improving the visibility of the images is an important issue.It presents a method for local detail enhancement and global brightness regulation.The input image was first divided into the base layer and detail layer using Guide Filter(GF).After the filtering is completed,the detail layer is first stretched with the arctan function,and then the brightness component of the base layer is used to adjust the detail layer,and then linearly combined with the base layer to enhance the detail.Finally,the image was adjusted using the maximum entropy-guided adaptive curve to adjust the brightness and improve the contrast of low and over-lit regions.Experimental results show that our proposed algorithm has significant results in enhancing uneven illumination images and has the advantage of low running time consuming over some popular algorithms.
维普
万方数据