Contrast enhancement by using global and local histogram information jointly

摘要

Global histogram equalization (GHE) is a common technique used to enhance image contrast by utilizing the histogram information of the input image to create its transformation function. However, it usually fails to adapt with the local image brightness features. By contrast, Local histogram equalization (LHE) is suitable for handling the local feature; whereas it suffers from halo or blocking artifacts and requires high computation power. This paper presents an effective hybrid method for image contrast enhancement that jointly combines local information and global information. Therefore, it provides a regional enhanced effect without generating visual artifacts or halo in local areas. Experimental results show that our proposed method produces higher visual quality on the enhanced images than the GHE method does.