Multi-scale nonlinear contrast enhancement of radiographics based on human contrast sensitivity

Abstract

Medical imaging techniques like computed/digital radiography (CR/DR) have introduced a formidably powerful tool in medicine. Image enhancement takes an important roll in the CR/DR computerized analysis process. Much effort has been put into the area of image enhancement. However, conventional multi-scale methods have the drawback of the introduction of severe visible artifacts while large structures are enhanced strongly. This paper presents a nonlinear multi-scale medical image contrast enhancement method for the improvement of medical image quality. More specifically, a novel nonlinear enhancement function is proposed incorporated with human visual local perceptual contrast. The proposed work provides the advantages of enhancing or preserving image contrast while suppressing visible artifacts. To quantitatively compare the performance of the proposed method, the average local variances are used as comparison criteria. Results demonstrate the superiority of the proposed method. Our results show that the proposed method has the potential to become useful for improvement of image quality of medical images.