Novel Anisotropic Diffusion Algorithm Based on PID Control Law Together with Stopping Mechanism

Abstract

Anisotropic diffusion is one of the widely used techniques in the field of image process. Tremendous applications show promising results by using anisotropic diffusion algorithm or its revisions. In this paper, we propose a new anisotropic diffusion method which is based on PID (Proportional-Integral-Derivative) control law together with a stopping mechanism. The PID control law is a simple and intuitive law in cybernetics and it has been proven to be effective to improve the system's performance. In this new approach, according to the different processing purpose for edge and noise, different PID control parameters are chosen. Therefore, anisotropic diffusion approach is carried out more efficiently. Moreover, the proposed stopping mechanism is based on the linear correlation theory, which indicates the similarity of two images. Incorporating with this mechanism, anisotropic diffusion is able to stop automatically when the processed image reaches the certain predefined performance target. The experiments show the correctness and effectiveness of the proposed method.