An utilization of edge detection in a modified Bicubic interpolation used for frame enhancement in a camera-based traffic monitoring

Abstract

Nowadays, the developed countries have widely utilized image processing in the Intelligent Transportation System (ITS), medical, biology and other fields. The image quality enhancement is an important task, especially enhancing the resolution of the image from new interpolated pixels. The conventional Bicubic interpolation (BI) algorithm used a square matrix obtained by the original neighbor pixels, which surrounds the interpolated pixels, however the image resolution is lost in the depths and and the boundary pixel value are blurred by a block artifact. This proposed algorithm aims to reduce blurred effect from block artifact by enhancing the interpolated pixel values along the edge pixels of the object of interest in the image. In other words, this paper modifies the conventional BI by adoption of Canny based edge mask for adjustment of the pixels along the detected edges in the interpolation process. The result of this proposed study can be applied to the field of camera-based traffic monitoring system.