Preprocessing of automobile engine connecting rod based on shadow removal and image enhancement

Abstract

As the main transmission parts, the quality of automobile engine connecting rod directly affects the engine performance, and ultimately affects the safety of the vehicle. Image preprocessing is the key basic technology that affects the precision of multi-parameters detection based on machine vision for the quality of automobile engine connecting rod. The image preprocessing based on multi threshold analysis and homomorphic filter is proposed to remove the shadow and enhance the contrast of the engine connecting rod image. The multi-parameters inspection vision system of automobile engine connecting rod quality is constructed to realize the simultaneous detection of geometric parameters, bending and twisting for connecting rod. The problem of shadow elimination method in single color space is analyzed for engine connecting rod image. The shadow elimination method of color image is studied based on multi threshold analysis. A Gaussian homomorphic filter is designed to enhance image contrast. An application example of image preprocessing for the connecting rod of a vehicle engine shows the effectiveness of the proposed method.