Research on Removing Shadow in Workpiece Image Based on Homomorphic Filtering
Weidong Yang, Wei Guo, Kai Peng, L.B. Liu · Procedia Engineering · 2012
The shadow in the image has a great influnece for the accuracy of the image segmentation and measurement, while the workpiece is detected in the machine vision system. In the paper a homomorphic filter algorithm is designed for removing shadow, through the analysis of the relationship between the shadow and the brightness of the image. Firstly the image from RGB form is converted to HSV form, based on the relationship between the variation of the Vcomponent and the shadow, the components of the related shadow can be identified. Then a filter is designed, which uses the principle of homomorphic filtering to filter the component of the related shadow in the image. Finally, a new complete image is formed by using the method of image synthesis to incorporate the treated V-component into the original image. The experiment results show that this method can remove the cast shadows in the image effectively, which has improved the image quality for the subsequence image processing.