Abstract:In recent years, pipeline endoscopic camera detection is a common drainage pipeline defect detection technology. Because most pipelines have a certain sewage filling degree, there are often white fog and spot interference in pipeline images, which affect the identification and interpretation of image defects. Therefore, a method for removing fog and spot interference in drainage pipeline images is proposed, Firstly, combined with the characteristics of pipeline fog, the dark channel a priori method(DCP) is used to remove the fog. Aiming at the problem of color deviation after defogging, a multi-scale Retinex color deviation recovery algorithm is used. Finally, the two-dimensional gamma function is improved to suppress the image spot by reducing lens pollution. The experimental results show that this method can effectively remove the interference of fog and light spot in the video image of drainage pipeline, improve the visibility, contrast and clarity of drainage pipeline image, and improve the accuracy of drainage pipeline disease interpretation.