Pipeline defect identification method based on computer vision and machine learning

Abstract

The invention relates to a pipeline defect identification method based on computer vision and machine learning. The method comprises the following steps: carrying out gray scale stretching on pipelineimages and videos collected by a pipeline robot; performing sliding filtering processing to filter noise interference; extracting the outline of the suspected defect area by adopting a Canny operator; dividing the acquired image into gradients or edge direction histograms of all pixel points in a cell unit, and combining the histograms to extract HOG features of the image; training a VGGNet modelby adopting a transfer learning strategy, and extracting image features by adopting a trained VGGNet feature extractor; fusing the HOG features of the pipeline image and the features extracted by theVGGNet, classifying the features by using an SVM, and identifying the pipeline defects; and counting the pipeline defect identification result and outputting the pipeline defect type. Defect characteristics of a real pipeline environment can be recognized, robustness and recognition accuracy of a classification model are improved, system image processing speed and recognition efficiency are improved, and the method is suitable for the field of urban pipeline and industrial pipeline maintenance.

Description

technical field [0001] The invention relates to the technical fields of image and video processing and feature recognition, in particular to a recognition method for detecting pipeline defects based on computer vision and machine learning. Background technique [0002] Pipeline defect detection and repair work is an important part of urban construction, and has become a hot research issue in computer vision. Pipeline defect recognition is to identify defect parts from video frames according to specific technical indicators. Defect identification plays a decisive role in later pipeline repair. At present, pipeline defect identification is mainly through pipeline robots carrying closed-circuit television (CCTV) hardware to obtain pipeline video, and technicians analyze the video, and manually identify pipeline defects. However, this method is highly dependent on labor and requires experience. Only qualified operators can accurately identify diseases, and human identification...

AI Technical Summary

Problems solved by technology

[0003] At present, there are some problems in the field of pipeline defect recognition, such as the complexity of the pipeline environment in the video, the differences of pipelines and the differences of defects, which lead to the fact that in the actual complex pipeline environment, the characteristics of pipeline defects are easily affected by changes in the pipeline environment, and the classification model is anti-interference The ability is poor, and the accuracy of pipeline defect identification is low in the pipeline environment with large differences

Images