Pipeline abnormality type automatic detection method based on deep learning

Abstract

The present invention relates to a pipeline abnormality type detection method based on deep learning. The method comprises: extracting video frames from the video captured by a pipeline robot to forma pipeline image set and an image tag set; dividing the pipeline abnormality types into 7 situations of normal pipelines, pipeline staggered joints, pipeline ring cracks, pipeline long cracks, tree roots entering, accumulated floats, and water in the pipeline; establishing a training data set, dividing pictures into a training set, a verification and test set, as well as a corresponding tag set; and using the produced data set to train the convolutional neural network through the error back-propagation algorithm BP, and outputting the probability that the image corresponds to 7 abnormality types.

Description

technical field [0001] The present invention relates to the fields of computer vision, computer image processing, deep learning and the like, and in particular to a method for judging abnormal types of sewer pipes based on deep learning technology. Background technique [0002] In recent years, with the development of my country's economy, the scale of urban construction has become larger and larger. my country plans to build a relatively complete urban underground drainage pipeline system in about 10 years, so that the construction and management level of underground drainage pipelines can meet the needs of economic and social development. , The ability of emergency disaster prevention has been greatly improved. However, due to various reasons, there are different degrees of abnormal phenomena in the underground drainage pipes of many cities in our country, which brings endless troubles to urban drainage and even ground transportation, and seriously affects people's daily li...

AI Technical Summary

Problems solved by technology

A large amount of drainage pipeline video data can be obtained through it. At present, the staff still need to manually watch the pipeline video and analyze the type and degree of pipeline anomalies based on experience. It often takes a lot of time and energy, and it is difficult to achieve accurate pipeline anomaly detection. automation, automation and intelligence

In order to improve the deficiencies of CCTV, Tang Yiping and others (Patent Publication No. CN104568983A) disclosed a pipeline internal defect detection device based on active panoramic vision. This device uses traditional computer vision and image processing methods to identify abnormalities in pipelines. The accuracy depends heavily on the quality of the captured pipeline images, so various sophisticated sensors need to be installed, the production cost is high, and it cannot be put into large-scale production

Images