Three-dimensional-based pipeline crack detection method, device and apparatus and storage medium

Abstract

The invention discloses a three-dimensional-based pipeline crack detection method, device and apparatus and a storage medium. A depth camera is utilized to perform three-dimensional reconstruction ona pipeline, data acquired by the depth camera is not limited by brightness, the data volume is more than that of an image by depth information, and the data can be better analyzed. Three-dimensional reconstruction is performed on the data acquired by the depth camera by using a PCL library, a three-dimensional model of the pipeline is reconstructed, the three-dimensional model is refined, and thegenerated model is textured to obtain the crack characteristics of the model. Through three-dimensional reconstruction of a pipeline in a normal state before, the normal state of the pipeline is compared with a to-be-detected state, if the to-be-detected state goes wrong, the object is identified again, and whether the object goes wrong or not is ensured through two times of reconstruction and comparison. According to the invention, misjudgment during image recognition is reduced by adding the depth information and fully utilizing the depth information, and the judgment accuracy is improved because the method is not limited by a brightness environment.

Description

technical field [0001] The invention relates to the technical field of pipeline crack detection, in particular to a three-dimensional-based pipeline crack detection method, device, equipment and storage medium. Background technique [0002] Pipelines play an extremely important role in petrochemical and other production. Therefore, timely and accurate detection of pipeline defects and hidden dangers has important practical significance for ensuring social production and life as well as people's lives and property. [0003] At present, in view of the characteristics of pipeline cracks, the image processing method is used to automatically detect and identify cracks, grayscale the collected pipeline crack images, and then perform image enhancement to reduce the noise in the image, and then select an appropriate threshold to proceed. Segmentation to extract crack features. However, the pipeline information obtained by image processing is only two-dimensional, and its informati...

AI Technical Summary

Problems solved by technology

However, the pipeline information obtained by image processing is only two-dimensional, and its information has limitations. If the brightness threshold of the image is not good, it will cause poor image quality, and image quality problems will lead to various problems in subsequent data processing. the emergence of

Images