Detection Method of Casing Joint based on Computer Vision

Aiming at the current problem of low detection efficiency and high detection error when manually inspecting the casing joint for calibration depth in the visual inspection of oil and gas wells, a casing joint intelligent recognition based on YOLOv5 algorithm is proposed, which can realize the intelligent detection of casing joint. Firstly, a large number of pictures of oil and gas well casing joints were collected and the dataset was made by data enhancement method. Then, the enhanced dataset which was annotated with Labelimg tool was sent to YOLOv5 network for training. Finally, use the best trained weights for testing result. The test results show that the method has a high detection accuracy, short detection time, wide applicability, and great advancement and practicality in casing joint detection.