Coal mine underground drill rod counting method based on machine vision

The invention provides a coal mine underground drill rod counting method based on machine vision, and belongs to the technical field of drill rod counting. The problems that an existing method for counting drill rods through machine vision is low in precision and large in error are solved. The method includes the following steps that feature targets are arranged on a drilling machine mechanical arm, and underground drilling machine construction videos with the feature targets are obtained to serve as a data set; preprocessing the data set; a target detection model based on deep learning is constructed and trained, the preprocessed video is input into the target detection model, and a feature target on the drilling machine mechanical arm is detected; after the feature target is detected, the motion trail of the feature target of the mechanical arm is obtained through the visual SLAM technology in combination with a target detection result; whether the drilling machine mechanical arm has rod mounting and rod di