The realization of intelligent mining is the only method for realizing high-quality development in the coal industry. As the forefront working link in mine production, achieving automatic roadway tunneling control is key to improving production efficiency, enhancing intelligence, and reducing accident rates at fully mechanized tunneling working faces. Among various detection techniques, machine vision technology stands out with advantages of non-contact measurement, rich information acquisition, and high detection accuracy. The detection and control of tunneling equipment groups based on machine vision has become a research hotspot in the intelligence process of coal mines. This study first introduces the key technologies of a visual detection system, including camera calibration, image preprocessing, feature extraction, visual matching, target segmentation and recognition, visual measurement, and 3D reconstruction. It then elaborates on detection principles, workflows, limitations, precautions, and development status of various vision detection systems in practical application scenarios at tunneling faces, such as tunneling equipments, anchoring systems, transportation systems, and safety auxiliary systems, which significantly improve production safety and efficiency. Finally, considering challenging work conditions and strong interference in mines, the successful adaptation of machine vision to excavation sites relies on addressing technical challenges related to poor environment adaptability, limited imaging field of view, and low intelligence level. Furthermore, according to existing research results and the current technical status, this paper forecasts key technologies that need to be developed in the future for coal mine intelligent equipment systems based on machine vision, including multi-sensor information fusion, equipment group collaborative control, and digital twin-driven remote monitoring.INDEX TERMS Machine vision, coal machinery equipment, detection and measurement, roadway fully mechanized tunneling faces, unmanned coal mine.