Multi-degree-of-freedom pose real-time adjustment binocular vision detection system, method and device

Abstract

The invention belongs to the technical field of mechanical engineering, and discloses a multi-degree-of-freedom pose real-time adjustment binocular vision detection system, method and device. The system comprises a multi-degree-of-freedom pose real-time adjustment binocular vision module and a binocular vision real-time calibration module. The multi-degree-of-freedom pose real-time adjustment binocular vision module comprises a sensing unit, a rotation control unit and a displacement control unit for sensing visual information; and the binocular vision real-time calibration module is used for calibrating the binocular vision module by using a high-definition camera to perform multi-degree-of-freedom pose real-time adjustment. Adaptive adjustment of binocular vision and stability of real-time calibration can be realized at the same time. On the basis of a basic binocular vision system, the servo motor, the electric push rod and the monocular high-definition camera are added to perform real-time pose evaluation on the sensing camera, so that real-time calibration is realized, the robustness is relatively high, and the adaptability to the environment is relatively high.

Description

technical field [0001] The invention belongs to the technical field of mechanical engineering, and in particular relates to a binocular vision detection system, method and device for real-time adjustment of multi-degree-of-freedom poses and postures. Background technique [0002] At present, with the rapid development of my country's national economy and the improvement of people's quality of life, the country pays more and more attention to a series of smart technologies such as smart manufacturing, smart cities, and smart parks. The binocular stereo vision that reflects artificial intelligence is becoming more and more important. It is in Intelligent production, traffic monitoring, machine navigation, aerospace, medical modeling, visual simulation, restoration of cultural relics, non-contact high-precision measurement, etc., provide technologies including virtual visual simulation, visual recognition and positioning, which can effectively improve production efficiency and Pr...

AI Technical Summary

Problems solved by technology

[0012] At present, the relatively mature calibration method is Zhang Zhengyou calibration method. Its calibration principle is to calculate the relative positional relationship of binocular cameras by shooting standard checkerboard image information, so as to complete the calibration. It is an indirect calibration method. Binocular cameras are widely used in actual use, but this solution cannot be calibrated in real time, which has great limitations

[0013] At present, the use of binocular cameras is mostly static calibration before use after installation, and then used. If repeated installation or loosening occurs during use, re-calibration is required. Therefore, the pose and angle of view of the camera are determined each time. can no longer be adjusted

However, in actual use, the range of viewing angles provided by the two completely fixed cameras is limited, and external factors such as obstructions will further increase the blind area of ​​the field of vision, greatly reducing the perception and recognition capabilities of the binocular camera.

[0014] Therefore, the limitations brought by the fixed viewing angle of the binocular camera form a contradiction with the real-time calibration required for real-time adjustment of the camera, that is, according to the existing related solutions and technologies, the real-time adjustment of the viewing angle of the binocular camera and the real-time adjustment of the binocular vision system cannot be satisfied at the same time. real-time calibration

Images