Development Of Aircraft Rivet Size Detection System Based On Image Recognition

Rivet, as an important aircraft fastener, its dimension accuracy and quality detecting are signif icant to the process ensuring quality of aircraft assembly. The traditional rivet detection methods have some disadvantages, for example, low detection efficiency, precision is not easy to control, it is influenced by artificial factors and easy to damage the rivet and other defects. A rivet detection system based on image recognition is proposed in this paper, and the related research work has been carried on. The main research content includes the following several aspects.Firstly, through analyzing the function demand of rivet detection size system and combining with the rivet characteristics, the measurement system is analyzed in theory. The overall des ign of the rivet detection system is completed and the final form of realizing the system is determined.Secondly, the key components such as monochrome industrial camera, lenses and light source are selected according to the detecting requirements. The transport structure and clamp structure of the rivet are designed. The assembly for the mechanical structure of the detection system is completed.Thirdly, through the analys is and research of image processing algor ithms,the noisy image is processed and analyzed using two classic filter ing algor ithms. In the end of this part, an improved median filter algorithm is proposed. This algorithm adds the judgment for marking signal and noisy pixels, which can effectively improve the effect.Finally, through the research both on the model of the pinhole camera and camera calibration method, the calibration of the system is completed. In order to facilitate human visual inspection system interface, a system is developed adopting the modular design concept.The system, as an automatic detecting equipment has follow ing advantages such as simple structure, low cost and high efficiency, the measurement precision can reach 0.01 mm.It can satisfy the requirements of on-line measurement and real-time control for modern aircraft assembly, and improve the level of assembly automation and assembly quality.