Real-time visual detection method for high-speed periodic motions

Abstract

The invention relates to a real-time visual detection method for high-speed periodic motions, and belongs to the technical field of visual detection. The real-time visual detection method comprises the following steps of: extracting features of a periodic motion; obtaining a circumferential angle of the periodic motion; establishing a high-frame rate image sample library on the basis of the circumferential angle of the periodic motion; and carrying out real-time visual detection and exception extraction on the basis of the high-frame rate image sample library. The method is a monocular cameraand periodic feature mark-based rapid detection algorithm, and is capable of carrying out lossless and visualized real-time detection on periodic mechanical motions, automatically extracting exceptionconditions and then realizing real-time visual detection for periodic motions at high speeds.

Description

technical field [0001] The invention relates to the technical field of visual detection, in particular to a visual real-time detection method for high-speed periodic motion. Background technique [0002] Visual detection technology is more and more used in the production and life of human society. The conventional visual technology is to use the camera to collect the target image, save the video image data in the computer, and study the corresponding image processing and video processing algorithms to complete the Automatic detection and recognition of visual images. According to the real-time processing speed, visual inspection technology can be divided into two categories: ordinary visual inspection and high-speed visual inspection. Ordinary visual inspection systems usually have a real-time processing speed of 25-30 frames per second, and the processing speed of high-speed visual real-time detection is generally above 128 frames per second, which is used to collect and d...

AI Technical Summary

Problems solved by technology

However, this method has the following technical defects: 1. This method needs to store a large amount of normal motion high frame rate image data, and use complex image processing algorithms to calculate and extract key pixels. The offline processing process is complex and time-consuming, usually taking more than ten minutes The offline calculation process, so the biggest disadvantage is the large amount of data, the amount of calculation, and the complexity of the algorithm

2. This method is mainly aimed at high-speed periodic motion without obvious circular motion area, so it is impossible to find a suitable mark point

Since there is no obvious periodic motion feature, the result of this method is unstable in the calculation of periodic eigenvalues, and more noise interference will be generated.

Images