Abstract :
To the pipeline welding defect detection, robot tracking weld reliably, the welds original image acquired by visual sensor need pretreatment to eliminate the impact of noise. A pipeline welding machine vision detection system proposed in this paper. With the help of embedded computer and takes full use of the traditional ultrasonic-testing device, FPGA, USB, electronic technology and wavelet transform signal processing the system realizes data collection, save, display, analysis and processing. Using of neighborhood mean filter for smoothing, feature prominent method to segmentation image. Image after smooth will be the binarization processing. Then remove the small area noise with labeling method, get a clear image of the weld. The level of projection method to the recognition of weld image and determine the location of weld. The experiments show that the advantage of small, plug and play, stability, high rate, efficiency and reliability, the pretreatment method solute the prevailing situation of uneven illumination, and also reducing the size of the processing image, meeting the needs of the pipeline weld tracking real-time detection.
Keywords :
failure (mechanical); image recognition; image segmentation; pipelines; production engineering computing; robot vision; robotic welding; smoothing methods; ultrasonic materials testing; wavelet transforms; welding; welds; FPGA; USB; binarization processing; electronic technology; embedded computer; feature prominent method; labeling method; machine vision; neighborhood mean filter; pipeline weld detection system; pipeline welding defect detection; robot tracking; segmentation image; smoothing method; ultrasonic-testing device; visual sensor; wavelet transform signal processing; weld image recognition; Computer displays; Embedded computing; Field programmable gate arrays; Image sensors; Machine vision; Pipelines; Robot sensing systems; Universal Serial Bus; Wavelet analysis; Welding; Image processing; Object recognition; computer vision; robot;
