Three-dimensional weaving guide rod online insertion detection method based on machine vision

Abstract

The invention discloses a three-dimensional weaving guide rod online insertion detection method based on machine vision, and the method specifically comprises the following steps: collecting an image above a guide template, training a Faster RCNN model until a loss function converges, verifying the Faster RCNN model through a test set, and when an output guide rod circular feature extraction frame is compared with an artificially marked guide rod, carrying out the verification of the Faster RCNN model. When the accuracy is higher than 95%, completing the training of the Faster RCNN model, inputting an image, collected in real time, above the guide template into the trained Faster RCNN model, outputting a feature extraction frame and a geometric center position thereof, and obtaining a feature extraction result; and the output feature extraction frame and the geometric center position thereof are mapped to a guide hole state detection matrix with the same row number and column number as the array holes in the guide template, and the guide holes without the guide rods inserted in the guide template are positioned. The machine vision-based online insertion detection method for the three-dimensional weaving guide rod has the characteristics of intelligence, high efficiency and accuracy, and the insertion detection efficiency and reliability of the guide rod can be greatly improved.

Description

technical field [0001] The invention relates to the field of visual inspection-assisted three-dimensional weaving machine guide rod placement, in particular to an on-line detection method for three-dimensional weaving guide rods based on machine vision. Background technique [0002] Three-dimensional woven composite materials are widely used in many industries such as automobiles, aerospace, and national defense because of their excellent properties such as light weight, high strength, high modulus, fatigue resistance, corrosion resistance, designability, and good manufacturability. With the rapid development of related industries, the requirements for the weaving and forming methods of three-dimensional woven composite preforms are getting higher and higher. Compared with traditional two-dimensional structural composite materials, three-dimensional woven composite materials are composite materials with three-dimensional overall fabrics as reinforcements, which interweave ea...

AI Technical Summary

Problems solved by technology

Arrange guide sleeve arrays with equidistant cam grooves on the guide template, and weave staggeredly around the guide sleeve cam grooves during weaving; while the 3D weaving equipment based on the digital guide template needs to be re-inserted on the guide template before each weaving For the guide rod, due to the small array aperture and the large number of holes on the guide template, the manual detection method is still used for the detection work, which has low detection efficiency and high cost, and manual detection is easy to miss, which affects the weaving effect

Therefore, manual detection methods are difficult to meet the needs of 3D weaving forming equipment based on digital guidance templates

Images