A multi-target full-pixel segmentation method for aerospace material damage detection images

Abstract

The invention discloses a multi-objective full-pixel segmentation method for damage detection images of aerospace materials, including: extracting typical transient thermal responses of each type of defect; obtaining infrared reconstruction images; combining multi-objective optimization algorithms with segmentation models for infrared reconstruction images Separation of background area and defect area; construction of infrared image segmentation function under the guidance of noise removal, detail preservation and edge preservation; multi-objective optimization algorithm combined with segmentation model to achieve one-time segmentation of defects; define sparseness according to Euclidean distance According to the size of the level of sparsity, the weight vector is adjusted based on the individual with a large level of sparsity; the segmented image of the damage defect in the infrared detection image is obtained. The invention utilizes the main features to reconstruct the sequence of infrared thermal images to obtain the infrared reconstructed image of the defect, which reflects the defect characteristics of the test piece. The result image obtained by target segmentation on the infrared reconstructed image can not only realize the noise elimination but also ensure the detail preservation, and the edge preservation improves the accuracy of the image segmentation.

Description

technical field [0001] The invention belongs to the field of damage detection application and pattern recognition technology, and more specifically, the invention relates to a multi-object full-pixel segmentation method of an aerospace material damage detection image. Background technique [0002] During launch and in-orbit operation, spacecraft are extremely vulnerable to accidental impacts from various tiny objects, such as space junk fragments, tiny meteoroids, and peeled off coatings. In particular, the ever-increasing space junk debris is the most harmful to on-orbit spaceflight. Because the tiny debris has a very high impact speed (usually reaching several kilometers per second or even tens of kilometers per second), it is easy to cause various types of debris on the surface of the spacecraft. Hypervelocity impact damage, such as perforation, impact crater, spallation, spalling, etc., causes the structure of the spacecraft to be damaged or the function of the component...

AI Technical Summary

Problems solved by technology

If only the segmentation model is used, the weight coefficients corresponding to the objective functions to achieve the three segmentation performances when forming the segmentation function are to be determined, and continuous debugging is required to determine the weight coefficients to control the balance between each objective function, and the computational efficiency and universality of the algorithm The performance is low, and the segmentation quality of the final infrared reconstructed image cannot be guaranteed

Images