A Method of Defect Feature Extraction in Infrared Thermal Image Based on Multi-objective Optimization

Abstract

The invention discloses a method for extracting defect features of infrared thermal images based on multi-objective optimization. The transient thermal response of pixels is selected by transforming the step length of thermal image sequences, and FCM is used for classification to obtain the transient thermal response of each pixel. The category of the thermal response, and then consider the similarity of the pixel value (temperature value) between each category of pixels and similar pixels, and at the same time consider the difference between the pixel (temperature point) and different categories of pixels (temperature point), construct The corresponding multi-objective function uses the decomposition-based multi-objective evolutionary algorithm to obtain the dimensionality reduction results of the thermal image sequence, and finally uses the pulse-coupled neural network for feature extraction to extract the defect features of the infrared thermal image. Through the above steps, the precise selection of representative pixel points (temperature points) is realized, which ensures the accuracy of defect feature extraction.

Description

technical field [0001] The invention belongs to the technical field of defect detection, and more specifically relates to a method for extracting defect features of infrared thermal images based on multi-objective optimization. Background technique [0002] Infrared thermal image detection technology obtains structural information on and below the surface of the material by controlling the thermal excitation method and measuring the temperature field change on the surface of the material, so as to achieve the purpose of detection. When acquiring structural information, infrared thermal imaging cameras are often used to record the temperature field information of the surface or subsurface of the specimen over time, and convert it into a sequence of thermal images for presentation. Due to the huge amount of data and strong noise interference of the thermal image sequence obtained by the infrared thermal imager, in order to obtain better detection results, it is necessary to ex...

AI Technical Summary

Problems solved by technology

In this process, the representative temperature points of each category are obtained by using the correlation between different categories, but the similarity between the representative temperature points and the same temperature points is not studied, and the selected representative temperature points are not enough to characterize the characteristics of this category, so Goals that need to consider both differences and similarities

In addition, the method is to search for a regionally representative thermal response temperature point in each category, which is the distance from other cluster centers and the largest thermal response data in the corresponding category, and the thermal response data of the representative temperature points of all categories The response data constitutes a two-dimensional matrix, and these representative temperature points are incomplete for the information representation of the corresponding category, so the defect features extracted after linear transformation are inaccurate, thus failing to achieve a certain accuracy

Images