Self-adaptive Canny aircraft skin infrared image edge detection method based on FFWA

Abstract

The invention discloses a self-adaptive Canny aircraft skin infrared image edge detection method based on FFWA. The method comprises the following steps: acquiring an aircraft skin damage infrared image; denoising the image; solving a gradient value and a maximum gradient matrix; performing non-maximum suppression on the maximum gradient matrix; solving a maximum between-class variance, and optimizing the maximum between-class variance by using a fast fireworks algorithm to obtain an optimal segmentation threshold value; and carrying out double-threshold division on the image to finally obtain an edge detection image and the like. The method has the advantages that the double thresholds of the Canny algorithm are calculated through the Otsu method, manual preset errors are avoided, the Canny algorithm has self-adaptability, the image edge detection effect is better and more accurate, and therefore the method is more suitable for actual operation. Compared with a traditional firework algorithm, the improved fast firework algorithm saves the calculation time by 36%, the stability is also considerably improved, and the calculation speed can be increased by 49% compared with the calculation speed before improvement under the condition that the precision is kept unchanged.

Description

technical field [0001] The invention belongs to the technical field of non-destructive detection of skin damage of civil aircraft, and in particular relates to an adaptive Canny aircraft skin infrared image edge detection method based on FFWA. Background technique [0002] Aircraft skin is generally a visual structure fixed on the outside of the aircraft by adhesives or rivets. The main function is to make the aircraft have better aerodynamic characteristics and transfer loads. Since the skin is in direct contact with the external environment, its structure is prone to damage in the form of water accumulation, debonding, cracks, and corrosion, resulting in structural damage to the aircraft, disintegration in the air, and other devastating disasters. Infrared detection is to use the principle of infrared radiation to detect the surface of materials. If the surface to be tested is damaged, the thermal conductivity of the material will change, and a large thermal radiation gra...

AI Technical Summary

Problems solved by technology

Better threshold accuracy can be obtained, but the maximum inter-class variance requires traversal calculations on image pixels, which is complex and slow when processing large images

Fireworks Algorithm (FWA) is an efficient intelligent optimization algorithm, which can greatly reduce the calculation amount of the maximum inter-class variance method, but FWA converges slowly when optimizing the fitness function near the optimal non-zero point And it is prone to local optimal situations. In addition, the Gaussian mutation spark in FWA has little effect on improving the diversity of feasible solutions, so the algorithm needs to be improved

Images