High-pressure container detection method based on infrared image splicing fusion algorithm

Abstract

The invention discloses a high-pressure container detection method based on an infrared image splicing fusion algorithm. The method comprises the following steps: performing data acquisition on the surface of a high-pressure container; performing K-means and GMM algorithm combined processing on all column vectors, and processing video stream signals to obtain a defect reconstruction image; for thereconstructed defect information images including the overlapping regions in pairs, adopting a FAST-SURF algorithm to perform feature extraction, adopting a bidirectional FLANN algorithm to perform rough matching, an MSAC algorithm is adopted to purify matching point pairs, and a geometric transformation model is estimated, so that an image registration process can be realized; processing the splicing result by adopting a fusion method based on brightness adjustment and distance weighting according to the gap and brightness and color difference generated by the splicing result; obtaining a panoramic spliced image for infrared detection of the high-pressure container; the invention provides the high-pressure container detection method based on an infrared image splicing fusion algorithm aiming at the requirements of large volume and in-situ detection of a high-pressure container, and the speed and the precision of image splicing are comprehensively considered.

Description

technical field [0001] The invention belongs to the technical field of infrared nondestructive testing, and more specifically, the invention relates to a high-pressure container detection method based on an infrared image splicing and fusion algorithm. Background technique [0002] High-pressure vessels are widely used in the aerospace industry, such as the construction of launch vehicles, space stations, wind tunnels and other equipment are inseparable from the use of high-pressure vessels. However, due to the lack of supervision during the construction of the equipment, there are hidden concerns about the quality. After the equipment has been in operation for a period of time, there will be reports of cracks and corrosion leaks. As time goes by, the probability of safety problems in the operation of these high-pressure vessels will continue to increase. Safety problems during the period will cause disastrous consequences, and the satellite launch bases and high-density tes...

AI Technical Summary

Problems solved by technology

The feature points distributed on the surface of the high-pressure vessel are difficult to extract. Traditional feature point-based feature detection methods such as SIFT and SURF are complex in calculation and the number of feature points detected at the same time is not ideal.

At the same time, the features extracted by this type of algorithm have the characteristics of scale invariance and rotation invariance, but they are partially invariant to illumination changes and affine and perspective transformations.

Therefore, the image stitching result obtained by simply using the above matching algorithm will be affected by factors such as shooting angle and ambient temperature, resulting in obvious brightness differences and stitching gaps.

Images