Multi-index dam defect image detection method based on DS optimization

Abstract

The invention discloses a multi-index dam defect image detection method based on DS optimization. The method comprises the following steps of a), carrying out multi-index defect detection, judging whether a defect exist or not through comparison and analysis of the difference between history normal image characteristic quantity parameters (an average information entropy, an average signal to noiseratio, average gradient average value contrast and average gray variance offset) and test image characteristic quantity, wherein the key point in the phase is setting of each index threshold; and 2),carrying out fusion decision based on DS optimization, combining detection results of four indexes through adoption of an optimized DS algorithm, and outputting a final detection result. Through utilization of a common DS evidence theory, the high collision problem among evidences under the uncertain condition cannot be solved well, so the decision accuracy is liable to reduce or even a decisionerror is liable to occur; and in order to reduce the evidence collision, through utilization of an improved DS evidence theory algorithm, the validity of each index evidence is defined through importing of adaptive weight, the evidence probability function allocation is corrected, and a relatively accurate defect detection effect is finally obtained.

Description

technical field [0001] The invention relates to a method for detecting defect images of dams with optimized multi-indicators based on DS, in particular to a method for realizing optimized detection of defect images, which belongs to the field of distributed data fusion. Background technique [0002] The world's dam construction has a long history, and my country is a big water conservancy country. At present, it has more than 98,000 reservoir dams, which is the country with the most reservoir dams in the world. However, with the passage of time and the growth of the age of the dam, affected by various factors such as temperature and environmental pressure, the operating form of the dam will change, and a series of dam surfaces such as deformation, cracks, leakage, and calcification precipitation will appear. Visual defects. [0003] The image acquisition device deployed inside the dam regularly collects the images of each collection point. Due to the large number of collect...

AI Technical Summary

Problems solved by technology

[0004] Purpose of the invention: Aiming at the problems of low resource utilization, low decision-making accuracy and even wrong decision-making in the multi-index defect image detection method, the present invention provides a DS-based optimized multi-index dam defect image detection method to improve the detection efficiency of multi-index defect images, To more accurate defect detection effect

Images