Faster R-CNN-based bridge crack instance segmentation method

Abstract

The invention belongs to the technical field of image target detection, and particularly relates to a Faster R-CNN-based bridge crack instance segmentation method, which comprises the steps of 1, constructing a bridge crack data set; 2, marking a training sample; 3, building a bridge crack instance segmentation model of the improved Faster R-CNN; 4, training the instance segmentation model built in the step 3; 5, testing the instance segmentation model trained in the step 4; 6, performing actual detection. Compared with the prior art, the invention is higher in robustness, accurate bridge crack classification and positioning results can be obtained, and a high-quality bridge crack segmentation mask can be generated and used for evaluating the damage degree of a bridge and making a corresponding maintenance scheme; in addition, the method can achieve the accurate detection of a plurality of cracks in the image, and can improve the detection efficiency and obtain a complete crack form incombination with the image splicing technology.

Description

technical field [0001] The invention belongs to the technical field of image target detection, and in particular relates to a bridge crack instance segmentation method based on Faster R-CNN. Background technique [0002] As an important carrier connecting two large-span locations, bridges play an important role in my country's road transportation. However, during the long-term sun, rain and load operation of the bridge, the internal stress generated will also be transmitted to some weak parts along the bridge structure, resulting in the occurrence and development of cracks on the surface of the structure at this position. The damage degree of cracks to bridge structures is also different. If the extension direction of surface cracks is perpendicular to the bearing surface of the structure, it is easy to cause unsafe accidents. [0003] According to engineering practice and theoretical analysis, most bridges in service work with cracks, and the potential damage caused by bri...

AI Technical Summary

Problems solved by technology

The traditional detection method based on artificial vision has high cost and low efficiency. The accuracy of detection is affected by subjective factors, and it is increasingly unable to meet the detection requirements of bridge cracks. The existing Faster R-CNN technology is used to detect cracks. The cracks are marked with a rectangular frame (such as the application number 201910526241.2), and the morphological characteristics of the cracks cannot be directly extracted, that is, the damage degree of the cracks cannot be detected intuitively. In addition, the existing technology is basically for a single picture (partial picture) detection, not only the detection efficiency Low, and crack detection is incomplete

Images