Method for detecting crack in bridge crack image in complex scene
A complex background and detection method technology, applied in the field of computer vision, can solve problems such as unsatisfactory results, complex background textures, and small image areas, and achieve good generalization ability, reduced training time, and strong recognition effects.
Active Publication Date: 2018-11-23
SHAANXI NORMAL UNIV
View PDF4 Cites 17 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
If the existing semantic segmentation model is directly applied to the detection of bridge pavement cracks, due to the linear topological structure of the cracks, which occupy a small area of the entire image, and the background texture is complex and there are many obstacles, satisfactory results cannot be achieved.
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0040] The present invention will be described in further detail below in conjunction with specific examples, but the embodiments of the present invention are not limited thereto.
[0041] Such as Figure 17 As shown, this embodiment provides a method for crack detection of bridge crack images under complex backgrounds, which includes the following steps: Step 1, performing geometric transformation, spatial filtering, and linear transformation on the collected original bridge crack images, and then generating sub-models and discriminating The sub-model performs data set amplification; the generated sub-model sequentially includes a fully connected layer, a dimension conversion layer, a first transposed convolution layer, a second transposed convolution layer, a third transposed convolution layer, and a fourth transposed convolution Convolution layer and fifth transposed convolutional layer; first transposed convolutional layer, second transposed convolutional layer, third tran...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention relates to a method for detecting a crack in a bridge crack image in a complex scene, the method comprising the steps of: firstly, according to a deep convolutional generative adversarial network principle, providing and using a bridge pavement crack image generation model to amplify a data set; secondly, constructing a bridge pavement crack image segmentation model based on semanticsegmentation for crack features; finally, extracting the crack in the crack image through the bridge pavement crack image segmentation model. According to the method for detecting a crack in a complex bridge crack image, the data set amplification effectively alleviates the poor fitting phenomenon due to insufficient training data, and the precision rate and the recall rate are increased by 79.4%and 74.7%, respectively. Compared with an existing semantic segmentation algorithm, the number of parameters used in the method of the invention is reduced, the training time is reduced, and the precision rate, the recall rate and the F1 score are increased and reach 92% or more. Compared with the existing bridge pavement crack detection and segmentation algorithm, the method of the invention ismore suitable for bridge pavement crack detection and segmentation in a complex scene, and has better identification effect and better generalization ability.
Description
technical field [0001] The invention belongs to the technical field of computer vision, and in particular relates to a method for detecting cracks in bridge crack images under complex backgrounds. Background technique [0002] Transportation is the basic need and prerequisite for economic development, the survival foundation and civilization symbol of modern society, and the infrastructure and important link of industrial development. It is related to the development of the national economy and carries the lifeline of social progress. According to the "Statistical Bulletin on National Economic and Social Development in 2017", in 2017, 6,796 kilometers of highways were newly rebuilt in my country, and 2,182 kilometers of new high-speed railways were put into operation. Influence has increased significantly. In the construction of modern transportation, bridges account for 50% of the lines in the high-speed railways that have been constructed, and the proportion of bridges in ...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More IPC IPC(8): G06T7/00G06T7/10G06N3/04
CPCG06T7/0004G06T7/10G06T2207/20081G06T2207/10004G06N3/045
Inventor 李良福孙瑞赟
Owner SHAANXI NORMAL UNIV