Unlock instant, AI-driven research and patent intelligence for your innovation.

Asphalt pavement crack rapid classification method based on semantic segmentation

A technology of asphalt pavement and semantic segmentation, applied in special data processing applications, instruments, biological neural network models, etc., can solve problems such as unsatisfactory application effects, inability to accurately extract target information, low signal noise in two-dimensional images, etc., to achieve Effects of improving efficiency and accuracy, eliminating subjectivity, and increasing robustness

Pending Publication Date: 2021-03-19
ZHEJIANG SCI RES INST OF TRANSPORT
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the signal-to-noise ratio of the two-dimensional image is low, and the target information cannot be accurately extracted, so the application effect is not satisfactory

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 more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Asphalt pavement crack rapid classification method based on semantic segmentation
  • Asphalt pavement crack rapid classification method based on semantic segmentation
  • Asphalt pavement crack rapid classification method based on semantic segmentation

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0052] 1. Use the Pathrunner comprehensive detection vehicle to obtain asphalt pavement images, and establish the original image set of asphalt pavement;

[0053] 2. Carry out manual cleaning according to the asphalt pavement images collected by the Pathrunner comprehensive inspection vehicle, and manually clean up 600 samples of non-disease-free pavement samples, distorted samples, and samples with possible diseases (1,800 samples in total);

[0054] 3 Manually classify and label the selected samples, and use 525 cases of each type of samples as training samples and 75 cases as test samples;

[0055] 4. Use the residual neural network ResNet50 to perform model training according to the training set data after classification and labeling. The model structure diagram of ResNet50 is as follows figure 2 Shown, and the model performance is verified by the test set data; when the F of the test set 1 When the value is greater than 90, it is considered that the cleaning algorithm c...

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

No PUM Login to View More

Abstract

The invention provides an asphalt pavement crack rapid classification method based on semantic segmentation, and the method comprises the steps: obtaining an asphalt pavement image, and building an asphalt pavement original image set; performing data cleaning on the acquired crack images by using a data cleaning algorithm, and constructing an asphalt pavement crack image sample data set; on the basis of the cleaned bituminous pavement crack image sample data set, adopting manual segmentation and labeling, and providing sufficient training data for the model through data augmentation; finally,performing crack image semantic segmentation by using a Unet network, and finally outputting crack information. The subjectivity of manual detection and manual analysis is eliminated, the efficiency and accuracy of asphalt pavement crack detection are improved, and the detection cost is reduced; the semantic segmentation network is used as the basis of automatic classification, and compared with traditional image processing analysis, the robustness of crack recognition is greatly improved.

Description

technical field [0001] The invention belongs to the technical field of road engineering detection, and in particular relates to a rapid classification method for asphalt pavement cracks based on semantic segmentation. Background technique [0002] As an important part of highway engineering, asphalt pavement is an important infrastructure to promote economic development. With the increase of service life, different forms of diseases appear in asphalt pavement, among which pavement cracks have become the most important disease of highways and urban roads. Detecting and identifying pavement cracks is one of the key tasks of regular maintenance and management of asphalt pavement. [0003] Road inspectors need to manually check whether there are cracks on the asphalt pavement, and judge the classification and severity of the detected cracks. With the development of hardware such as computers and cameras, detection equipment can be mounted on vehicles, and detection work can be ...

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
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06K9/62G06F16/215G06N3/04
CPCG06F16/215G06N3/045G06F18/241G06F18/214
Inventor 章天杰韩海航王洋洋
Owner ZHEJIANG SCI RES INST OF TRANSPORT