Retaining wall crack identification method based on CNN and SVM

Abstract

The invention relates to a retaining wall crack identification method based on a CNN and an SVM. An unmanned aerial vehicle equipped with a monocular camera is adopted to collect and establish a retaining wall crack identification database; image data preprocessing and data enhancement are carried out; a retaining wall crack recognition algorithm model is established based on the CNN and the SVM;the algorithm model is trained to obtain an optimal retaining wall crack recognition algorithm training model; the unmanned aerial vehicle collects retaining wall pictures in real time and uploads theretaining wall pictures to the upper computer through a cloud network; and the upper computer carries out preprocessing and carries out crack identification on the picture of the retaining wall by using a pre-trained model. Based on the images collected by the unmanned aerial vehicle, the cracks of the retaining wall can be accurately recognized through the intelligent algorithm, the retaining wall can be repaired easily, the manual inspection cost is reduced, and the efficiency is improved.

Description

technical field [0001] The invention relates to the field of crack identification in retaining walls, in particular to a method for identifying cracks in retaining walls based on CNN and SVM. Background technique [0002] A retaining wall refers to a structure that supports roadbed fill or hillside soil and prevents the deformation and instability of fill or soil. The existence of the retaining wall reduces soil erosion and landslides, which is of great significance to the protection of natural features and the safety of people's property. With the progress of time and the washing of wind and rain, cracks will inevitably occur in the retaining wall, which will greatly weaken the performance of the retaining wall, so the accurate identification of the cracks in the retaining wall needs to be solved. [0003] At present, the method of manual regular inspection is mostly used, but there is great subjectivity, manual inspection is time-consuming and labor-intensive, the labor c...

AI Technical Summary

Problems solved by technology

[0003] At present, the method of manual regular inspection is mostly used, but there is great subjectivity, manual inspection is time-consuming and labor-intensive, the labor cost is high, and the work efficiency is low

Moreover, the geographical location of some retaining walls is rugged and steep, which is not conducive to inspectors walking and visual observation, which may easily lead to missed inspections

Images