Identification, 3D-Reconstruction, and Classification of Dangerous Road Cracks

Abstract

Advances in semiconductor technology and wireless sensor networks have permitted the development of automated inspection at diverse scales (machine, human, infrastructure, environment, etc.). However, automated identification of road cracks is still in its early stages. This is largely owing to the difficulty obtaining pavement photographs and the tiny size of flaws (cracks). The existence of pavement cracks and potholes reduces the value of the infrastructure, thus the severity of the fracture must be estimated. Annually, operators in many nations must audit thousands of kilometers of road to locate this degradation. This procedure is costly, sluggish, and produces fairly subjective results. The goal of this work is to create an efficient automated system for crack identification, extraction, and 3D reconstruction. The creation of crack-free roads is critical to preventing traffic deaths and saving lives. The proposed method consists of five major stages: detection of flaws after processing the input picture with the Gaussian filter, contrast adjustment, and ultimately, threshold-based segmentation. We created a database of road cracks to assess the efficacy of our proposed method. The result obtained are commendable and outperform previous state-of-the-art studies.

Keywords: 3D reconstruction; crack characterization; crack classification; crack detection; image processing; machine learning.

Figure 1

Some examples of real road cracks.

Figure 2

Block diagram of the crack detection and classification system.

Figure 3

Principle of morphological filtering. (a) Binarized Image, (b) Small hole filling results, (c) Skeletonization result.

Figure 4

3D reconstruction of diverse types of cracks.

Figure 5

Region of Interest Extraction (a) Binarized transverse crack (b) Image without crack, (c,d) Noise suppression.

Figure 6

Length and width of the crack. The approximate size of the crack is depicted in the red box using a rectangle.

Figure 7

Crack severity.

Figure 8

Determination of horizontal projection peaks.

Figure 9

Determination of vertical projection peaks.

Figure 10

Hough Transformation for an example of crack.

Figure 11

Choice of the best Hough line (a,c) No line detected, (b) Hough line. The green box illustrates the shape of the crack approximated using a rectangle.

Figure 12

Determination of attributes by the Hough transform.

Figure 13

Variation of interclass primitives (X-axis: Primitives associated with each class of crack; Y-axis: Percentage).

Figure 14

Variation of primitives for a transversal crack. The red, blue, and green colours represent three samples from the same transversal class to prove the intraclass variations of the primitives.

Figure 15

Location of the crack: (a): Original image; (b): Binarized image; (c): Denoising of the binarized image; (d): Skeletonization; (e): Crack detection.

Figure 16

An example of a crack for each class.

Figure 17

Examples of images from our dataset obtained in static mode.

Figure 18

Crack detection system.

Figure 19

Example of false classification of cracking defect.

Figure 20

Recognition rate for each class of crack.