Evaluation of the low-cost depth cameras for non-destructive testing
The primary aim of this paper is to assess the effectiveness of a low-cost stereo (depth) camera as a non-destructive tool for the detection and measurement of cracks in concrete surfaces. The experiment was carried out on four concrete beams with cracks, created with different concrete mixes. The...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Vilnius Gediminas Technical University
2025-01-01
|
| Series: | Geodesy and Cartography |
| Subjects: | |
| Online Access: | https://journals.vilniustech.lt/index.php/GAC/article/view/20152 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | The primary aim of this paper is to assess the effectiveness of a low-cost stereo (depth) camera as a non-destructive tool for the detection and measurement of cracks in concrete surfaces. The experiment was carried out on four concrete beams with cracks, created with different concrete mixes. The mixes of the four beams were made up of lightweight aggregates with 12% of normal weight aggregates. One beam was used as a reference without fibers, while 3D steel fiber reinforcement, 5D steel fibers reinforcement, and a hybrid fibers mix of 5D steel fiber and synthetic were used for the other three beams. The cracks in the beams were measured manually followed by taking their stereo images with a ZED camera. The ZED images were processed to produce 3D models of the concrete surfaces, which are useful for crack measurement in a three-dimensional framework. The project results are particularly significant in the measurement of all three dimensions (length, width and depth), with less than a 10% error between the actual and the experimental procedure. Relatively, multiple differential approaches gave a less accurate result of a 15% error mainly due to syntax errors. Results suggest that the ZED depth camera is an effective tool for robust detection and measurement of cracks in concrete surfaces.
|
|---|---|
| ISSN: | 2029-6991 2029-7009 |