Non destructive evaluation of mixed concrete performance using image processing and machine learning

The study prepared HC samples with different mix ratios (1:3:6, 1:2:4, 1:1.

Embedment Plate

5:3), water cement ratios (0.

4-0.

6), and CRM substitution rates (0%, 15%, 25%) using 162 cylindrical specimens (diameter 150mm, height 300mm).

Horizontal and vertical cutting surface images were collected, and statistical features were extracted to train SVM, BRE, and GPR models.

The overall process is shown in Figure 1.

.

Leave a Comment