Development of a software for automated visual and measuring control of steel ropes based on computer vision and deep learning technologies

Hello there,

I take up researching work "Development of a software and hardware complex for automated visual and measuring control of steel ropes based on computer vision and deep learning technologies". It is necessary to solve a number of issues to detect defects on steel ropes and measure their metric characteristics:

1) Detection of external broken wires, counting their number at a fixed length of 6d for comparison with critical values, measuring the distance between the wires.

2) Detection of surface wear of the rope, measurement of the length and diameter of the detected surface wear.

3) Detection of surface corrosion of the rope.

4) Measuring the diameter of the rope at a fixed length of the rope 6d.

5) Detecting and counting loop formations of the rope.

6) Determination of the waveness by measuring the diameter of the waveness and the step of the waveness relative to the axes drawn.

7) Detection of birdcage defect.

8) Rope kinks detection.

9) Detection of a burn defect due to temperature exposure.

10) Detection thinning of the ropes.

The biggest issues we have is the algorithm to measure diameter, waveness and surface wear of ropes. We're using three steel ropes that's why we're going to develop the software for damages detection on each steel ropes at the picture. You can find some pictures of steel rope's defects below:

What should I get started? What MATLAB tools can be effective for solving those issues? Please, help me with algorithms of damages detection.

I am looking forward to a reply.

Yours faithfully,

Darya Yakovleva

South Russian State Polytechnic University by Platov