You are here: Start » Tutorial Exercises » Gray-based Template Matching: Basic (gbtm_basic)

Gray-based Template Matching: Basic (gbtm_basic)

Aim

Your task is to create a simple algorithm which will be able to detect a marker on a printed circuit board.

Input

Images of a printed circuit board at different rotations.

The input image is stored in gbtm_basic directory.

Output

An application that detects the marker.

Hints

This exercise shows the way to find an object in an image using Gray-based Template Matching. To use GbTM please use the LocateSingleObject_NCC filter.

Solution (AVS)

  1. Add the EnumerateImages filter to load consecutive images from a disk directory.

  2. Add the LocateSingleObject_NCC filter and connect it with the input image.

  3. Open a template matching editor on inGrayModel and set model parameters as on the image below.

  4. In the next tab mark object like an image below.

  5. Add the outImage and outObject.Match outputs to a data preview.

  6. Perform several iterations of the program and verify if the results are correct.

  7. If the object is not always found, this is most likely due to too high value of the inMaxPyramidLevel input (try 3 and also see what is on the diagImagePyramid diagnostic output). Another solution might be to reduce the value of the inMinScore input (try 0.6).