June 2019
Intermediate to advanced
348 pages
8h 28m
English
In this example, we are going to train a cascade classifier that will be used to detect a traffic sign, namely, the no-entry sign:

The first thing to do is to prepare the negative samples—the background images. As we mentioned, the negative samples can be arbitrary images that don't contain the objects of interest, so we can collect some images for that purpose easily. When these images are collected, we put their paths in a text file. In that file, the paths are listed in one path per line format, and the path can be either absolute or relative.
Read now
Unlock full access