There are several different ways that we can implement object detection using the OpenMV camera. In this section, we are going to use two different methods:
- First, we are going to use a pretrained convolutional neural network (CNN) that was trained by OpenMV using the CIFAR-10 dataset. We will load the example and then provide it with several images to see how the network behaves and to become familiar with OpenMV operations.
- Second, we will move beyond the pretrained network and train a network of our own that we will then deploy to the OpenMV camera. By the time you have completed this section, you will be able to train a network to detect whatever object it is that you are interested in detecting and be able to start building your own control application.
