In this step, we have to detect all the plates in a current camera frame. To do this task, we divide it in to two main steps: segmentation and segment classification. The feature step is not explained because we use the image patch as a vector feature.
In the first step (segmentation), we will apply different filters, morphological operations, contour algorithms, and validations to retrieve parts of the image that could contain a plate.
In the second step (classification), we will apply an SVM classifier to each image patch, our feature. Before creating our main application, we will train with two different classes: plate and non-plate. We will work with parallel frontal view color images with 800 pixels of width and that are taken between two and four meters from a car. These requirements are important for correct segmentation. We can perform detection if we create...