Video-Based Vehicle Counting for Expressway: A Novel Approach Based on Vehicle Detection and Correlation-Matched Tracking Using Image Data from PTZ Cameras

Figure 4

Framework of SSD model (512 × 512) [13]. As is shown, the red box in the figure contains the structure of our SSD 512 × 512 model. If an image is input, it will be resized to 512 × 512. Moreover, VGG-16 is still employed as the base network for the SSD model to extract features.