The goal of this project is to recognize motorcycle license plates from images. Given an image containing a motorcycle, the system extracts the license plate, aligns it using image processing techniques, segments individual characters, and finally recognizes them using a classification model.
-
License Plate Detection
- The input is an image containing a motorcycle.
- The license plate region is detected and cropped from the image using YOLOv8.
-
License Plate Alignment
- Hough Transform and Contour detection are used to align and correct the perspective of the license plate.
-
Character Segmentation
- Connected Component Lablelings (CCL) is applied to segment individual characters from the plate.
-
Character Recognition
- A classification model is used to recognize the segmented characters.
