License Plate Extraction Based on Vertical Edge Detection and Mathematical Morphology

License plate recognition plays an important role in intelligent transport system; however, plate region extraction is the key step before the final recognition. In this paper, an effective license plate extraction algorithm based on vertical edge detection and mathematical morphology is proposed, which can exactly extract the plate from complex background. The proposed algorithm mainly consists of two modules: license plate region´s rough detection and license plate accurate detection. The former aims to find out region of interest (ROI) through a sequence of steps including vertical edge detection, edge detection and mathematical morphology, while the latter analyzes the ROI by taking full advantage of outer shape feature and inner texture feature of license plate. Experiments have been conducted for this algorithm. 400 images taken from various scenes were employed, including diverse angles, different lightening conditions, etc. The proposed algorithm can detect the region of license plate quickly with 98.1 percent accuracy.