Detecting License Plates from Videos Using Morphological Operations and Adaboosting Algorithm

Abstract

This paper presents a novel hybrid detection method for extracting license plates from cluttered images using morphological operations and Adaboosting algorithm. First, the hybrid method uses the Adaboosting algorithm for training a detector to detect license plates whose embedded characters are horizontally aligned. Then, the method uses a morphology-based scheme to detect inclined license plates. The morphology-based scheme extracts important contrast features to search possible license plate candidates. The contrast feature is robust to lighting changes and invariant to different transformations. The hybrid method can avoid the significant growth of training samples for training the detector to detect any oriented license plates. The proposed detection technique can locate multiple plates with different orientations and even with very low intensities. Experimental results show that the proposed method improves the state-of-the-art work in terms of effectiveness and robustness for license plate detection.