Multiple License Plates Location Using the Optical Flow Algorithm and Blob analysis

Content-based image retrieval (CBIR) techniques have been applied to forensic science databases, such as fingerprints, bullets and cartridge cases of firearms, drug tablet, etc. Similar techniques are also extended to video data. Automatic license plate recognition (ALPR) is one of the most important applications. In Taiwan, ALPR has been equipped in many intelligent surveillance systems. It helps police to check suspected vehicles if they are stolen from other people. One of the core techniques in ALPR is license plate location. We need to find and locate license plates in an image before identifying the plate numbers. Efficient plate location is useful for license plate numbers recognition. A number of methods have been proposed for license plate location. Most literatures focus on detecting the accurate location of single license plate from a vehicle image or video. However, there are usually more than one vehicle appear within an image frame simultaneously in practical cases. That is, we need to locate multiple license plates before identifying their license plate numbers. In this paper, we used the optical flow algorithm and blob analysis to locate the multiple license plates in video sequences. We use video data of surveillance systems from real cases to show the capability of the proposed method. From the experimental results, our method supplies exhilarating capability to locate multiple license plates.