Face Recognition System For Border Inspection In Shanhaiguan Shipyard

Border Inspection is a kind of inspection which aims at safeguarding the sovereignty and security of a country. The targets of Border Inspection include: people’s passports, visas, Exit & Entry registration cards, baggage and property. The duties of Border Inspection include: checking transport vehicles and goods as well as handling personnel who violate the rules and regulations of Exit & Entry administration. Different from the land border inspection and the airport one, the harbor border inspection has unique characteristics such as long-stay transport vehicles, complex but regular people who enter and exit the restricted areas(ship-owners, crew, passengers or workers), bigger flow rate than average. While exercising the powers, the harbor border inspection authorities are supposed to prevent harms to national security and social order, to deal with hostiles, terrorists and criminals who enter and exit the restricted areas, to prevent and combat with illegal entry and exit behaviors and other criminal activities, to prevent the entry and exit transport vessels from carrying criminal, terrorists, weapons and ammunition. To accomplish the tasks above, we must ensure that the access to the restricted area and the board are kept in control. Therefore, we introduce the face recognition system to strengthen border control on the restricted areas and the management of the personnel. In this paper, a face recognition system is designed which is based on the background of the personnel access management in Shanhaiguan shipyard.First of all, we make a general technical analysis on face recognition system. According to the practical tasks of border inspection in Shanhaiguan shipyard, we select the approach of face detection, face feature extraction and face recognition method which is based on the algorithm of feature recognition and the method of video recognition.Then, a detailed necessity analysis is made on the border inspection in Shanhaiguan shipyard. According to the actual situation, the functional design of the face recognition system is completed, including the functions of face recognition, personnel classification, information early warning, permission settings and so on. The principal component analysis(PCA) method is used to construct the face recognition algorithm. The hardware necessity is also analyzed and designed.Secondly, the overall system is designed, including personnel information collection module, face detection module, face recognition module, face retrieval module, information management module. As a result, the integration and erection of equipment is completed so that the management of personnel access to the restricted area is achieved.Finally, a combination classifier is used to verify the weighted distance classifier and neural network classifier, and a test on face detection and location, personnel identification check is carried out. The results of the experiments show that the system is stable and the algorithm and design are verified by the experimental results.