Content area
Full Text
Abstract: With the rapid development of artificial intelligence in recent years, the application of intelligent security has become increasingly widespread. This paper presents a new intelligent system that uses Convolutional Neural Network (CNN) combined with a high-resolution camera to identify the theft behavior of customers. The CNN extracts relevant information from the theft and non-theft behavior of customers in supermarkets to establish a recognition model. Our results show that, by updating the data sets, the recognition model can be continuously optimized, and the average recognition accuracy finally reaches 83%. The proposed system can independently identify the theft and non-theft behavior in video surveillance and sound alarm on the theft behavior in time. The advantages of the system are its low cost and high precision, which show excellent commercial value and application prospects.
Key words: artificial intelligence, convolutional neural network, information extraction, sound alarm
ACM CCS:
(1) Computing methodologies-Modeling and simulation-Model development and analysis-Model verification and validation
(2) Computing methodologies-Artificial intelligence-Computer vision-Image and video acquisition
(3) Computing methodologies-Machine learning-Machine learning approach
(4) Computer systems organization-Real-time systems-Real-time operating systems
(5) Mathematics of computing-Probability and statistics
1.Introduction
With a continued prosperity of the national economy and an increasing demand for household goods, the supermarket has become a core place for residents to purchase daily necessities. A series of surveys show that the primary business model of modern supermarkets is independent shopping [1]. This shopping mode inevitably brings some management problems, the most serious of which is the occurrence of the theft in supermarkets. The Global Retail Theft Barometer released a survey of 222 retailers in 24 countries. The survey showed that global retailers lost $ 128 billion in 2013, of which US retailers accounted for $ 42 billion. The most significant cause of the losses comes from the theft [2,3]. Therefore, solving the theft problem becomes one of the essential things during the supermarket operation. At present, most supermarkets adopt the traditional surveillance mode, which monitors the video surveillance terminal through the human eye. It has many limitations, such as extended response time and a high error rate. Some supermarkets have an acoustic and magnetic detection system at entrance and exit [4], but none of them can effectively prevent the theft.
The development of intelligent security technology...