A real-time monitoring system and a monitoring method for the number of people on buses

Abstract

The invention provides a real-time monitoring system and a monitoring method thereof. The monitoring system comprises a video recording module, which is used for acquiring video data of passengers boarding the bus; a voice module for alerting passengers to get off and issuing an overload alarm; a storage unit for storing passenger file data; a detection and recognition unit connected with the video recording module, the voice module and the storage unit through optical fiber, receives the video data and combines the passenger file data, performs image preprocessing, face detection and recognition, population statistics and target tracking, and sends the control information to the voice module at the same time. The monitoring system can automatically detect the number of passengers on and off the bus and count the passenger information on the bus, monitor the number of passengers on the bus in real time, judge whether the bus is overloaded and send out an alarm, improve the safety of the bus, and obtain the travel data of the city residents; the monitoring method avoids the interference of background changes, and is accurate, efficient and real-time.

Description

technical field [0001] The invention relates to a monitoring system, in particular to a real-time monitoring system for the number of people on board a bus and a monitoring method thereof, belonging to the field of artificial intelligence. Background technique [0002] With the growth of population, the pressure of urban traffic has increased sharply, and the problem of congestion during peak hours and overloading of buses has caused potential hidden dangers in urban public safety. Especially in recent years, a large number of energy-saving and environmentally friendly new energy unmanned bus tickets have been put into use, and the driver focuses on driving the car, and no one on the car monitors whether the car is overloaded. As a result, various electronic testing equipment has emerged, such as based on infrared sensor Electronic equipment, and pedestrian detection equipment based on static background in some public places. Since the monitoring video screen in the bus is ...

AI Technical Summary

Problems solved by technology

Among them, the recognition method based on the geometric features of the face is a bottom-up face detection method. The face has characteristics or attributes that do not change due to changes in external conditions, especially in a static state, such as organ shape, skin color, etc. , size ratio, etc., this method is to find the above-mentioned invariable facial features, and use these features to locate the human face, but this type of method is not sensitive to dynamic changes such as facial expression, posture or rotation; The recognition method based on the skin color model is easy to be affected by factors such as light, race, age, etc., especially when the background and skin color are close, it is easy to cause misjudgment and the accuracy is not high; the face detection method based on statistical theory will detect the face The problem is transformed into a statistical pattern recognition problem. The main methods are based on feature space, artificial neural network, support vector machine, hidden Markov model and Boosting method. High accuracy, but low detection efficiency and poor real-time performance

Therefore, the above methods are not applicable to the bus environment

Images