Method for evaluating queuing time based on vision

Abstract

The invention relates to a method for evaluating the queuing time based on vision, which has the advantages that the cost is low (the hardware cost of a queuing time evaluation device is less than 1000 yuan); the device can be fixed and can be carried, thereby being suitable for various fixed or temporary locations; and under the condition of a plurality of passages, a plurality of equipment can be accessed to a management PC 204 through networks so as to automatically identify the efficiency of different queues and lead users to be distributed evenly in each queue, thereby effectively enhancing the user satisfaction and improving the queue efficiency. The method is characterized in that a camera is arranged above the queuing entrance for detecting people in the queues and carrying out video tracking on a plurality of target people, and the queuing time evaluation device counts the number of people before the target people in a definite period of time and performs weighted average on the target people to obtain the average forward speed of the queues, thereby evaluating the approximate time needed to wait according to the number of people before the users.

Description

technical field [0001] The invention relates to a vision-based queuing time evaluation method, belonging to the technical fields of image recognition and pattern recognition. technical background [0002] Occasions that require queuing in daily life are very common, such as airport registration and security checks, supermarket cash registers, fast food restaurants, competitions, concerts, large-scale exhibitions (World Expo, exhibitions), hospitals, museums, science and technology museums, ticket offices, etc., but due to lack of The evaluation and prediction of the queuing time will bring a lot of inconvenience to both parties. For the sponsor, it is difficult to effectively improve behaviors such as cashiering, ticket sales or ticket inspection, causing dissatisfaction among users; for users, it is impossible to evaluate the waiting time and effectively arrange their own behavior. [0003] The existing queuing machine technology can effectively improve the queuing experie...

AI Technical Summary

Problems solved by technology

[0002] Occasions that require queuing in daily life are very common, such as airport registration and security checks, supermarket cash registers, fast food restaurants, competitions, concerts, large-scale exhibitions (World Expo, exhibitions), hospitals, museums, science and technology museums, ticket offices, etc., but due to lack of The evaluation and prediction of queuing time will bring a lot of inconvenience to both parties

For the organizer, it is difficult to effectively improve behaviors such as cash register, ticket sales or ticket inspection, which causes dissatisfaction among users; for users, it is impossible to evaluate the waiting time and effectively arrange their own behavior

[0003] The existing queuing machine technology can effectively improve the queuing experience, but there are still several defects: first, it is not suitable for large-scale queuing, such as train ticket sales, exhibitions, and concerts, and users are unwilling to queue up in order to get the queuing number; Second, the waiting time cannot be assessed

Images