Monitoring system working method

Abstract

The invention provides a monitoring system working method. The monitoring system working method mainly comprises the following steps: step 1: a monitoring camera sends video information to a snapshotserver, and the snapshot server parse a human face image from the video stream and sends the image and related parameters of the image to a snapshot theme in a distributed issuance subscription message system as snapshot records; step 2: a crowd counting module reads the video information of the monitoring camera, counts the number of people from the video stream, and sends the counted number of people, the camera information and the current time to a crowd counting theme in the distributed issuance subscription message system as crowd counting records; and step 3: a storage module consumes the crowd counting records, and displays corresponding thermodynamic diagram distribution on a display module according to the counting records and a monitoring range corresponding to the camera. Structured extraction is performed on the offline data of customers by using the computer vision technology, the online and offline data are fused by using a human face ID as a bridge, the reconstruction ofpeople, goods and fields is achieved at least, an effective structured data support is provided, and a foundation is laid for the subsequent big data analysis application.

Description

technical field [0001] The invention relates to the field of face recognition, in particular to a working method of a monitoring system. Background technique [0002] It is difficult to count the information of the personnel entering the store [0003] In the Internet environment, one of the core problems that traditional stores have to solve is the control of the store's own data. How many people came to the store today? What is the traffic flow of store A and store B? Comparing traffic and revenue, what is the conversion rate of the two stores? At present, most of the solutions adopt the deployment of top-mounted people counting cameras at the entrances and exits, and identify and judge the people's heads and shoulders. The biggest problem with this solution is that the statistics obtained are only the number of people, and it is impossible to rule out the number of times the clerk enters and exits, the same person enters and exits multiple times, etc., resulting in ina...

AI Technical Summary

Problems solved by technology

The biggest problem with this solution is that the statistics obtained are only the number of people, and it is impossible to rule out the number of times the clerk enters and exits, the same person enters and exits multiple times, etc., resulting in inaccurate data; at the same time, it is impossible to know the age, gender, etc. of the personnel entering the store. attribute data

[0009] Adhere to consumer demand-oriented, customer-centered, and accelerate the return to the essence of commercial retail has become the consensus of the physical retail industry. Consumer demand, consumption habits, and consumption hotspots may vary from place to place, which will inevitably lead to physical retail performance in various places. There are more and more differences, and the differences are prominent, which makes it difficult to implement personalized services

Images