CN102289908A - Image analysis-based independent association real-time anti-theft system - Google Patents

Abstract

The present invention provides an autonomous correlation real-time anti-theft system based on image analysis, which mainly includes the establishment of image data collection and target object feature information collection part, the camera system for comparing the correlation degree of the collected images and the automatic alarm system part, which are connected through the Internet . The invention has relatively high real-time performance, security, anti-theft performance and reliability, and is suitable for hotels, kindergartens and other places.

Description

Image analysis-based autonomous correlation real-time anti-theft system

technical field

The invention relates to an anti-theft system, in particular to an autonomously associated real-time anti-theft system based on image analysis.

Background technique

At present, in some specific public places, such as: museums, movie theaters, Internet cafes, restaurants, hotels, parking lots, children's playgrounds, primary schools or kindergartens, property theft or children being taken away by strangers often occur. In these places, cameras and other similar monitoring devices are often installed, but they cannot avoid the occurrence of incidents in time. They can only call the police afterwards and the public security personnel can solve the case based on on-site video and other materials. In addition, there are some anti-theft measures. Prevent criminal suspects from stealing collections and computer components such as museums and Internet cafes by installing some micro-switches and alarms; set up some jack-like devices in the parking lot to raise the driving part of the parked car to prevent theft of the vehicle; For personal items such as suitcases, travel bags, handbags, etc., set up a pair of small devices with response emission function, and when the object leaves a certain distance from the owner, an alarm will sound; leader. From the above measures, it can be seen that the anti-theft measures in these occasions are passive, rigid and unsafe.

Many public places are equipped with monitoring systems such as cameras and self-sensing infrared sensors. This system can only conduct video surveillance of restricted areas to provide evidence for evidence collection, analysis and processing after the incident. This monitoring system is widely used in museums, Internet cafes, restaurants, hotels, kindergartens and primary schools, public bathing centers and parking lots.

The shortcoming of this monitoring system is mainly passive evidence collection after the event, which cannot effectively prevent the theft of property and cars, and the occurrence of false claims of children. For example, when a customer enters a hotel with a handbag, when the handbag is stolen from the guest room by others, the monitoring system only records the process of the thief entering and leaving the hotel, but fails to immediately and automatically recognize that the thief is carrying the guest’s handbag and leaves and calls the police , this process can only be investigated and dealt with after the incident based on relevant evidence. Therefore, the process of the entire handbag being stolen and the criminal suspect fleeing the scene lacked real-time alarming, resulting in the embarrassment of "monitoring" rather than "control" by the monitoring system. The situation has brought a lot of mental and material losses to customers and businesses.

Most Internet cafes now adopt the following main anti-theft forms:

(1) Closed-circuit monitoring system installed at the entrance and exit of Internet cafes and indoors: it mainly monitors the general situation of people entering and leaving Internet cafes, and can help to grasp the movement of mobile personnel in Internet cafes as a whole.

(2) Internal host monitoring system of Internet cafes: Mainly use modified anti-theft lock chassis or install electronic magnet switches inside the chassis to monitor the host computer conditions in Internet cafes. Each computer is connected to the network management monitoring equipment. Once someone opens the main box or removes the monitor, the network management system will give a voice alarm.

(3) Increase the inspection staff of the network management.

Its main disadvantage is that for the first form, the use of closed-circuit surveillance cameras for monitoring protection, the effect is not only affected by factors such as the working hours, emotional attitudes, and working methods of the monitoring personnel, but also has monitoring dead spots. When the indoor environment is dirty or the lighting Dim monitoring images may not be clear, and Internet cafe theft is bound to happen. Generally, the whole set of closed-circuit monitoring system is also more expensive. Second kind of form is to adopt anti-theft lock and can only play anti-theft effect to general thief, and professional thief can still unlock and steal. The use of magnet induction switches also requires stricter environmental requirements. Once there is electromagnetic interference equipment, it will fail. Therefore, the working stability of the device is also relatively poor, and it cannot function as an alarm. Although the third form increases the number of network management patrol personnel, the security will also be improved, but the theft of computer components in Internet cafes is a small probability event, which brings a waste of human resources and financial resources, and its cost performance is not high. The fundamental purpose of Internet cafes to make money.

Contents of the invention

In view of the defects in the above-mentioned various measures, the present invention proposes a self-associated real-time anti-theft system based on image analysis, which adopts image processing technology to actively identify, analyze and judge the degree of association of the target, and establishes an association model, which includes Protected people, objects related to people, attribute relationships between target objects, target appearance time and specific space, etc., each element in the model is processed to form an interrelated module, and all elements in the module are compulsively composed A collection with a certain degree of association. The system thresholds all associated elements in the set by the comprehensive database, and stores these thresholded data in the host computer for retrieval at any time. For example, objects and their geographic location, people and objects, adults and children, passengers and guest rooms, parking lots and cars, etc. Only all the elements in the collection are judged to be related to each other, and they cannot be too close to other non-related objects such as other people or stay close for too long, otherwise, the system will give an early warning prompt or alarm to the target, and carry out The owner confirms that the protection targets such as property, vehicles and children are in a safe state.

Its technical solution is:

An autonomous correlation real-time anti-theft system based on image analysis, which mainly includes the establishment of image data collection and target object feature information collection, the camera's correlation analysis system for the collected images, and the automatic alarm system, which are connected through the Internet.

Further preferably, the part for establishing image data collection and target object characteristic information collection includes a camera, an audio and video transmitter, an application system host, and a database comprehensive application server.

Further, the camera is a wired or wireless camera.

In one embodiment of the present invention, the camera is a camera integrated with a high-definition camera, a lens and a rotating pan/tilt.

The present invention is an image analysis-based self-associated real-time anti-theft system. The system for comparing and analyzing the correlation degree of images collected by the camera includes a host computer with an anti-theft system function and a networked host computer.

Correlation analysis: it means that in the process of system development, if the change trend of two factors is consistent, that is, the degree of synchronous change is high, it can be considered that the correlation between the two is relatively large; otherwise, the correlation between the two is small. Therefore, correlation analysis provides a quantitative measurement of the development and change of a system, which is very suitable for dynamic process analysis.

The 'correlation analysis' in the present invention is not quite consistent with the existing concept of correlation analysis. The existing concept of correlation analysis mostly means that if there is a certain correlation between two or more things, one of them can pass through other things. Prediction, for example, the stock market can use the concept of correlation analysis to predict the general trend of the stock market, the proper placement of supermarket products can increase the sales of shopping malls, and the use of correlation analysis in data mining can classify and cluster common classes. Data trends and trends. The degree of association in the present invention is a mandatory association relationship. Individuals who are too far apart are not within the scope of protection of the system. For example, if two objects appear in the same system, the system will forcibly associate these two objects together. If the distance between the two targets exceeds the threshold range stipulated by the system, this correlation does not exist. The system independently analyzes and compares the two correlation degrees before and after entering the system range, and the system finds two correlation degree analysis results If it is damaged, the system will trigger the alarm system for early warning or alarm after judgment. The second meaning of the correlation degree analysis of the system of the present invention is that, for example, a person is associated with the items he carries to form a correlation degree set, and the person has nothing to do with the type and attribute of the specific item. The person can carry any item, as long as the person and the item If it falls into the protection scope of the system, the system will forcibly associate the two objects together. Those that are not within the stated scope are not within the protection scope of the system. For example, if a guest enters a hotel with a handbag, then the guest is associated with the bag. If the guest enters the hotel to pick up a child, then the guest is associated with the child. This guest has nothing to do with anything outside the system.

Further preferably, the networked hosts are multiple hosts located in different geographic locations.

The present invention is an image analysis-based self-associated real-time anti-theft system, wherein the automatic alarm system is an alarm, a user's mobile phone, a telephone, and other hosts.

Further preferably, the alarm is wired or wireless.

Compared with the prior art, the key technology of the present invention is to process the collected target image data and pictures in real time, and the data elements processed by the system have the following properties:

(1) All elements in the set have mandatory association;

(2) All elements in the set are thresholded;

(3) When the correlation of any element in the set is destroyed, the system will trigger an alarm warning prompt or alarm after the system judges.

Establish a collection of association thresholds for the data processed by the system and store them in the host comprehensive database. The comprehensive application database performs related operations such as loading, editing, retrieval, updating, classification and reorganization of the stored data. When the system obtains the target association threshold again Just perform target matching, which can achieve real-time processing. When the alarm condition is reached, the alarm will be called in time. The alarm process of this system is very short, and has relatively high real-time performance, safety, anti-theft, and reliability. The present invention can realize effective anti-theft without adding additional hardware resources, and has the advantages of saving resources, saving costs, and improving the efficiency of solving crimes, etc. advantage.

Description of drawings

The technical solution of the present invention will be further described below in conjunction with the accompanying drawings and embodiments.

Fig. 1 is a structural diagram of the present invention's self-associated real-time anti-theft system based on image analysis;

Fig. 2 and Fig. 3 are the flowcharts of the self-associated real-time anti-theft system based on image analysis of the present invention.

Detailed ways

Referring to Figure 1, Figure 2, and Figure 3, an image analysis-based self-associated real-time anti-theft system mainly includes the establishment of image data collection and target object feature information collection, a system for comparing and analyzing the image collected by the camera, and an automatic alarm The system part is connected through the Internet.

(1) Establish image data collection and target object feature information collection part

The main content of this part is: the camera probe monitors the real-time status of information points in real time (multiple camera devices can be set up through networking in different environments), and performs real-time camera and picture capture of the target objects that appear. At the same time, the system collects other relevant information of the target object. The system processes the collected image data and feature set elements of the target object in real time. The main operations include loading, editing, retrieval, updating, classification and reorganization of the set elements, and builds a set of correlations and stores it in the comprehensive database. The implementation form is shown in the following example:

Example 1 The monitoring system constitutes a system for the guests entering the hotel and the related items they carry with them. The camera collects the characteristics of the guests, the characteristics of the items related to the guests, the time and place of the guests, and at the same time, the identity and contact information of the guests are registered. , the number of people in a group, the room number and other information.

Example 2 Install this system at the gate of the kindergarten or at a specific pick-up location for children. The system collects images of children who appear for the first time and their parents, and enters other information of the target object to establish a set of specific correlations. This collection includes the basic characteristics of the child and the parent, specifically the appearance characteristics, relationship attributes, name, age, address, and contact information of the child and the parent.

(2) The camera performs a correlation analysis system on the collected images

The camera performs real-time correlation degree threshold analysis on the collected image data, and compares the correlation degree with the relevant data entered in the database for the first time. When it does not match, it will judge and send an alarm signal.

For example, the guest who enters the hotel for the first time and the bag he carries with him (or the guest enters the hotel without a bag), the system has bundled them into a mandatory association set, and the set elements include the guest's appearance characteristics, ID card, room number, Information such as contact information is associated with the bag. When it is found that the bag has been taken away by someone else (or the guest is carrying someone else’s bag), and the monitoring system finds that this illegal relationship does not match the preset relationship, the anti-theft The system automatically analyzes and judges and alarms.

(3) Automatic alarm system part

When the automatic alarm controller receives the alarm command, the on-site alarm can forward the information to other MTSW (alarm receiving center software) management computer or user mobile phone through Modem, network, GSM and other equipment at the same time to achieve the purpose of joint defense .

Further preferably, the part for establishing image data collection and target object characteristic information collection includes a camera, an audio and video transmitter, an application system host, and a database comprehensive application server.

The audio and video transmitter transmits the picture or video data collected by the camera to the comprehensive application system of the database. The use of this transmitter can ensure that the image and picture data collected by the camera are clearer and undistorted after being transmitted through the cable. It has the advantages of strong anti-interference ability, long transmission distance, easy wiring, and low price.

Database comprehensive application system, the system can establish, use and maintain the database, manage and control the database in a unified way, so as to ensure the security, reliability, real-time and integrity of the database.

The database integrated application server can realize multi-users and multi-application programs to create, modify and query databases in parallel, and provide users with integrated database services.

In one embodiment of the present invention, the camera is a wired or wireless camera.

Further preferably, the camera is a camera integrated with a high-definition camera, a lens and a rotating pan/tilt.

The present invention is based on an image analysis-based self-associated real-time anti-theft system. The system for comparing and analyzing the correlation degree of images collected by the camera includes a host with an anti-theft system function and a networked host.

The 'correlation analysis' in the present invention is not quite consistent with the existing concept of correlation analysis. The existing concept of correlation analysis mostly means that if there is a certain correlation between two or more things, one of them can pass through other things. Prediction, for example, the stock market can use the concept of correlation analysis to predict the general trend of the stock market, the proper placement of supermarket products can increase the sales of shopping malls, and the use of correlation analysis in data mining can classify and cluster common classes. Data trends and trends. The degree of association in the present invention is a mandatory association relationship. Individuals who are too far apart are not within the scope of protection of the system. For example, if two objects appear in the same system, the system will forcibly associate these two objects together. If the distance between the two targets exceeds the threshold range stipulated by the system, this correlation does not exist. The system independently analyzes and compares the two correlation degrees before and after entering the system range, and the system finds two correlation degree analysis results If it is damaged, the system will trigger the alarm system for early warning or alarm after judgment. The second meaning of the correlation degree analysis of the system of the present invention is that, for example, a person is associated with the items he carries to form a correlation degree set, and the person has nothing to do with the type and attribute of the specific item. The person can carry any item, as long as the person and the item If it falls into the protection scope of the system, the system will forcibly associate the two objects together. Those that are not within the stated scope are not within the protection scope of the system. For example, if a guest enters a hotel with a handbag, then the guest is associated with the bag. If the guest enters the hotel to pick up a child, then the guest is associated with the child. This guest has nothing to do with anything outside the system.

The host with the anti-theft system function has the function of independent correlation analysis of processing images and image data. This function is a key step to independently analyze and process images and picture data collected by cameras to establish the correlation degree of target objects. Elementize, threshold, and establish an association degree set for the target object of mandatory association, so that only all elements in the collection have an association degree relationship. When the association degree of the elements in the real-time analysis set is destroyed, when the alarm warning or alarm condition is reached , then send relevant alarm information to other hosts, alarms, and user mobile phones through networking. The information can be forwarded to other alarm reception centers through Modem, network, GSM and other equipment.

In one embodiment of the present invention, the networked hosts are multiple hosts located in different geographic locations.

The present invention is an autonomous associated real-time anti-theft system based on image analysis, and the automatic alarm system is an alarm, a user's mobile phone, a telephone, and other hosts.

Further preferably, the alarm is wired or wireless.

The working process of the self-associated real-time anti-theft system based on image analysis in the present invention:

(1) The image and picture data collected by the camera in real time are sent to the audio and video transmitter for processing. The processed data can achieve low distortion, data restoration, and reduce distortion and interference during data transmission. The audio and video transmitter transmits the processed image and picture data to the database comprehensive application server for processing and storage.

(2) Hosts equipped with anti-theft system functions and networked hosts, apply system-related functions to process the collected data, including independent correlation analysis of data, thresholding associated data, establishing associated element sets, and People, objects, time, specific space and other related elements are combined into a set of elements protected by the system and stored in the comprehensive application system of the database.

(3) When the corresponding association threshold comparison analysis relationship of the target data object processed by the system is destroyed, the system will issue a real-time early warning prompt or alarm after the system judges, and the system will carry out real-time monitoring of arming at ordinary times.

(4) The host computer conducts real-time analysis on the collected data. When the alarm condition is met, the host computer sends an early warning or alarm command to the alarm device, and at the same time, the host computer can forward the information to other alarm receiving centers for management through Modem, network, GSM and other equipment. Computer or user mobile phone, in order to achieve the purpose of joint defense.

The embodiments of the present invention are not limited thereto, and any equivalent replacement of devices with the same function may fall within its protection scope.

Claims (8)

1. real-time burglary-resisting system of autonomous association based on graphical analysis, it is characterized in that, mainly comprise set up image data acquiring and destination object characteristic information set part, camera carries out degree of association comparative analysis system and automatic alarm system part to the image of gathering, be connected by the internet.

2. the real-time burglary-resisting system of autonomous association based on graphical analysis according to claim 1, it is characterized in that described image data acquiring and the destination object characteristic information set part of setting up comprises camera, audio frequency and video transmitter, application system main frame, database integrated application server.

3. the real-time burglary-resisting system of autonomous association based on graphical analysis according to claim 2 is characterized in that the wired or wireless camera of described camera.

4. the real-time burglary-resisting system of autonomous association based on graphical analysis according to claim 2 is characterized in that described camera is to have the integrated camera of high-definition camera, camera lens and rotary head.

5. the real-time burglary-resisting system of autonomous association based on graphical analysis according to claim 1 is characterized in that, described camera carries out degree of association comparative analysis system to the image of gathering and comprises having burglary-resisting system function host and networked hosts.

6. the real-time burglary-resisting system of autonomous association based on graphical analysis according to claim 5 is characterized in that described networked hosts is the multiple host that is in diverse geographic location.

7. the real-time burglary-resisting system of autonomous association based on graphical analysis according to claim 1 is characterized in that described automatic alarm system is alarm, user mobile phone, phone, other main frames.

8. the real-time burglary-resisting system of autonomous association based on graphical analysis according to claim 7 is characterized in that described alarm is wired or wireless.

Images