Event driven video tracking system

The first event-driven video tracking system Technical Field The present invention relates to video tracking of moving targets, and in particular to video tracking initiated by event triggers, in which targets are tracked through multiple visible areas.

BACKGROUND OF THE INVENTION Such systems are disclosed in US Patent No. 5,70 & 423, which can track the position of the person or object by connecting a transponder to a person or object and receiving an identification signal transmitted from the transponder. The commonly used technology to accomplish this tracking is called radio frequency identification (RFID). Generally, each transponder transmits a unique identification signal, which is used to identify the person or object connected to the transponder. Generally speaking, the identification signal generated by the transponder in this system is received by the antenna and connected with the receiving circuit. In some cases, the antenna is associated with the door or entrance that allows entry into the room. It is usually also considered to determine the direction of the tracked object relative to the entrance. The building of the tracked item can contain several controlled areas and uncontrolled areas. It is possible to allow items to enter certain areas, but not to enter other areas. Alarms are usually used to warn individuals of items being moved into or out of a specific controlled area.

Controlled areas and associated entrances can also be under video surveillance. Alarms are usually used to trigger video recording equipment to record related areas. As disclosed in US Patent No. 6,02&626, a video surveillance system can determine the speed, direction, and acceleration of objects within the video coverage area.

When certain events occur, some problems arise in the current technology. When a trigger event occurs, for example, when an item with an RFID tag attached is transported through the RFID monitoring entrance, the video surveillance of the entrance and/or the associated area can be started. When the item is triggered to move into a different area, if the area may not contain an RFID antenna, the video surveillance in the area is not triggered, thereby losing the video surveillance of the item. What we need is video surveillance of specific assets from region to region. In another example, only a certain person is authorized to remove items from the controlled area. Without human intervention, if the item is removed from the area, it can be triggered

200380104849.3 report to the police, even if the individual who removed the item has the proper authorization to remove the item. Not only do we need to associate the asset with the region, but we also need to associate it with a specific individual.

SUMMARY OF THE INVENTION The present invention provides a system for tracking assets from area to area in a controlled location. The tracking operation is initiated by a trigger event. The system can also use electronic methods to identify assets, and selectively record and display assets and logos.

The event-driven video tracking system and method include sensors that detect preselected events. The pre-selected event is located in the first area. The video coverage of the first video surveillance camera determines the first area. The first video surveillance camera is responsive to a sensor that activates the first video surveillance camera in response to a preselected event. The first video surveillance camera monitors the images associated with the preselected event and tracks the images in the first area. The second video surveillance camera has a video coverage area that determines the second area. There is an overlap between the first area and the second area. The second video surveillance camera captures the image in the overlapping part and tracks the image in the second area.

The system may have an asset database, so that the pre-selected event is related to the pre-selected information, and the pre-selected information is related to the pre-selected event.

The preselected information may include the identification of the asset and the area authorized to find the asset.

The preselected event may include reading the RFID tag.

The system can include a video recorder and/or a video display.

The system can have multiple sensors and multiple video surveillance cameras that define multiple areas.

A prediction algorithm that predicts the speed and direction of the image can be used to track the image.

When capturing an image in the overlapped part, it can be used to translate the image coordinates from the first area to the second area.

Through the following detailed description of the embodiments of the present invention, the purpose, advantages and applications of the present invention are obvious.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a block diagram of an embodiment of the present invention.

Fig. 2 is a typical block diagram of handover assets in the visible area of the present invention.

Detailed ways

200380104849.3 Referring to Figure 1, the video controller 2 controls each of the "n" video surveillance areas 4, where "11" is greater than or equal to (3) 2. Each video surveillance area 4 can have a fixed focal length, a variable focal length, a pan and tilt, a wide angle, or other cameras controlled by the video controller 2. The video controller 2 also controls a video recorder 6 and a video display 8. The video recorder 6 can be one or more analog or digital video recorders, and the video display 8 can be one or more video displays. The event-driven tracking controller 10 guides the video controller 2 how to control the video cameras in the video surveillance area 4 to track assets. The event-driven tracking controller 10 receives input signals from the area sensor 12. The area sensor 12 uses one or more sensors "m" to monitor selected areas called areas, where the sensor "m" can be an RFID tag, an infrared sensor, a motion sensor, a light sensor, a sound sensor, or other sensors. The output can be used to trigger the event-driven tracking controller 10. The event-driven tracking controller 10 communicates with an asset database 14. The asset database 14 contains all asset tables under the control of the system, and may contain authorized areas for each asset. If the asset leaves its contact area, video surveillance is triggered. The asset database 14 may contain other rules required for asset control. An example of a rule is that if a specific individual enters a controlled area, video tracking is triggered. Another rule is that asset number X is not allowed to leave controlled area Y unless it is accompanied by one of A, B or C, where A, B and C are individuals authorized to remove asset X from area Y. If the rules are violated, video tracking is triggered. Other related actions can be initiated, such as alarms, automatic calls to authorized individuals, and other required actions such as video tracking can be triggered. The choice of rules for specific applications is flexible and can be implemented through programming.

After the trigger input from the area sensor 12 to the event-driven tracking controller 10, the event-driven tracking controller 10 guides the video controller 2 to start tracking the asset in the appropriate video surveillance area 4, and if the trigger event contains an identifier, the asset is identified. The trigger event can be selected as needed, and it can be read from the RFID that contains the asset identification in the relevant area, or it can be other sensor input that does not contain the identification. Once the video controller starts to track the asset, it can store and display the asset's identification and video images in the video recorder 6 and the video display 8, respectively. If multiple trigger events occur, multiple assets can be tracked and identified.

Referring to Figure 2, we describe an embodiment in detail to explain the present invention. Sensor "m" is one or more RFID tags 20, from area sensor 12 to event-driven tracking control

200380104849.3 The input of the first device 10 is the output of the RFID reader 22, and the reader 22 can be placed near the entrance of the video surveillance area 1. The video surveillance area 1 and the video surveillance area 2 are under the surveillance of the video camera 24 and the video camera 26 respectively, and these two cameras are controlled by the video controller 2. After entering the area 1, the RFID reader 22 reads the RFID tag 20. The RFID reader 22 communicates with the event-driven tracking controller 10, as shown by the area sensor 12 in FIG. 1, and transmits asset-related information to the connected, adhered, or associated RFID tag 20. The event-driven tracking controller 10 communicates with the asset database 14 and associates the information received from the RFID tag 20 with relevant stored information.

As mentioned above, the asset database 14 may contain rules for determining trigger events. Triggering events are events that require video tracking, as well as recording and/or displaying assets. Let us assume that the information obtained from the RFID tag 20 and transmitted to the event-driven tracking controller 10 is a trigger event, and the event-driven tracking controller 10 commands the video controller 2 to start tracking the trigger asset. At this time, the location of the asset is known because it is adjacent to the RFID reader 22, and therefore, the tracking video camera 24 can easily capture the image of the asset. Now, the asset is under video surveillance and is identified based on the information obtained from the RFID tag 20 and/or the asset database 14. Assets can be RFID-tagged individuals or RFID-tagged objects, which are concealed, connected, adhered, or connected in some way. When the asset moves in the area 1, the camera 24 uses an electronic and/or mechanical method to track the asset under the control of the video controller 2, which is related to the camera used. When the RFID tag 20 moves into the area 1+2, the video controller 2 activates the camera 26 and passes along the image position associated with the RFID tag 20, which is learned from the camera 24. There may be many ways to pass the position of the RFID tag 20, for example, establishing a coordinate system associated with each camera and translating the coordinates from the coordinate system of the camera 24 to the coordinate system of the camera 26. When the RFID tag 20 moves When entering area 2, the camera 26 continues to monitor, but the camera 24 loses surveillance. The increase in the number of areas and cameras is virtually unlimited.

A combination of trigger sensors "m" can be used. For example, in some areas, only motion sensors can be used, and assets are tracked, but they may not be recognized. When tracking assets, it can be moved into areas with RFID sensors. Once the RFID reader 22 communicates with the controller 10, the asset is identified. Through flexible programming of video controller 2 and event controller 10,

200380104849.3 Section can easily implement many other changes of sensors and areas.

As shown in Fig. 1, in order to facilitate asset tracking, the video controller 2 can utilize a predictive algorithm 16 to determine useful asset tracking characteristics. Useful tracking features include: image speed, direction, acceleration, and other related parameters. The optical flow calculation and other known techniques can be used in the prediction algorithm 16. When the surveillance camera used uses electronic or mechanical means to complete the pan, tilt, and zoom functions, the use of the predictive algorithm 16 is particularly beneficial.

It should be understood that various changes and modifications can be made to the present invention without departing from the scope of the present invention. The example given in Figure 2 is only an embodiment, and we can change the specific implementation details. For example, a single controller or computer, or multiple controllers or computers can be used to accomplish the above-mentioned functions. It should also be understood that the scope of the present invention should not be construed as being limited by the specific embodiments disclosed herein. After reading the above disclosure, the scope of the present invention is only limited by the appended claims.

200380104849.3