System and method for the cooperation of a fixed camera and an unmanned mobile device to improve the identification reliability of an object

Abstract

Disclosed is a process for collaborating between a fixed camera and an unmanned mobile device to enhance identification of an object of interest. A first point of view (POV) of a detected object is obtained, and it is determined, with a first level of confidence, that the detected first POV of the object matches a stored object of interest. Then, one or more camera-equipped unmanned mobile vehicles are identified in a particular direction of travel of the first detected object, and dispatch instructions and intercept information are then transmitted to the one or more camera-equipped unmanned mobile vehicles. Subsequently, a detected second POV of the first detected object is received via the one or more camera-equipped unmanned mobile vehicles.The captured second POV of the detected object is used to determine with a second level of confidence that the detected object matches the stored object of interest.

Description

BACKGROUND OF THE INVENTION

[0001] As the number of available and accessible fixed cameras in urban and suburban areas increases, so too does the possibility and ability to identify objects against objects of interest in real time. Using digital image processing, certain objects can be identified against objects of interest with an associated level of certainty or confidence. In situations where the level of certainty that an imaged object matches an object of interest is relatively high, for example, 90 to 100 percent, it becomes relatively easy to make related decisions based on this match. Similarly, in situations where the level of certainty is relatively low, for example, 0 to 10 percent, it becomes relatively easy to foresee related decisions based on the lack of match.However, there is a range between relatively high and relatively low security levels, for example, between ten and ninety percent or twenty and eighty percent security, where it becomes more difficult to make relevant decisions based on such an intermediate security level, namely that an imaged object matches an object of interest. A method for object identification and tracking, involving object recognition in a camera image with an associated security level and a determination of the direction of movement, which is also used to recognize the object in further camera images, is already known from the prior art in the form of "XU, Yuanlu [et al.]: Multi-view people tracking via hierarchical trajectory composition. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 27-30 June 2016, pp. 4256-4265. - ISBN 978-1-4673-8852-8".Furthermore, methods are known from US 9 237 307 B1 and "JIANG, Wenbin [et al.]: Vehicle tracking with non-overlapping views for multi-camera surveillance system. In: Proceedings / IEEE international conference on high performance computing and communications & IEEE international conference on embedded and ubiquitous computing, 13-15 November 2013, pp. 1213-1220. - ISBN 978-0-7695-5088-6" by which it is possible to switch between different views from different cameras in order to track an object.

[0002] Therefore, there is a need for an improved technical procedure, device and system to improve the security level of object detection of objects recorded by fixed cameras through intelligent transmission and coordination between the fixed cameras and one or more camera-equipped unmanned mobile vehicles. BRIEF DESCRIPTION OF THE MULTIPLE VIEWS OF THE DRAWINGS

[0003] The accompanying illustrations, in which the same reference numerals refer to identical or functionally similar elements in the individual views, are included in the description together with the following detailed description and form part of it, serving to further illustrate embodiments and concepts that include the claimed invention and explain various principles and advantages of these embodiments. Fig. Figure 1 is a system sketch which provides an exemplary operating environment for improving the object recognition security of objects imaged by fixed cameras through intelligent dispatch and coordination between fixed cameras and camera-equipped unmanned mobile vehicles, according to one embodiment. Fig. Figure 2 is a plan representation of a geographical area that schematically illustrates the positioning and coordination of a stationary camera and a camera-equipped unmanned mobile vehicle, according to one embodiment. Fig. Figure 3 is a device sketch showing a device structure of a data processing device for improving the object recognition security of objects of interest imaged by fixed cameras, by intelligent transmission and coordination between fixed cameras and camera-equipped unmanned mobile vehicles, according to one embodiment. Fig. Figure 4 illustrates a flowchart showing the process steps for operating a data processing device. Fig. 3 represents an embodiment to improve the object recognition security of objects imaged by fixed cameras through intelligent dispatch and coordination between fixed cameras and camera-equipped unmanned mobile vehicles. Fig. Figure 5 illustrates a flowchart depicting process steps for operating a distributed system of fixed cameras and camera-equipped unmanned mobile vehicles to improve the object recognition security of objects imaged by fixed cameras, according to one embodiment. Fig. Figure 6 is a sketch illustrating several different possible viewpoints for capturing and comparing a person's face as a captured object for comparison with the face of another captured person as an object of interest, according to one embodiment.

[0004] Experts will recognize that elements in the figures are illustrated for the sake of simplicity and clarity and are not necessarily drawn to scale. For example, the dimensions of some of the elements in the figures may be exaggerated compared to other elements to help improve the understanding of embodiments of the present invention.

[0005] Where appropriate, the apparatus and process components have been represented by conventional symbols in the drawings, showing only those specific details essential for understanding the embodiments of the present invention, so as not to obscure the disclosure with details that are readily apparent to those skilled in the art who benefit from this description. DETAILED DESCRIPTION OF THE INVENTION

[0006] Disclosed is an improved technical method, device and system for improving the object recognition security of objects imaged by fixed cameras through intelligent dispatch and coordination between fixed cameras and identified camera-equipped unmanned mobile vehicles.

[0007] In one embodiment, a process for cooperation between a stationary camera and an unmanned mobile device, for improving the identification of an object of interest, includes: receiving, at an electronic data processing device, from a stationary camera, a captured first view of a first captured object and determining, with a first level of certainty within a predetermined range of certainty levels, that the captured first view of the first object matches a first stored object of interest; identifying, by the electronic data processing device, one or more camera-equipped unmanned mobile vehicles in a specific direction of travel of the first captured object;a transmission, by the electronic data processing device to the one or more identified camera-equipped unmanned mobile vehicles, of a transmission instruction and interception information, wherein the interception information includes the specified direction of travel of the first detected object, information sufficient to identify either the first detected object or a vehicle in which the first detected object is traveling, and information identifying a desired second viewpoint of the first detected object that is different from the first viewpoint; a reception, by the electronic data processing device, via the identified one or more camera-equipped unmanned mobile vehicles, of a detected second viewpoint of the first detected object;and the use, by the electronic data processing device, of the captured second viewpoint of the first captured object to determine, with a second level of security, that the first captured object corresponds to the stored object of interest.

[0008] In a further embodiment, an electronic data processing device for the cooperation of a stationary camera and an unmanned mobile device to improve the identification of an object of interest comprises: an interface for a stationary camera; an interface for an unmanned mobile device; a memory; a transceiver; and one or more processors configured to: receive, via the interface for a stationary camera and from a stationary camera, a captured first view of a first captured object and determine, with a first security level within a predetermined range of a security level, that the captured first view of the first object corresponds to a first stored object of interest; identify one or more camera-equipped unmanned mobile vehicles in a specific direction of travel of the first captured object;Transmitted, via the transceiver to the one or more identified camera-equipped unmanned mobile vehicles, a transmission instruction and interception information, wherein the interception information includes the specified direction of travel of the first detected object, information sufficient to identify either the first detected object or a vehicle in which the first detected object is traveling, and information identifying a desired second viewpoint of the first detected object that is different from the first viewpoint; received, via the transceiver and via the one or more identified camera-equipped unmanned mobile vehicles, a captured second viewpoint of the first detected object; and used the captured second viewpoint of the first detected object to determine, with a second level of assurance, that the first detected object matches the stored object of interest.

[0009] Each of the aforementioned embodiments is explained in more detail below, beginning with an exemplary communication system and device architectures of the system in which the embodiments can be implemented, followed by a description of processing steps to achieve an improved method, device, and system for enhanced object recognition security of objects imaged by stationary cameras, through intelligent transmission and coordination between stationary cameras and identified camera-equipped unmanned mobile vehicles. Further advantages and features consistent with this disclosure are set forth in the following detailed description with reference to the figures. 1. Communication system and device structures

[0010] Now referring to the drawings and in particular Fig. Figure 1, an exemplary communication system diagram illustrates a system 100 comprising a first fixed video camera 102, a first camera-equipped unmanned mobile vehicle 104, and a first object to be detected 106 in a first vehicle 108. Each of the two, the first fixed video camera 102 and the first camera-equipped unmanned mobile vehicle 104, can be capable of communicating directly wirelessly via a direct-mode wireless link 142 or a wired connection, and / or can be capable of communicating wirelessly via a wireless infrastructure radio access network (RAN) 152 or corresponding wireless infrastructure links 140, 144.

[0011] The stationary video camera 102 can be any image recording device capable of recording still or moving images in an associated area of ​​interest located in Fig. Figure 1 is depicted as a road, but in other embodiments it may also include a building entrance, a bridge, a sidewalk, or another area of ​​interest. The stationary video camera 102 is stationary in the sense that it cannot physically move itself in any substantial direction (for example, more than one foot or one inch in any horizontal or vertical direction). However, this does not mean that it cannot pan, tilt, or zoom from its stationary position to cover a larger area of ​​interest than it could without such panning, tilting, or zooming.The fixed video camera 102 can be continuously switched on, can periodically capture images at a regular cadence, or can be triggered to begin capturing images and / or video as a result of another action, such as the detection of movement in the associated area of ​​interest by a separate motion detector communicatively coupled to the fixed video camera 102. The fixed video camera 102 can include a CMOS or CCD imager, for example, for digitally capturing images and / or video of an associated area of ​​interest. Images and / or video captured by the fixed video camera 102 can be stored in the fixed video camera 102 itself and / or can be transmitted to a separate storage or processing device via a wireless direct mode connection 142 and / or RAN 152. While the fixed video camera 102 is in . Fig. 1 shown as attached to a street light or street pole, in other embodiments the stationary video camera 102 may be attached to a building, traffic light, street sign or other structure.

[0012] The first camera-equipped unmanned mobile vehicle 104 can be a camera-equipped, flyable aerial drone with an electromechanical propulsion element, an image-capturing camera, and a microprocessor capable of flying under its own control, under the control of a remote operator, or a combination thereof, and capturing images and / or video of an area of ​​interest before, during, or after flight. The image-capturing camera attached to the unmanned mobile vehicle 104 can be fixed in its orientation (and thus require repositioning of the mobile vehicle 104 to which it is attached) or can include a pan, tilt, or zoom motor for independent control of the pan, tilt, and zoom functions of the image-capturing camera.The first camera-equipped unmanned mobile vehicle 104 could additionally or alternatively, among many other possibilities, be a ground-based or water-based mobile vehicle, although it is in . Fig. Figure 1 depicts an airborne drone. The image-capturing camera attached to the unmanned mobile vehicle 104 can be permanently switched on, can periodically capture images at a regular cadence, or can be triggered to begin capturing images and / or video as a result of another action, such as the unmanned mobile vehicle 104 being deployed to a specific area of ​​interest or being sent with instructions to intercept a particular type of person, vehicle, or object. The image-capturing camera may include a CMOS or CCD imager, for example, for digitally capturing images and / or video of the associated area of ​​interest, a person, a vehicle, or an object of interest.Images and / or video captured by the image-capturing camera can be stored on the unmanned mobile vehicle 104 itself and / or transferred to a separate storage or processing device via a wireless direct connection 142 and / or RAN 152. Although the unmanned mobile vehicle 104 is located in . Fig. 1 is shown as being temporarily positioned on a street light or road pole (possibly acting as a charging station for charging a battery in the unmanned mobile vehicle 104 while it is not flying), in other embodiments the unmanned mobile vehicle 104 may be positioned on a building, on a traffic light or on another structure.

[0013] The RAN 152 infrastructure can implement a conventional or trunked land mobile radio (LMR) standard or protocol over wireless links 140, 144, such as ETSI Digital Mobile Radio (DMR), a Project 25 (P25) standard defined by the Association of Public Safety Communications Officials International (APCO), terrestrial trunked radio (TETRA), or other LMR radio protocols or standards. In other embodiments, the RAN 152 infrastructure can additionally or alternatively implement a Long Term Evolution (LTE) protocol over wireless links 140, 144, including Multimedia Broadcast Multicast Services (MBMS), an open mobile alliance (OMA) Push to Talk (PTT) over cellular (OMA-PoC) standard, a Voice over IP (VolP) standard, or a PTT over IP (PolP) standard.In further embodiments, the RAN 152 infrastructure can additionally or alternatively implement a Wi-Fi protocol, perhaps according to an IEEE 802.11 standard (for example, 802.11a, 802.11b, 802.11g), or a WiMAX protocol, perhaps operating according to an IEEE 802.16 standard, via wireless connections 140, 144. Other types of wireless protocols could also be implemented. The RAN 152 infrastructure is in . Fig. 1 is shown as providing coverage for the fixed camera 102 and the unmanned mobile vehicle 104 via a single fixed terminal 154 connected to a controller 156 (for example, a radio controller, a call controller, a PTT server, a zone controller, an MME, a BSC, an MSC, a site controller, a push-to-talk controller, or another network device), and may include a transmit console 158 operated by a sender. In other embodiments, more or different types of fixed terminals may provide RAN services for the fixed camera 102 and the unmanned mobile vehicle 104 and may or may not include a separate controller 156 and / or a transmit console 158.

[0014] The first object to be recorded, 106, is in Fig. 1 is represented as a facial profile of the driver of a vehicle 108. The stationary camera 102 can acquire an initial image and / or video capture of the first object 106 while the vehicle 108 passes the stationary camera 102. However, the initial image and / or video capture may only be one point of view (POV, for example, a profile, a frontal view, or a three-quarter profile) or may only be a partial capture (for example, due to interfering objects, such as other cars, people, or traffic objects, or due to reflections or weather, among other possible interferences).The limitations may result in a data processing device that processes the initial image or video capture and attempts to match an object in the initial image and / or video capture with a first stored object of interest with a level of certainty lower than absolute certainty (for example, less than 100%, less than 90%, less than 80%, 70%, or 60%). Furthermore, the data processing device that processes the initial image and / or video capture may match the initial object in the initial image and / or video capture with the first stored object of interest with a level of certainty higher than absolute uncertainty (for example, greater than 0%, 10%, 20%, 30%, or 40%).Within these different areas of less than absolute certainty and more than absolute uncertainty, it would be helpful to obtain a second POV or less obstructed POV of the first object through coordination and cooperation with the unmanned mobile vehicle 104 in order to determine with a higher level of certainty whether the first object matches the object of interest.

[0015] Preferably, and in accordance with the Fig. 3 and Fig. 4, the fixed camera 102 and the unmanned mobile vehicle 104 (along with other potentially camera-equipped mobile vehicles that are in Fig. (1 not shown) cooperate or coordinate to obtain a second POV of an image and / or video capture of the object and to obtain an associated second assurance of a match.

[0016] Now referring to Fig. Figure 2, an exemplary overview diagram 200, illustrates the positioning and coordination of a fixed camera 202 and a camera-equipped unmanned mobile vehicle 204 for capturing a second object 206 in a second vehicle 208 with respect to an underlying cartographic road map. The fixed camera 202 can be the same or a similar camera to the fixed camera 102, which is positioned with respect to Fig. 1 described above, and the unmanned mobile vehicle 204 may be the same or a similar vehicle to the unmanned mobile vehicle 104 described above in relation to Fig. 1 is described. Vehicle 208 is in Fig. 2 is depicted as traveling west along a road 209 when the fixed camera 202 captures an initial image and / or video of the second object 206 in the second vehicle 208 and associates it with a stored object of interest with an initial certainty that is less than but greater than absolute certainty. For example, the initial image or video of the second object 206 could be a capture of a driver's face in a side profile. The initial certainty could be less than 90%, or less than 80%, or 70%, but greater than 10%, or greater than 20%, 30%, or 40%.

[0017] In some embodiments, the range of the security level within which the cooperation or coordination described herein can be triggered may vary on the basis of metadata associated with the object of interest, which may be stored alongside the object of interest or may be obtained separately from another database but linked to the object of interest.For example, a facial image as an object of interest associated with a person charged or convicted of a crime may have an associated security level range of 10% (or 20%) to 80% (or 70%) to trigger the cooperation or coordination processes described herein, whereas a facial image as an object of interest associated with a person charged or convicted of a misdemeanor may have an associated security level range of 35% (or 45%) to 85% (or 75%). In some embodiments, the described technical processes may only be applicable to certain types of offenses, such as crimes, and not to all other types of offenses, such as misdemeanors.

[0018] If a match is made with a security level that exceeds the upper limit of the aforementioned areas, a police officer or other responder may be automatically dispatched.To avoid dispatching a police officer or otherwise alerting law enforcement or other responders in the case of an uncertain match, but because there is a higher than absolute uncertainty for a match (0%), the fixed camera 202 or some other data processing devices processing the first image and / or video may cause a nearby camera-equipped unmanned mobile vehicle 204 to be dispatched to capture a second viewpoint as an image of the second object 206 driver (preferably different from the first, such as a three-quarter profile view or a frontal profile, or perhaps an unobstructed side profile compared to a distorted first image or video). As in . Fig. As shown in Figure 2, the data processing device can access a cartographic database of roads and, based on a detected direction of movement from the first image and / or video, or as reported by a sensor other than the stationary camera 202, identify one or more possible future paths 210, 212 that the second object 206 (and / or the vehicle 208 carrying the second object 206) is likely to take. In one example, the data processing device can identify all possible paths and transmit, via a transceiver, a transmit instruction (directing the unmanned mobile vehicle to position itself to capture a second image / video of the second object of interest, including any necessary power-up and / or movement from its temporary loading platform) and intercept information (including information sufficient to detect the second object 206 and / or the vehicle 208).to identify the second object 206, such as, but not limited to, the first image and / or video of the second object, the location of the fixed camera 202, the make, model and / or color of the vehicle in which the second object is traveling, a license plate associated with the vehicle in which the second object is traveling, a road on which the vehicle is traveling, the status of any detected turn signals associated with the vehicle, a desired type of second view (for example, frontal, three-quarter profile and / or profile) and / or other information necessary or helpful in identifying the second object 206 and / or the vehicle 208.

[0019] In some embodiments, a data processing device can be located in the stationary camera 202 or a data processing device can be located in the RAN 152. Fig. 1 (in Fig. 2 (not shown) from all possible future paths using interception information provided by the fixed camera, such as a current lane or the turn signal of vehicle 208, and preselect fewer than all unmanned mobile vehicles 204 for transmission. For example, if the left turn signal of vehicle 208 is on, the data processing device can send the transmission instruction and interception information only to the vehicle 208 that is currently traveling in the same direction. Fig. 2 lower unmanned mobile vehicles 204 transferred and not to the upper unmanned mobile vehicle 204.

[0020] In further embodiments, a data processing device arranged in or coupled to the stationary camera 202 can wirelessly transmit the transmission instruction and the interception information for direct wireless reception by one or more of the unmanned mobile vehicles 204 within the transmission range of the stationary camera 202. These unmanned mobile vehicles 204, which receive the transmission instruction and the interception information, can then determine individually or as a group whether and which can intercept the second object and / or vehicle and can transmit an acknowledgment and response to the stationary camera, including their current position and / or image parameters of their associated cameras (for example, resolution, field of view, focal length, light sensitivity, and so on).The data processing device could then determine which, or possibly all, of the confirming unmanned mobile vehicles will be authorized to send an intercepting image and / or video of the second object based on, among other possibilities, geographical proximity (preferably closer to the direction of travel of the second object and / or vehicle) and / or imaging parameters (preferably enhanced / higher imaging parameters).

[0021] Now referring to Fig. Figure 3, a schematic sketch, illustrates a data processing device 300 according to some embodiments of the present disclosure. The data processing device 300 can, for example, be embedded in the stationary video camera 102, in a processing unit adjacent to the stationary video camera 102 but communicatively coupled to the stationary video camera 102, in a remote server device in the RAN 152 (such as the controller 156), which is accessible to the stationary video camera 102 and the unmanned mobile vehicle 104 via the RAN 152, in the unmanned mobile vehicle 104, or at any other network location. As shown in Figure 3, the data processing device 300 can be embedded in the stationary video camera 102, in a processing unit adjacent to the stationary video camera 102 but communicatively coupled to the stationary video camera 102, in a remote server device in the RAN 152 (such as the controller 156), which is accessible to the stationary video camera 102 and the unmanned mobile vehicle 104 via the RAN 152, in the unmanned mobile vehicle 104, or at any other network location. Fig. As shown in Figure 3, the data processing device 300 includes a communication unit 302, which is coupled to a general data and address bus 317 of a processing unit 303. In some embodiments, the data processing device 300 may also include an input unit (for example, a keyboard, a pointing device, a touch-sensitive surface, and so on) 306 and a screen 305, each of which is coupled such that it communicates with the processing device 303.

[0022] A microphone 320 may be present for capturing audio simultaneously with an image or video, which is further encoded by the processing unit 303 and transmitted as audio / video stream data through the communication unit 302 to other devices. A communication speaker 322 may be present for reproducing audio sent to the data processing device 300 via the communication unit 302, or may be used to play alarm tones or other types of pre-recorded audio when a match with an object of interest is found, in order to alert nearby police officers.

[0023] The processing unit 303 can include a code read-only memory (ROM) 312, which is coupled to the general-purpose data and address bus 317, for storing data to initialize system components. The processing unit 303 can also include a microprocessor 313, which is coupled via the general-purpose data and address bus 317 to a random-access memory (RAM) 304 and a static memory 316.

[0024] The communication unit 302 can include one or more wired or wireless input / output (I / O) interfaces 309 that are configurable to communicate with other devices, such as a portable radio, a tablet, wireless RAN and / or vehicle transceivers.

[0025] The communication unit 302 can include one or more wireless transceivers 308, such as a DMR transceiver, a P25 transceiver, a Bluetooth transceiver, a Wi-Fi transceiver that may operate according to an IEEE 802.11 standard (for example, 802.11a, 802.11b, 802.11g), an LTE transceiver, a WiMAX transceiver that may operate according to an IEEE 802.16 standard, and / or similar types of wireless transceivers that are configurable to communicate over a wireless radio network. The communication unit 302 may additionally or alternatively include one or more wired transceivers 308, such as an Ethernet transceiver, a Universal Serial Bus (USB) transceiver, or similar transceivers that are configurable to communicate with a wired network via a twisted-pair line, a coaxial cable, a fiber optic connection, or a similar physical connection.The transceiver 308 is also connected to a combined modulator / demodulator 310.

[0026] The microprocessor 313 has connections for connecting to the input unit 306 and the microphone unit 320, and to the screen 305 and the speaker 322. The static memory 316 can store operating code 325 for the microprocessor 313, which, when executed, performs one or more of the data processing device steps described in Fig. 4 and the accompanying text and / or Fig. 5 and the accompanying text.

[0027] Static storage 316 can include, for example, a hard disk drive (HDD), an optical drive such as a CD drive or a DVD drive, a solid-state drive (SSD), a tape drive, a flash memory drive, or a tape drive, to name a few. 2. Processes to improve object recognition reliability of objects compared to objects of interest by coordinating stationary and camera-equipped unmanned mobile vehicles.

[0028] Now to the Fig. 4 and Fig. 5. The flowcharts illustrate procedures 400 and 500 for improving the security level of object detection of objects compared to objects of interest by coordinating stationary and camera-equipped unmanned mobile vehicles. While a specific sequence of processing steps, message receptions, and / or message transmissions in the Fig. 4 and Fig. 5 is given for illustrative purposes; the timing and sequence of these steps, receptions, and transmissions may vary without negating the purpose and benefits of the examples detailed later in this disclosure. An associated data processing device, such as the one described in Fig. 3 described above, the method 400 and / or 500 can be executed upon power-up, a predetermined period thereafter, in response to a trigger initiated locally on the device via an internal process, or in response to a trigger generated externally by the data processing device and received via an input interface, among other possibilities.In particular, Method 400 describes a process for improving the security of object detection of objects compared to objects of interest by coordinating stationary and camera-equipped and unmanned mobile vehicles from the perspective of a centralized data processing device, while Method 500 describes a similar process for improving the security of object detection of objects compared to objects of interest by coordinating stationary and camera-equipped unmanned mobile vehicles from the perspective of a distributed wireless network of devices.

[0029] Method 400 begins at step 402, in which a data processing device communicatively connected to a first stationary camera receives a first point of view (POV) of an image and / or video of a first object from the stationary camera. The stationary camera could be, for example, a fixed camera attached to a power pole, a camera mounted on an ATM, a camera attached to a traffic light, or any other stationary camera with a field of view that covers the first object.

[0030] In step 404, the data processing device compares the first POV of the first object with an object of interest that is stored at the data processing device or stored remotely from the data processing device but is accessible to the data processing device.For example, the stored object of interest may be an image of a person of interest (inside or outside a vehicle, and for example a facial capture of the person of interest or distinctive features, such as a tattoo or scar of the person of interest) or a vehicle (for example, an image of the vehicle, a specific make and model of the vehicle, or a specific license plate number of the vehicle) or it may be an image of another object (such as a specific type of firearm, a specific word or phrase on a car sticker, a specific hat worn by a person, a specific warning sticker associated with a flammable, explosive, or biological substance, and / or other types of objects of interest).

[0031] The data processing device can determine whether the first point of view (POV) of the image and / or video of the first object of interest corresponds to a first specified security level within a predetermined security level range. As outlined above, this predetermined security level range extends below absolute security (for example, less than 100%, less than 90%, less than 80%, 70%, or 60%) and above absolute uncertainty (for example, greater than 0%, greater than 10%, greater than 20%, greater than 30%, or 40%) and can vary based on an identity or category associated with the object of interest (for example, a larger range but possibly shifted downwards overall, such as 10% to 60% security for more serious or dangerous objects of interest, such as persons convicted of crimes).or graphics associated with biological hazards, and a smaller range, but possibly shifted upwards overall, such as 40 to 70% for less serious or dangerous objects of interest, such as individuals convicted of misdemeanors or graphics associated with criminal organizations). For certain security levels that exceed the upper limit of the predetermined security level range, such as 60% for more serious or dangerous objects of interest, a responder may always be dispatched, and the disclosed security confirmation processes will not be executed.

[0032] Various text, image, and / or object recognition algorithms can be used to compare the captured point of view (POV) of the first object with the stored object of interest (OFI). These include, but are not limited to, geometric hashing, edge detection, scale-invariant feature transformation (SIFT), speeded-up robust features (SURF), neural networks, deep learning, genetic optical character recognition (OCR), gradient-based and derivative-based matching approaches, the Viola-Jones algorithm, template matching, image segmentation, and blob analysis. The confidence level of a match is provided by the matching algorithm and represents a numerical measure of how confident the algorithm is that the first object (from the first captured image and / or video of the first POV) matches the object of interest.This numerical representation can be a percentage (between 0 and 100%, as described above), a decimal value (between 0 and 1), or any other predefined numerical range with upper and lower bounds. One or more additional views of the first object may be necessary to make the algorithm more confident that there is a match between the first object and the object of interest stored in a text, image, or object database. If the match at step 404 does not fall within the predefined confidence level range, processing returns to step 404, and the first object is compared to another stored object of interest.Once all objects of interest have been compared, the procedure can stop at 400 and the system can ultimately refrain from sending a first responder or otherwise taking action concerning the first object.

[0033] On the other hand, and assuming that the first point of view (POV) of the first object matches a stored object of interest with a first determined security level within the predetermined security level range, processing continues at step 406, where the data processing device identifies one or more camera-equipped unmanned mobile vehicles (UMVs) in a direction of travel of the first object. The data processing device can maintain a predetermined list of UHV locations or dynamically update a managed list of UHV locations as they report their current locations. Each UHV in the list can be identified by a unique alphanumeric identifier and can also be associated with a status, such as whether it is available for dispatch, its battery charge levels,its estimated flight times based on the remaining energy level, imaging characteristics associated with the unmanned vehicle's camera, and other information useful for the data processing device to determine which camera-equipped unmanned mobile vehicle(s) to dispatch to intercept the first object for additional image acquisition. For example, the data processing device can select an available camera-equipped unmanned mobile vehicle from the list, with a position closest to an expected interception point of the first object (or the vehicle carrying the first object), taking into account speed, direction, lane, and / or cartographic information of roads and / or other routes along which the first object or vehicle is traveling.a camera-equipped unmanned mobile vehicle with a highest image acquisition parameter relative to the first object (for example, preferring a better shutter speed if the first object / vehicle is traveling at high speed, preferring a higher resolution if the first object / vehicle is traveling at a low / nominal speed below, for example, 25 or 35 miles per hour, preferring a higher zoom capability and / or optical image stabilization capability if the object of interest is relatively small, for example, less than 25 cm, 2 in the area or 25 cm 3in volume, wherein an airborne mobile vehicle or a land-based mobile vehicle is selected based on an altitude at which the first object exists and a desired second point of view (POV) of the first object, or a combination of the foregoing. Parameters such as those foregoing may be pre-filled in the list maintained by the data processing device or made accessible to the data processing device, or may be transmitted to the data processing device from each camera-equipped unmanned mobile vehicle and stored in the list. In some cases, the data processing device may identify only a single best or most capable camera-equipped unmanned mobile vehicle at step 406, while in other embodiments, the data processing device may identify two or more camera-equipped unmanned mobile vehicles at step 406 using one or more of the foregoing parameter sets.

[0034] Once the data processing device has identified one or more preferred camera-equipped unmanned mobile vehicles at step 406, it transmits a send instruction and intercept information to the identified one or more camera-equipped unmanned mobile vehicles via a transceiver at step 408. The send instruction and intercept information can be transmitted to the one or more camera-equipped unmanned mobile vehicles via a RAN, such as RAN 152, and the wireless infrastructure links 140, 144, as described in Fig. 1 shown, or via wireless direct mode connections, such as those in Fig. 1. Wireless connection 142, as shown, is transmitted. In further examples, a wired connection can be used to communicatively link the data processing device with the identified one or more camera-equipped unmanned mobile vehicles, in order to ultimately reach the identified one or more preferred camera-equipped unmanned mobile vehicles via a near-field or wired coupling at a charging point where the mobile vehicle is at rest.

[0035] The transmission instruction may be integrated into the interception information message or sent as a separate message requiring the receiving identified one or more preferred camera-equipped unmanned mobile vehicles to take action to intercept the first object and capture a second POV of the image and / or video of the first object to improve certainty that the first object matches the object of interest, alone or in combination with the first POV of the first captured object from the fixed camera obtained at step 402.

[0036] The interception information includes information sufficient for the identified one or more preferred camera-equipped unmanned mobile vehicles to identify the first object and capture a second point of view (POV) of the image / video of the first object. For example, the interception information may include a copy of the captured first POV of the image and / or video of the first object, taken by the stationary camera at step 402, which the identified one or more preferred camera-equipped unmanned mobile vehicles can use to monitor an area around them for the first approaching object.In other embodiments, the interception information may include information that identifies a vehicle in which the first object is traveling (which may or may not be included in the first POV of the image and / or video provided by the fixed camera, but which may have been separately imaged and / or processed by the fixed camera), such as the make and model of a vehicle on or in which the first object is traveling, or a license plate number of a vehicle on or in which the first object is traveling.Additionally or alternatively, the interception information can include information identifying the direction and / or travel speed of the first object or vehicle on or in which it is traveling. It can also include the lane the vehicle is traveling in from a variety of available lanes, and the status of a turn signal, such as a flashing light or hand gesture indicating an impending turn. Furthermore, the interception information can include the position of the fixed camera and the time at which the first object was detected by the fixed camera. Other information can also be recorded. This information can then be used by the received, identified one or more preferred camera-equipped unmanned mobile vehicles to locate the first object more quickly and accurately.

[0037] In some embodiments, the transmission instruction or interception information may also include information sufficient to identify a desired point of view (POV) of the first object, thus maximizing security. For example, and as described in Fig. 6. As shown, if the first object is the face of a driver in a vehicle, captured by the fixed camera in a first POV of a profile view 604 or a three-quarter profile view 606, either due to obstacles, imaging characteristics of the fixed camera, or the characteristics of the profile view / three-quarter profile view itself, the certainty of a match between the first POV of the profile view 604 / three-quarter profile view 606 and a stored object of interest (perhaps a stored frontal view 602 of a particular suspect wanted on a warrant or for another reason) may be 60% and may fall into a predetermined range of certainty levels where it is not completely certain that there is a match, but it is also not completely certain that there is no match.Accordingly, the data processing device can identify a specific desired POV as a function of the first POV already provided, which is an improvement on the first POV and / or complementary to the first POV to maximize a second level of security, such as a frontal view 602 of the first object that the data processing device wishes to receive from the identified one or more preferred camera-equipped unmanned mobile vehicles, and can transmit such a request or instruction in the transmission instruction or interception information. The receiving identified one or more preferred camera-equipped unmanned mobile vehicles can then use this additional information to position themselves to obtain the specific desired POV of the first object.

[0038] In step 410, in response to the transmission of the sending instruction and the interception information to the identified one or more preferred camera-equipped unmanned mobile vehicles in step 408, the data processing device receives a captured second POV of the image and / or video of the first object. The second POV of the image and / or video may be captured from the same or a different POV than the first POV of the image and / or video of the first object, may be captured from the same or a different distance than the first POV of the image and / or video, or may be captured using a camera with different, preferably higher, imaging parameters than the first POV of the image and / or video.In some cases where the transmission instruction and interception information are transmitted to multiple preferred camera-equipped unmanned mobile vehicles at step 408, multiple captured second POVs of the images and / or videos of the first object can be received by the data processing device at step 410.

[0039] In step 412, the captured second POV of the image and / or video (or images and / or videos if there are multiple) is fed back into the same or a similar object identification / matching algorithm as described above in step 404, accompanied or not by the first POV of the image and / or video, and a second level of assurance is determined that the first object matches the object of interest from step 404.In most cases, the second determined security level will differ from the first determined security level and preferably be significantly greater than the first determined security level (for example, twenty percentage points or more indicating that a match exists, and in some embodiments above the upper limit of the predetermined range of the security level, meaning that the second POV of the first object has helped to confirm that the first object matches the object of interest) or significantly lower than the first determined security level (for example, twenty percentage points or more lower indicating that a match exists, and in some embodiments lower than the lower limit of the predetermined range of the security level, meaning that the second POV of the first object has helped to confirm that the first object does not match the object of interest).

[0040] In some embodiments, and as a result of finding a match with a higher second security level that is above the upper limit of the predetermined security area, for example, greater than 60%, 70%, 80%, or 90%, the data processing device can cause an alarm to be transmitted to a dispatch console and / or a mobile data processing device whose position is determined within a threshold distance of the one or more identified camera-equipped unmanned mobile vehicles, for example, 0.5, 1, or 2 miles. In some embodiments, the alarm can include the captured first and / or second point of view (POV) images and / or videos of the first object, or links thereto, and / or one or both of the determined first and second security levels.The alarm may also include the current position(s) of the one or more identified camera-equipped unmanned mobile vehicles (UMVs) if they are still tracking the initial object, and / or the positions of the fixed camera and the one or more identified camera-equipped UHVs when the first and second point-of-view images and / or videos were captured. In yet other examples, a live video stream (or streams) from the one or more identified camera-equipped UHVs of the initial object (assuming the initial object is still being tracked) may be provided after the alarm or in response to a receiving device activating a link in the alarm requesting such a live video stream.

[0041] The type of alarm provided may depend on the second security level and / or an identity or category associated with the object of interest (for example, a more conspicuous alarm, such as a haptic alarm and / or a flashing full-screen visual alarm message, for more serious or dangerous objects of interest, such as persons convicted of criminal offenses or graphics associated with biological hazards, and a more subtle alarm, such as an audio tone or text message, for less serious or dangerous objects of interest, such as persons convicted of a misdemeanor or graphics associated with known criminal organizations).

[0042] Now to Fig. 5, Method 500 describes a similar process for improving the security of object detection of objects compared to objects of interest by stationary and camera-equipped unmanned mobile vehicles from the perspective of a distributed wireless network of devices. Where similar steps and processes in Fig. 5 as in Fig. The descriptions already presented in section 4 are shown below. Fig. 4 included by reference, without repeating it here ipsis verbis.

[0043] Procedure 500 begins at step 502, similar to step 402 of Fig. 4, where a second data processing device receives a first point of view (POV) of a second object from a stationary camera. The second data processing device compares the first POV of the second object with an object of interest that is stored at the second data processing device or stored remotely from the second data processing device but made accessible to the second data processing device.

[0044] At step 504 and similarly to step 404 of Fig. 4. The second data processing device determines whether the first POV of the image and / or video of the second object matches an object of interest with a first determined security level within a predetermined range of security levels.

[0045] If the match at step 504 does not fall within the specified range of the security level, processing returns to step 504 and the second object is compared with another stored object of interest. Once all objects of interest have been compared, procedure 500 can stop, and the system can ultimately refrain from requesting a first responder or taking any other action regarding the second object.

[0046] On the other hand, and assuming that the first POV of the second object matches a stored object of interest with a first determined security level within the predetermined security level range, processing continues with step 508, similar to step 408 of Fig. 4, where the second data processing device this time causes a transmission, via the transceiver, of a transmit instruction and intercept information, which is wirelessly transmitted for reception by one or more camera-equipped unmanned mobile vehicles in an environment (for example, a wireless direct-mode transmission range) of the second data processing device and / or the transceiver. The transmit instruction and intercept information may be configured in a manner that contains the same or similar information as already described above in connection with step 408 of Fig. 4 is set out.

[0047] In some embodiments, the second data processing device can rely on receiving from one or more camera-equipped unmanned mobile vehicles to determine, individually or via additional wireless communications between them, which of the one or more camera-equipped unmanned mobile vehicles receiving the transmission should actually be deployed to intercept the second object and capture a second point of view of the images and / or videos of the second object. The individual or group determination can be performed using similar decision trees and parameters as those already described above in connection with the Fig. 1 and Fig. 4 are set out.

[0048] In alternative embodiments, and after initiating the wireless transmission of the transmission instruction and interception information, the second data processing device can wirelessly receive, via the transceiver, from one or more camera-equipped unmanned mobile vehicles receiving the transmission(s), respective acknowledgments of receipt of the transmission instruction and interception information and respective camera-equipped unmanned mobile vehicle parameter information, wherein the camera-equipped unmanned mobile vehicle parameter information includes one or more of the following: position information, battery level information, flight capabilities (such as speed, acceleration, flight ceiling) and camera imaging parameter information.The second data processing device can then use the parameter information received from the camera-equipped unmanned mobile vehicles to identify one of the two or more camera-equipped unmanned mobile vehicles for capturing the second POV of the image and / or video of the second object, similar to what has already been described above (for example, preferring a position that is closest to an expected interception point of the first object, and / or preferring image parameters of the camera of the mobile vehicle, and / or preferring a type of mobile vehicle).Once one specific object is identified, the second data processing device can cause the transmission of an additional electronic instruction, or transmit it directly to the specific one of the two or more camera-equipped unmanned mobile vehicles, which further instructs that specific one of the two or more camera-equipped unmanned mobile vehicles to capture the second point of view (POV) of the image and / or video of the second object. Other camera-equipped unmanned mobile vehicles that do not receive the further instruction from the second data processing device (or that receive the transmission but determine that they are not selected as the specific one of the two or more camera-equipped unmanned mobile vehicles to capture the second object) will refrain from capturing the second object for additional image and / or video capture.In some cases, the second data processing device can identify only a single, best, or most powerful particular camera-equipped unmanned mobile vehicle, while in other embodiments, the second data processing device can identify two or more camera-equipped unmanned mobile vehicles using one or more of the parameters shown above.

[0049] At step 510 and somewhat similar to step 410 of Fig. 4. In response to the transmission of the transmission instruction and interception information to the one or more camera-equipped unmanned mobile vehicles at step 508, the second data processing device receives a captured second POV of the image and / or video of the first object. In some cases where multiple camera-equipped unmanned mobile vehicles have received the transmission at step 508 and captured a second POV of the second object, the second data processing device may receive multiple captured second POVs of the images and / or video of the second object at step 510.

[0050] At step 512 and similarly to step 412 of Fig.4. The captured second POV of the image and / or video (or images and / or videos if there are multiple) is fed back into the same or a similar object identification / matching algorithm as described above in step 504, accompanied or not by the first POV of the image and / or video, and a second level of certainty is determined that the second object matches the object of interest from step 504.

[0051] In some embodiments, and as a result of finding a match with a higher second security level that is higher than the upper limit of the predetermined security level range, for example, greater than 60%, 70%, 80%, or 90%, the second data processing device can cause an alarm to be sent to a dispatch console or mobile data processing device located within a threshold distance, such as 0.5, 1, or 2 miles, from the one or more identified camera-equipped unmanned mobile vehicles. In some embodiments, the alarm can include the captured first and / or second point of view (POV) images and / or videos of the second object, or links thereto, and / or one or both of the specified first and second security levels.The alarm may further include the current position(s) of the one or more identified camera-equipped unmanned mobile vehicles while they are still tracking the second object (and sending location information back to the second data processing device), and / or the positions of the fixed camera and the one or more identified camera-equipped unmanned mobile vehicles when the first and second points of view (POV) of the images and / or videos are captured. In yet other examples, a live video stream (or streams) from the one or more identified camera-equipped unmanned mobile vehicles of the second object (assuming the second object is still being tracked) may be provided following the alarm or in response to a received device that activates a link in the alarm. 3. Summary

[0052] In accordance with the foregoing, an improved device, method, and system are disclosed for enhancing the reliability of object detection of objects imaged by fixed cameras through intelligent cooperation and coordination between the fixed cameras and one or more identified camera-equipped unmanned mobile vehicles. As a result of the foregoing, false positives and false negatives can be significantly reduced and / or eliminated, thereby improving the reliability of the imaging systems and reducing the number of police officers or other responders unnecessarily dispatched simply to confirm an initially low or medium level of certainty.

[0053] Specific embodiments have been described in the preceding specification. However, it is clear to those skilled in the art that various modifications and changes can be made without departing from the spirit of the invention, as set forth in the claims below. Accordingly, the specification and the figures are to be understood in an illustrative rather than a restrictive sense, and all such modifications are to be included within the scope of protection of the present teachings. The benefits, advantages, problem solutions, and any conceivable element that leads to the occurrence or enhancement of any benefit, advantage, or solution are not to be construed as critical, necessary, or essential features or elements of any claim or of all claims.The invention is defined exclusively by the attached claims, including any amendment made during the pendency of the present application and all equivalents of such claims as published.

[0054] Furthermore, in this document, relational expressions such as first and second, above and below, and the like are to be used solely to distinguish one entity or action from another, without necessarily requiring or implying any actual relationship or order between such entities or actions. The expressions "includes," "comprising," "has," "having," "include," "containing," "containing," or any variation thereof are to cover non-exclusive inclusion, so that a process, procedure, article, or device that includes, has, includes, or contains a list of elements may not only include such elements but may also include other elements not expressly listed or inherent in such processes, procedures, articles, or devices. An element that continues with "includes... a," "has..."The terms "one," "includes... one," and "contains... one" do not, without further stipulations, exclude the existence of additional identical elements in the process, method, article, or apparatus that comprise, have, include, or contain the element. The terms "one" and "a" are defined as one or more unless explicitly stated otherwise herein. The terms "essentially," "essentially," "approximately," "about," or any other version thereof are defined as "being close to" as is clear to those skilled in the art, and in one non-limiting embodiment, the term is defined as being within 10%, in another embodiment within 5%, in another embodiment within 1%, and in yet another embodiment within 0.5%. The term "coupled," as used herein, is defined as "connected," although not necessarily directly and not necessarily mechanically.A device or structure that is "configured" in a certain way is configured at least in that way, but may also be configured in at least one other way not listed. The terms "air interface" and "wireless connection" should also be considered interchangeable.

[0055] It is desired that some embodiments include one or more generic or specialized processors (or “processing devices”), such as microprocessors, digital signal processors, custom processors and freely programmable field-gate arrays (FPGAs) and unique stored program instructions (comprising both software and firmware) that control the one or more processors to implement, in conjunction with certain non-processor circuitry, some, most or all of the functions of the method and / or device described herein.Alternatively, some or all functions can be implemented by a state machine that has no stored program instructions, or in one or more application-specific integrated circuits (ASICs) where each function, or some combinations of certain functions, are implemented as custom logic. Naturally, a combination of the two approaches can be used.

[0056] Furthermore, an embodiment can be implemented as a computer-readable storage medium containing computer-readable code stored thereon for programming a computer (which, for example, includes a processor) to perform a method described and claimed herein. Examples of such computer-readable storage media include, but are not limited to: a hard disk, a CD-ROM, an optical storage device, a magnetic storage device, a ROM (read-only memory), a PROM (programmable read memory), an EPROM (erasable programmable read memory), an EEPROM (electrically erasable programmable read memory), and flash memory.Furthermore, it is to be expected that a person skilled in the art, regardless of potentially considerable effort and a wide range of design choices, which may be based, for example, on available time, current technology, and economic considerations, guided by the concepts and principles disclosed herein, will readily be able to generate such software instructions, programs, and integrated circuits with minimal experimental effort. The summary of the disclosure is provided to allow the reader to quickly grasp the nature of the technical disclosure. It is submitted with the understanding that it is not intended to interpret or limit the spirit or meaning of the claims. Additionally, it is clear from the preceding detailed description that various features in different embodiments are grouped together to streamline the disclosure.This disclosure procedure should not be interpreted as reflecting an intention that the claimed embodiments require more features than are expressly stated in each claim. Rather, as is evident from the following claims, an inventive subject matter exists in fewer than all the features of a single disclosed embodiment. Thus, the following claims are hereby integrated into the detailed description, with each claim standing alone as a separately claimed subject matter.

Claims

A method (400) for collaborating between a fixed camera and an unmanned mobile device to improve identification of an object of interest, the method comprising:receiving (402), at an electronic data processing device (300) from a fixed camera, a detected first view angle of a first detected object and determining, with a first confidence level within a predetermined range of confidence levels, that the detected first view angle of the first object matches a first stored object of interest;identifying (406), by the electronic data processing device, one or more camera-equipped unmanned mobile vehicles in a particular direction of travel of the first detected object;Transmitting (408), by the electronic data processing device, to the one or more identified camera-equipped unmanned mobile vehicles, a dispatch instruction and interception information, the interception information including the determined direction of travel of the first detected object, information sufficient to identify either the first detected object or a vehicle in which the first detected object is traveling, and information identifying a desired second view angle of the first detected object that is different from the first view angle;Receiving (410), by the electronic data processing device, via the identified one or more camera-equipped unmanned mobile vehicles, a detected second view angle of the first detected object;andusing (412), by the electronic data processing device, the detected second view angle of the first detected object to determine with a second level of confidence that the first detected object matches the stored object of interest; The method (400) of claim 1, further comprising using, by the electronic data processing device, the detected first view angle of the first detected object and the detected second view angle of the first detected object to determine, with the second level of confidence, whether the first detected object matches the stored object of interest. Method (400) according to claim 1, wherein the predetermined range of the security level is below 70%. The method (400) of claim 1, wherein the first stored object of interest is a person of interest, the first view angle is a lateral view angle of the person of interest, and the second view angle is a frontal view angle of the person of interest. The method (400) of claim 4, wherein the person of interest travels in a motor vehicle. The method (400) of claim 5, wherein the intercept information includes information sufficient to identify the motor vehicle. The method (400) of claim 6, wherein the intercept information further includes a particular lane from a plurality of available lanes in the particular direction of travel that the motor vehicle had at the time the first view angle was detected. The method (400) of claim 7, wherein identifying one or more camera-equipped unmanned mobile vehicles in a particular direction of travel of the first detected object comprises identifying, by the electronic data processing device, a plurality of camera-equipped unmanned mobile vehicles based on the particular direction of travel of the first detected object and identifying a plurality of possible navigation paths using cartographic navigation information obtained from a cartographic database, each of the plurality of camera-equipped unmanned mobile vehicles being associated with one of the plurality of possible navigation paths. The method (400) of claim 8, wherein a number of the identified plurality of possible navigation paths is greater than an available number of the identified plurality of camera-equipped unmanned mobile vehicles, and the method further comprises using, by the electronic data processing device, the determined lane included in the intercept information to select a subset of the plurality of possible navigation paths to assign to the plurality of camera-equipped unmanned mobile vehicles. The method (400) of claim 1, wherein the desired second viewing angle is selected to maximize the second security level, and the second security level is higher than the first security level. Method (400) according to claim 1, wherein the predetermined range of the security level is greater than a minimum positive threshold of the security level which is not equal to zero. Method (400) according to claim 11, wherein the minimum positive threshold of the security level which is not equal to zero is greater than 15%. Method (400) according to claim 11, wherein the predetermined range of the security level is below a maximum threshold of the security level, which is itself below a security level of 100%. Method (400) according to claim 13, wherein the maximum threshold of the security level is below 85%. The method (400) of claim 1, wherein the fixed camera has a low relative image capture parameter causing the first security level to be lower than desired, and wherein the one or more identified camera-equipped unmanned mobile vehicles have a higher relative image capture parameter causing the second security level to be higher than the first security level. Method (400) according to claim 1, wherein the one or more identified camera-equipped unmanned mobile vehicles comprise mobile aircraft and / or land-based motorized vehicles. The method (400) of claim 1, wherein the method further comprises, when the second security level is above a second security level threshold, causing, by the electronic data processing device, an alarm to be transmitted to a dispatch console and / or a mobile data processing device determined to be in an area within a threshold distance of the one or more identified camera-equipped unmanned mobile vehicles. The method (400) of claim 17, wherein the alarm includes the detected first and second viewpoints or links thereto and the first and / or second security level. Method (400) according to claim 17, wherein the second threshold of the security level is greater than 60%. An electronic data processing device (300) for cooperation between a fixed camera and an unmanned mobile device to improve the identification of an object of interest, the device comprising: a fixed camera interface; an unmanned mobile device interface; a memory; a transceiver; and one or more processors configured to: receive (402), via the fixed camera interface and from a fixed camera, a detected first view angle of a first detected object and determine, with a first confidence level within a predetermined range of a confidence level, that the detected first view angle of the first object matches a first stored object of interest; identify (406) one or more camera-equipped unmanned mobile vehicles in a particular direction of travel of the first detected object;Transmitting (408), via the transceiver to the one or more identified camera-equipped unmanned mobile vehicles, a dispatch instruction and intercept information, the intercept information including the determined direction of travel of the first detected object, information sufficient to identify either the first detected object or a vehicle in which the first detected object is traveling, and information identifying a desired second view angle of the first detected object that is different from the first view angle;Receiving (410), via the transceiver and via the one or more identified camera-equipped unmanned mobile vehicles, a detected second view angle of the first detected object;andusing (412) the detected second view angle of the first detected object to determine, with a second level of confidence, that the first detected object matches the stored object of interest;

Images