# US 11,445,150 B2 · Multi-camera collaboration-based image processing method and video surveillance system > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 11,445,150 B2 - Document ID: US11445150B2 - Status: Granted - Kind: B2 - Application number: 17/120,471 - Filed: 2020-12-14 - Published: 2021-05-06 - Granted: 2022-09-13 - Earliest claimed: 2018-08-01 - Assignees: Huawei Technologies Co., Ltd. - Applicants: Huawei Technologies Co., Ltd. - Inventors: Ligang Yang - Web page: https://patentgrep.com/patent/US11445150B2 ## CPC - H04N 7/181: {for receiving images from a plurality of remote sources} - G06F 18/22: Matching criteria, e.g. proximity measures - G06V 20/46: {Extracting features or characteristics from the video content, e.g. video fingerprints, representative shots or key frames} - G06V 20/52: Surveillance or monitoring of activities, e.g. for recognising suspicious objects (recognising microscopic objects G06V20/69) - H04L 67/025: for remote control or remote monitoring of applications - H04L 67/125: involving control of end-device applications over a network - H04N 23/611: where the recognised objects include parts of the human body - H04N 23/661: Transmitting camera control signals through networks, e.g. control via the Internet - H04N 23/90: Arrangement of cameras or camera modules, e.g. multiple cameras in TV studios or sports stadiums - H04N: PICTORIAL COMMUNICATION, e.g. TELEVISION - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING - H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M) ## Abstract A video surveillance system includes at least two camera and an analysis device. The at least two cameras include a first image camera and a second image camera. The method includes: separately obtaining, by the first image camera and the second image camera, a photographed first image and a photographed second image; obtaining indication information of a target task; separately obtaining first execution result information and second execution result information of the target task based on the indication information of the target task; sending, the first execution result information and the second execution result information to the analysis device; and determining, by the analysis device, third execution result information of the target task based on the first execution result information and the second execution result information. ## Claims ### Claim 1 A multi-camera collaboration-based image processing method, applied to a video surveillance system, wherein the video surveillance system comprises at least two cameras and an analysis unit, the at least two cameras comprise a first camera and a second camera, and the method comprises: separately obtaining, respectively by the first camera and the second camera, a photographed first image and a photographed second image; receiving, by the first camera and the second camera, indication information of a target task, wherein the target task is image recognition and the indication information of the target task indicates characteristic information of a to-be-recognized image, a similarity condition and a process to be performed on a target object in the first and second photographed images; separately generating, respectively by the first camera and the second camera, first execution result information and second execution result information of the target task based on the indication information of the target task, wherein the first execution result information comprises first processing result information related to characteristic information of the target object in the first image, and the second execution result information comprises second processing result information related to characteristic information of the target object in the second image; sending, by the first camera and the second camera, the first execution result information and the second execution result information to the analysis unit; and determining, by the analysis unit, third execution result information of the target task based on the first execution result information and the second execution result information; wherein the step of generating, by the first camera, first execution result information of the target task comprises: extracting, by the first camera, first characteristic information of the target object from the first image; and generating, by the first camera, based on the first characteristic information and the characteristic information of the to-be-recognized image, first similarity information between a to-be-recognized object and the target object in the first image; and wherein the step of sending, by the first camera, the first execution result information to the analysis unit, the method further comprises: determining, by the first camera, the first similarity information meets the similarity condition. ### Claim 2 (depends on 1) The method according to claim 1, wherein: the analysis unit is a third camera different from the first camera and the second camera. ### Claim 3 (depends on 1) The method according to claim 1, wherein: the first camera comprises the analysis unit; or the second camera comprises the analysis unit. ### Claim 4 (depends on 1) The method according to claim 1, wherein: the first execution result information is related to object recognition in the first image; and the second execution result information is related to object recognition in the second image. ### Claim 5 (depends on 1) The method according to claim 1, wherein: the first processing result information is related to similarity between a to-be-recognized object and the target object in the first image; the second processing result information is related to similarity between the to-be-recognized object and the target object in the second image; the third execution result information is related to similarity between the to-be-recognized object and the target object. ### Claim 6 (depends on 1) The method according to claim 1, wherein the video surveillance system further comprises a task allocation unit, and before the receiving, by the first camera, indication information of the target task, the method further comprises: determining, by the task allocation unit, the indication information of the target task; and sending, by the task allocation unit, the indication information of the target task to the first camera. ### Claim 7 (depends on 6) The method according to claim 6, wherein after the receiving, by the first camera, indication information of the target task, and before the generating, by the first camera, first execution result information of the target task based on the indication information of the target task, the method further comprises: determining, by the first camera, whether to accept the target task; and in response to the first camera accepting the target task, performing, by the first camera, the step of generating first execution result information of the target task based on the indication information of the target task. ### Claim 8 (depends on 6) The method according to claim 6, wherein after the receiving, by the task allocation unit, the indication information of the target task, and before the sending, by the task allocation unit, the indication information of the target task to the first camera, the method further comprises: selecting, by the task allocation unit, the first camera from the at least two cameras. ### Claim 9 (depends on 8) The method according to claim 8, wherein the selecting, by the task allocation unit, the first camera from the at least two cameras comprises: receiving, by the task allocation unit, resource information sent by the at least two cameras; and selecting, by the task allocation unit, the first camera from the at least two cameras based on the resource information. ### Claim 10 (depends on 8) The method according to claim 8, wherein the selecting, by the task allocation unit, the first camera from the at least two cameras comprises: determining, by the task allocation unit, a type of the target task; and determining, by the task allocation unit, based on a pre-established correspondence between a task type and a camera, that the target task corresponds to the first camera. ### Claim 11 (depends on 8) The method according to claim 8, wherein the selecting, by the task allocation unit, the first camera from the at least two cameras comprises: determining, by the task allocation unit, area information corresponding to the target task; and selecting, by the task allocation unit, the first camera from the at least two cameras based on the area information. ### Claim 12 (depends on 6) The method according to claim 6, wherein after the obtaining, by the second camera, the second image, and before the determining, by the task allocation unit, the indication information of the target task, the method further comprises: determining, by the second camera, whether the second image meets a preset condition; if the second image meets the preset condition, sending, by the second camera, initiation information of the target task to the task allocation unit, wherein the initiation information indicates to the task allocation unit to send the indication information of the target task; and the determining, by the task allocation unit, the indication information of the target task comprises: determining, by the task allocation unit, the indication information of the target task based on the initiation information of the target task. ### Claim 13 (depends on 12) The method according to claim 12, wherein the determining, by the second camera, whether the second image meets a preset condition comprises: extracting, by the second camera, second characteristic information of the target object from the second image; generating, by the second camera, second similarity information, wherein the second similarity information indicates a similarity between a to-be-recognized object and the target object in the second image; determining, by the second camera, whether the second similarity information meets a similarity condition; and if the second similarity information meets the similarity condition, determining, by the second camera, that the second image meets the preset condition. ### Claim 14 (depends on 1) The method according to claim 1, wherein the first processing result information comprises the first similarity information, the second processing result information comprises a second similarity information, and the second similarity information indicates the similarity between the to-be-recognized object and the target object in the second image; and the determining, by the analysis unit, third execution result information of the target task based on the first execution result information and the second execution result information comprises: separately determining, by the analysis unit, weights of the first similarity information and the second similarity information; and generating, by the analysis unit, based on the first similarity information and a weight of the first similarity information, and the second similarity information and a weight of the second similarity information, third similarity information between the target object and the to-be-recognized object. ### Claim 15 (depends on 14) The method according to claim 14, wherein after the generating, by the analysis unit, based on the first similarity information and the weight of the first similarity information, and the second similarity information and the weight of the second similarity information, the third similarity information between the target object and the to-be-recognized object, the determining third execution result information of the target task based on the first execution result information and the second execution result information further comprises: determining, by the analysis unit, whether the third similarity information meets a similarity condition; and if the third similarity information meets the similarity condition, determining, by the analysis unit, that the to-be-recognized object appears in the video surveillance system. ### Claim 16 A multi-camera collaboration-based video surveillance system, comprising at least two cameras and a processor, wherein the at least two cameras comprise a first camera and a second camera; and the first camera is configured to: obtain a photographed first image; receive indication information of a target task, wherein the target task is image recognition and the indication information of the target task indicates characteristic information of a to-be-recognized image, a similarity condition and a process to be performed on a target object in the photographed first image; generate first execution result information of the target task based on the indication information of the target task, wherein the first execution result information comprises first processing result information related to characteristic information of the target object in the first image; and send the first execution result information to the processor; the second camera is configured to: obtain a photographed second image; receive the indication information of the target task, wherein the indication information of the target task further indicates a process to be performed on the target object in the photographed second image; generate second execution result information of the target task based on the indication information of the target task; wherein the second execution result information comprises second processing result information related to characteristic information of the target object in the second image; and send the second execution result information to the processor; and the processor is configured to determine third execution result information of the target task based on the first execution result information and the second execution result information; wherein the first camera is further configured to: extract first characteristic information of the target object from the first image, generate, based on the first characteristic information and the characteristic information of the to-be-recognized image, first similarity information between a to-be-recognized object and the target object in the first image; and determine whether the first similarity information meets the similarity condition. ### Claim 17 (depends on 16) The system according to claim 16, further comprising a third camera, wherein: the third camera comprises the processor, the third camera is different from the first camera and the second camera. ### Claim 18 (depends on 16) The system according to claim 16, wherein: the first camera comprises the processor; or the second camera comprises the processor. ### Claim 19 (depends on 16) The system according to claim 16, wherein: the first execution result information is related to object recognition in the first image; and the second execution result information is related to object recognition in the second image. ### Claim 20 (depends on 16) The system according to claim 16, wherein: the first processing result information is related to similarity between a to-be-recognized object and the target object in the first image; the second processing result information is related to similarity between the to-be-recognized object and the target object in the second image; the third execution result information is related to similarity between the to-be-recognized object and the target object. ### Claim 21 (depends on 16) The system according to claim 16, wherein the processor is configured to receive the indication information of the target task, and send the indication information of the target task to the first camera. ### Claim 22 (depends on 21) The system according to claim 21, wherein the first camera is configured to: after receiving the indication information of the target task, and before generating the first execution result information of the target task based on the indication information of the target task, determine whether to accept the target task, and in response to determining to accept the target task, perform the step of generating the first execution result information of the target task based on the indication information of the target task. ### Claim 23 (depends on 21) The system according to claim 21, wherein the processor is configured to: after receiving the indication information of the target task, select the first camera from the at least two cameras. ### Claim 24 (depends on 23) The system according to claim 23, wherein the processor is configured to: receive resource information sent by the at least two cameras; and select the first camera from the at least two cameras based on the resource information. ### Claim 25 (depends on 23) The system according to claim 23, wherein the processor is further configured to: determine a type of the target task, and determine, based on a pre-established correspondence between a task type and a camera, that the target task corresponds to the first camera. ### Claim 26 (depends on 23) The system according to claim 23, wherein the processor is further configured to: determine area information corresponding to the target task, and select the first camera from the at least two cameras based on the area information. ### Claim 27 (depends on 21) The system according to claim 21, wherein the second camera is configured to: after obtaining the second image, determine whether the second image meets a preset condition, and if the second image meets the preset condition, send initiation information of the target task to the processor, wherein the initiation information indicates to the processor to send the indication information of the target task; and the processor is further configured to receive the indication information of the target task based on the initiation information of the target task. ### Claim 28 (depends on 27) The system according to claim 27, wherein the second camera is further configured to: extract second characteristic information of the target object from the second image; generate second similarity information, wherein the second similarity information indicates a similarity between a to-be-recognized object and the target object in the second image; determine whether the second similarity information meets a similarity condition; and if the second similarity information meets the similarity condition, determine that the second image meets the preset condition. ### Claim 29 (depends on 16) The system according to claim 16, wherein the first processing result information comprises the first similarity information, the second processing result information comprises a second similarity information, and the second similarity information indicates the similarity between the to-be-recognized object and the target object in the second image; and the processor is further configured to: separately determine weights of the first similarity information and the second similarity information; and generate, based on the first similarity information and a weight of the first similarity information, and the second similarity information and a weight of the second similarity information, third similarity information between the target object and the to-be-recognized object. ### Claim 30 (depends on 29) The system according to claim 29, wherein the processor is further configured to: after generating, based on the first similarity information and the weight of the first similarity information, and the second similarity information and the weight of the second similarity information, the third similarity information between the target object and the to-be-recognized object, determine whether the third similarity information meets a similarity condition, and if the third similarity information meets the similarity condition, determine that the to-be-recognized object appears in the video surveillance system.