# US 12,190,692 B2 · Monitoring device, suspicious object detecting method, and recording medium > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,190,692 B2 - Document ID: US12190692B2 - Status: Granted - Kind: B2 - Application number: 18/235,745 - Filed: 2023-08-18 - Published: 2023-12-07 - Granted: 2025-01-07 - Earliest claimed: 2019-03-27 - Assignees: NEC Corporation - Applicants: NEC Corporation - Inventors: Ryoma Oami - Web page: https://patentgrep.com/patent/US12190692B2 ## CPC - G06T 7/246: using feature-based methods, e.g. the tracking of corners or segments - G08B 13/19613: {Recognition of a predetermined image pattern or behaviour pattern indicating theft or intrusion} - G06T 7/70: Determining position or orientation of objects or cameras (camera calibration G06T7/80) - G06V 10/768: {using context analysis, e.g. recognition aided by known co-occurring patterns} - G06V 20/52: Surveillance or monitoring of activities, e.g. for recognising suspicious objects (recognising microscopic objects G06V20/69) - G06V 40/172: {Classification, e.g. identification} - G06V 40/197: {Matching; Classification} - G06V 40/25: {Recognition of walking or running movements, e.g. gait recognition} - G08B 13/19608: {Tracking movement of a target, e.g. by detecting an object predefined as a target, using target direction and or velocity to predict its new position} - G06T 2207/10016: Video; Image sequence - G06T 2207/10024: Color image - G06T 2207/30196: Human being; Person - G06T 2207/30201: Face - G06T 2207/30232: Surveillance - G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL - G08B: SIGNALLING SYSTEMS, e.g. PERSONAL CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS - G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING ## Abstract A monitoring device and the like are provided which are capable of detecting an attribute change in a suspicious object that cannot be determined from the behavior of the object. An associating unit associates, among a plurality of objects detected from time-series image data, identical objects with one another. An attribute change detecting unit detects from the time-series image data a change in an attribute of at least one of the identical objects and an attendant item. A suspicious object detecting unit detects a suspicious object on the basis of the change in attribute. ## Claims ### Claim 1 A monitoring device comprising: a memory storing instructions; and at least one processor coupled to the memory and configured to execute the instructions to: detect multiple objects from time-series image data; associate the multiple objects that are a same object with one another by determining sameness; calculate a reliability of a determination result of the sameness, the reliability indicating a certainty of the sameness; detect an attribute change of at least one of the same object and an accessory of the same object from the time-series image data; determine authenticity of the attribute change based on the reliability; and determine whether the same object is a suspicious object according to a determination result of the authenticity, wherein determination of the authenticity includes specifying a context in the time-series image data in which the attribute change has been detected, determination of whether the same object is the suspicious object includes detecting the same object as the suspicious object when a combination of the detected attribute change and the specified context is not included in exception information indicating that the attribute change is not suspicious in the context, and the exception information includes information associating the context indicating a locker room with the attribute change indicating a change in clothing. ### Claim 2 (depends on 1) The monitoring device according to claim 1, wherein the at least one processor is configured to execute the instructions to further: determine the authenticity of the attribute change based on the reliability of the determination result of the sameness in predetermined time-series image data, the predetermined time-series image data including first image data in which the attribute change is detected and image data temporally before and after the first image data. ### Claim 3 (depends on 1) The monitoring device according to claim 1, wherein the at least one processor is configured to execute the instructions to further: when the attribute change is determined to be false, remove a detection result of the attribute change. ### Claim 4 (depends on 1) The monitoring device according to claim 1, wherein the at least one processor is configured to execute the instructions to further: when the attribute change is determined to be true, determine the same object as the suspicious object. ### Claim 5 (depends on 1) The monitoring device according to claim 1, wherein the reliability depends on a distribution situation of other objects in the time-series image data including a target object of determination of the sameness, the other objects being different from the target object. ### Claim 6 (depends on 1) The monitoring device according to claim 1, wherein the at least one processor is configured to execute the instructions to further: extract attribute information for each of multiple objects from the time-series image data; and determine the sameness based on the extracted attribute information. ### Claim 7 (depends on 6) The monitoring device according to claim 6, wherein the attribute information is information regarding at least one of a face, an iris, gait and a hairstyle of person. ### Claim 8 A monitoring method performed by a computer and comprising: detecting multiple objects from time-series image data; associating the multiple objects that are a same object with one another by determining sameness; calculating a reliability of a determination result of the sameness, the reliability indicating a certainty of the sameness; detecting an attribute change of at least one of the same object and an accessory of the same object from the time-series image data; determining authenticity of the attribute change based on the reliability; and determining whether the same object is a suspicious object according to a determination result of the authenticity, wherein determination of the authenticity includes specifying a context in the time-series image data in which the attribute change has been detected, determination of whether the same object is the suspicious object includes detecting the same object as the suspicious object when a combination of the detected attribute change and the specified context is not included in exception information indicating that the attribute change is not suspicious in the context, and the exception information includes information associating the context indicating a locker room with the attribute change indicating a change in clothing. ### Claim 9 (depends on 8) The monitoring method according to claim 8, further comprising: determining the authenticity of the attribute change based on the reliability of the determination result of the sameness in predetermined time-series image data, the predetermined time-series image data including first image data in which the attribute change is detected and image data temporally before and after the first image data. ### Claim 10 (depends on 8) The monitoring method according to claim 8, further comprising: when the attribute change is determined to be false, removing a detection result of the attribute change. ### Claim 11 (depends on 8) The monitoring method according to claim 8, further comprising: when the attribute change is determined to be true, determining the same object as the suspicious object. ### Claim 12 (depends on 8) The monitoring method according to claim 8, wherein the reliability depends on a distribution situation of other objects in the time-series image data including a target object of determination of the sameness, the other objects being different from the target object. ### Claim 13 (depends on 8) The monitoring method according to claim 8, further comprising: extracting attribute information for each of multiple objects from the time-series image data; and determining the sameness based on the extracted attribute information. ### Claim 14 (depends on 13) The monitoring method according to claim 13, wherein the attribute information is information regarding at least one of a face, an iris, gait and a hairstyle of person. ### Claim 15 A non-transitory computer-readable recording medium storing a program for causing a computer to execute processing comprising: detecting multiple objects from time-series image data; associating the multiple objects that are a same object with one another by determining sameness; calculating a reliability of a determination result of the sameness, the reliability indicating a certainty of the sameness; detecting an attribute change of at least one of the same object and an accessory of the same object from the time-series image data; determining authenticity of the attribute change based on the reliability; and determining whether the same object is a suspicious object according to a determination result of the authenticity, wherein determination of the authenticity includes specifying a context in the time-series image data in which the attribute change has been detected, determination of whether the same object is the suspicious object includes detecting the same object as the suspicious object when a combination of the detected attribute change and the specified context is not included in exception information indicating that the attribute change is not suspicious in the context, and the exception information includes information associating the context indicating a locker room with the attribute change indicating a change in clothing. ### Claim 16 (depends on 15) The non-transitory computer-readable recording medium according to claim 15, wherein the processing further comprises: determining the authenticity of the attribute change based on the reliability of the determination result of the sameness in predetermined time-series image data, the predetermined time-series image data including first image data in which the attribute change is detected and image data temporally before and after the first image data. ### Claim 17 (depends on 15) The non-transitory computer-readable recording medium according to claim 15, wherein the processing further comprises: when the attribute change is determined to be false, removing a detection result of the attribute change. ### Claim 18 (depends on 15) The non-transitory computer-readable recording medium according to claim 15, wherein the processing further comprises: when the attribute change is determined to be true, determining the same object as the suspicious object. ### Claim 19 (depends on 15) The non-transitory computer-readable recording medium according to claim 15, wherein the reliability depends on a distribution situation of other objects in the time-series image data including a target object of determination of the sameness, the other objects being different from the target object. ### Claim 20 (depends on 15) The non-transitory computer-readable recording medium according to claim 15, wherein the processing further comprises: extracting attribute information for each of multiple objects from the time-series image data; and determining the sameness based on the extracted attribute information.