# US 11,908,242 B2 · Efficient calculation of a robust signature of a media unit > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 11,908,242 B2 - Document ID: US11908242B2 - Status: Granted - Kind: B2 - Application number: 16/835,368 - Filed: 2020-03-31 - Published: 2020-10-01 - Granted: 2024-02-20 - Earliest claimed: 2019-03-31 - Assignees: Cortica, Ltd. - Applicants: Cortica, Ltd. - Inventors: Igal Raichelgauz - Web page: https://patentgrep.com/patent/US11908242B2 ## CPC - G06V 40/23: {Recognition of whole body movements, e.g. for sport training} - 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) - G06V 40/10: Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands - G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) ## Abstract Systems, and method and computer readable media that store instructions for calculating signatures, utilizing signatures and the like. ## Claims ### Claim 1 A method for object detection, the method comprises: sensing, by a visual light camera, an input image; wherein the sensing comprises generating pixels that represent the input image; receiving, by an object detector the input image; generating, by a signature generator, a signature of the input image; wherein the generating of the signatures comprises: (i) applying multiple image processing iterations on the input image to provide a first number of outputs values, wherein each imaging processing iteration of the multiple image processing iterations comprises an expansion operations that is followed by a merge operation, the expansion operation of the imaging processing iteration comprises performing the expansion operation by selected spanning elements that are associated with the imaging processing iteration, while reducing power consumption of other spanning elements that are not associated with the imaging processing iteration, wherein the reducing of the power consumption comprises shutting down the other spanning elements or entering the other spanning elements to an idle mode; and (ii) selecting, out of the first number of output values, a second number of selected output values, wherein the first number exceeds the second number by a factor that exceeds one hundred; wherein the signature of the input image comprises multiple matching identifiers that are multiple indexes for retrieval of the second number of selected output values; comparing, by an object detection determination unit, the signature of the input image to signatures of a concept structure; determining, by the object detection determination unit, whether the signature of the input image matches any of the signatures of the concept structure based on a signature matching criteria, wherein each signature of the concept structure is associated within a signature matching criterion that is determined based on an object detection parameter of the signature of the concept structure; wherein the signature matching criteria is adjusted to statistics of appearances of identifiers in images acquired under different scenarios; and concluding, by the object detection determination unit, that the input image comprises an object associated with the concept structure based on an outcome of the determining; wherein the signature generator and the object detection determination units comprise one or more processing circuitries. ### Claim 2 (depends on 1) The method according to claim 1 wherein each signature matching criterion is determined by evaluating object detection capabilities of the signature of the concept structure under different signature matching criteria. ### Claim 3 (depends on 2) The method according to claim 2 wherein the evaluating of the object detection capabilities of the signature of the concept structure under different signature matching criteria comprises: receiving or generating signatures of a group of test images; calculating the object detection capability of the signature of the concept structure, for each signature matching criterion of the different signature matching criteria; and selecting the signature matching criterion based on the object detection capabilities of the signature of the concept structure under the different signature matching criteria. ### Claim 4 (depends on 3) The method according to claim 3 wherein the object detection capability reflects a percent of signatures of the group of test images that match the signature of the concept structure. ### Claim 5 (depends on 4) The method according to claim 4 wherein the selecting of the signature matching criterion comprises selecting the signature matching criterion that one applied results in a percent of signatures of the group of test images that match the signature that is closets to a predefined desired percent of signatures of the group of test images that match the signature of the concept structure. ### Claim 6 (depends on 1) The method according to claim 1 wherein the signature matching criteria is a minimal number of matching identifiers that are indicative of a match. ### Claim 7 (depends on 1) The method according to claim 1, wherein each imaging processing iteration of the multiple image processing iterations comprises determining which are the selected spanning elements that are associated with the imaging processing iteration. ### Claim 8 (depends on 7) The method according to claim 7, each spanning element is associated with a unique set of one or more numbers, wherein during an image processing iteration that is preceded by a previous image processing iteration the selecting is based on a comparison between the unique sets of the spanning elements to an outcome of the previous image processing iteration. ### Claim 9 A non-transitory computer readable medium for object detection, the non-transitory computer readable medium that stores instructions which when executed by processing circuits, cause the processing circuits to perform operations comprising: sensing, by a visual light camera, an input image; wherein the sensing comprises generating pixels that represent the input image; receiving the input image by the processing circuits; generating a signature of the input image; wherein the generating of the signatures comprises: (i) applying multiple image processing iterations on the input image to provide a first number of outputs values, wherein each imaging processing iteration of the multiple image processing iterations comprises an expansion operations that is followed by a merge operation, the expansion operation of the imaging processing iteration comprises performing the expansion operation by selected spanning elements that are associated with the imaging processing iteration, while reducing power consumption of other spanning elements that are not associated with the imaging processing iteration, wherein the reducing of the power consumption comprises shutting down the other spanning elements or entering the other spanning elements to an idle mode; and (ii) selecting, out of the first number of output values, a second number of selected output values, wherein the first number exceeds the second number by a factor that exceeds one hundred; wherein the signature of the input image comprises multiple matching identifiers that are multiple indexes for retrieval of the second number of selected output values; comparing the signature of the input image to signatures of a concept structure; determining whether the signature of the input image matches any of the signatures of the concept structure based on a signature matching criteria, wherein each signature of the concept structure is associated within a signature matching criterion that is determined based on an object detection parameter of the signature of the concept structure; wherein the signature matching criteria is adjusted to statistics of appearances of identifiers in images acquired under different scenarios; and concluding that the input image comprises an object associated with the concept structure based on an outcome of the determining. ### Claim 10 (depends on 9) The non-transitory computer readable medium according to claim 9, wherein each signature matching criterion is determined by evaluating object detection capabilities of the signature of the concept structure under different signature matching criteria. ### Claim 11 (depends on 10) The non-transitory computer readable medium according to claim 10, wherein the evaluating of the object detection capabilities of the signature of the concept structure under different signature matching criteria comprises: receiving or generating signatures of a group of test images; calculating the object detection capability of the signature of the concept structure, for each signature matching criterion of the different signature matching criteria; and selecting the signature matching criterion based on the object detection capabilities of the signature of the concept structure under the different signature matching criteria. ### Claim 12 (depends on 11) The non-transitory computer readable medium according to claim 11 wherein the object detection capability reflects a percent of signatures of the group of test images that match the signature of the concept structure. ### Claim 13 (depends on 12) The non-transitory computer readable medium according to claim 12 wherein the selecting of the signature matching criterion comprises selecting the signature matching criterion that one applied results in a percent of signatures of the group of test images that match the signature that is closets to a predefined desired percent of signatures of the group of test images that match the signature of the concept structure. ### Claim 14 (depends on 11) The non-transitory computer readable medium according to claim 11 wherein the object detection capability reflects a percent of signatures of the group of test images that match the signature of the concept structure. ### Claim 15 (depends on 12) The non-transitory computer readable medium according to claim 12 wherein the selecting of the signature matching criterion comprises selecting the signature matching criterion that one applied results in a percent of signatures of the group of test images that match the signature that is closets to a predefined desired percent of signatures of the group of test images that match the signature of the concept structure. ### Claim 16 (depends on 9) The non-transitory computer readable medium according to claim 9 wherein the signature matching criteria is a minimal number of matching identifiers that are indicative of a match. ### Claim 17 (depends on 9) The non-transitory computer readable medium according to claim 9, wherein each signature matching criterion is determined by evaluating object detection capabilities of the signature of the concept structure under different signature matching criteria. ### Claim 18 (depends on 10) The non-transitory computer readable medium according to claim 10, wherein the evaluating of the object detection capabilities of the signature of the concept structure under different signature matching criteria comprises: receiving or generating signatures of a group of test images; calculating the object detection capability of the signature of the concept structure, for each signature matching criterion of the different signature matching criteria; and selecting the signature matching criterion based on the object detection capabilities of the signature under the different signature matching criteria. ### Claim 19 (depends on 9) The non-transitory computer readable medium according to claim 9 wherein the matching identifiers are indexes for retrieval of values that are outputs of multiple input image processing iterations. ### Claim 20 A non-transitory computer readable medium for object detection, the non-transitory computer readable medium that stores instructions which when executed by processing circuits, cause the processing circuits to perform operations comprising: sensing, by a visual light camera, an input image; wherein the sensing comprises generating pixels that represent the input image; receiving the input image by the processing circuits; generating a signature of the input image; wherein the generating of the signatures comprises (i) applying multiple image processing iterations on the input image to provide a first number of outputs values, wherein each imaging processing iteration of the multiple image processing iterations comprises an expansion operations that is followed by a merge operation, the expansion operation of the imaging processing iteration comprises performing the expansion operation by selected spanning elements that are associated with the imaging processing iteration, while reducing power consumption of other spanning elements that are not associated with the imaging processing iteration, wherein the reducing of the power consumption comprises shutting down the other spanning elements or entering the other spanning elements to an idle mode; and (ii) selecting, out of the first number of output values, a second number of selected output values, wherein the first number exceeds the second number by a factor that exceeds one hundred; wherein the signature of the input image comprises multiple matching identifiers that are multiple indexes for retrieval of the second number of selected output values; comparing the signature of the input image to signatures of a concept structure; determining whether the signature of the input image matches any of the signatures of the concept structure based on signature matching criteria, wherein each signature of the concept structure is associated within a signature matching criterion that (i) is determined based on an object detection parameter of the signature of the concept structure, and (ii) is adjusted to statistics of appearances of identifiers in images acquired under different scenarios; wherein the signature matching criteria is a minimal number of matching identifiers that are indicative of a match; and concluding that the input image comprises an object associated with the concept structure based on an outcome of the determining.