# US 2021/0240982 A1 · Image Recognition Method and Terminal > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2021/0240982 A1 - Document ID: US20210240982A1 - Status: Published application - Kind: A1 - Application number: 17/240,103 - Filed: 2021-04-26 - Published: 2021-08-05 - Earliest claimed: 2019-01-30 - Assignees: Huawei Technologies Co., Ltd. - Applicants: Huawei Technologies Co., Ltd. - Inventors: Changzhu Li; Xiyong Wang - Web page: https://patentgrep.com/patent/US20210240982A1 ## CPC - G06V 20/70: Labelling scene content, e.g. deriving syntactic or semantic representations - G06F 18/214: Generating training patterns; Bootstrap methods, e.g. bagging or boosting - G06F 18/24: Classification techniques - G06F 18/40: Software arrangements specially adapted for pattern recognition, e.g. user interfaces or toolboxes therefor - G06N 3/0464: Convolutional networks [CNN, ConvNet] - G06N 3/09: Supervised learning - G06V 10/764: using classification, e.g. of video objects - G06V 10/82: using neural networks - G06V 10/95: {structured as a network, e.g. client-server architectures} - G06V 20/10: Terrestrial scenes (scenes under surveillance with static cameras G06V20/52; scenes perceived from the exterior of a vehicle G06V20/56; scenes perceived from the interior of a vehicle G06V20/59) - G06F 3/04842: Selection of displayed objects or displayed text elements (G06F3/0482 takes precedence) - G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS ## Abstract An image recognition method and a terminal, where the method includes obtaining, by the terminal, an image file comprising a target object, recognizing, by the terminal, the target object based on an image recognition model in the terminal to obtain object category information of the target object, and storing, by the terminal, the object category information as first label information of the target object. Hence, image recognition efficiency of the terminal can be improved, and privacy of a terminal user can be effectively protected. ## Claims ### Claim 1 A terminal, comprising: a processor configured to: obtain a first image file comprising a target object; and recognize the target object based on an image recognition model in the terminal to obtain object category information of the target object; and a storage coupled to the processor and configured to store the object category information as first label information of the target object. ### Claim 2 (depends on 1) The terminal of claim 1, wherein the processor is further configured to: receive, from a user, a selection operation on a display image on the terminal, determine, based on the selection operation, the target object corresponding to the selection operation; and generate the first image file comprising the target object, wherein when the selection operation corresponds to n target objects and when n is a positive integer, the processor is further configured to: generate a second image file comprising the n target objects; or generate m image files comprising at least one of the n target objects, wherein m is an integer, and wherein m≤n. ### Claim 3 (depends on 1) The terminal of claim 1, further comprising a receiver coupled to the processor, wherein before recognizing the target object based on the image recognition model in the terminal, either: the receiver is configured to receive and the processor is configured to store the image recognition model from a server; or the processor is further configured to perform image recognition training based on a picture in the terminal to obtain the image recognition model. ### Claim 4 (depends on 3) The terminal of claim 3, further comprising a transmitter coupled to the processor and configured to send the first label information to the server to share the first label information on the server. ### Claim 5 (depends on 4) The terminal of claim 4, wherein before sending the first label information to the server, the processor is further configured to revise the first label information based on a revision operation received from a user for the first label information, and wherein the transmitter is further configured to send revised first label information to the server. ### Claim 6 (depends on 4) The terminal of claim 4, wherein the receiver is further configured to receive first label revision information from the server, and wherein the processor is further configured to update the first label information with the first label revision information. ### Claim 7 (depends on 1) The terminal of claim 1, wherein the processor is further configured to: search a target object detail base for second label information matching the first label information; and determine whether the second label information is found, wherein the storage is further configured to: store the second label information and label the target object based on the second label information when the second label information is found; and label the target object based on the first label information when the second label information is not found. ### Claim 8 (depends on 1) The terminal of claim 1, wherein the image recognition model is a convolutional neural network model. ### Claim 9 An image recognition method implemented by a terminal, wherein the image recognition method comprises: obtaining a first image file comprising a target object; recognizing the target object based on an image recognition model in the terminal to obtain object category information of the target object; and storing the object category information as first label information of the target object. ### Claim 10 (depends on 9) The image recognition method of claim 9, wherein obtaining the first image file comprises: receiving, from a user, a selection operation on a display image on the terminal; determining, based on the selection operation, the target object corresponding to the selection operation; and generating the first image file comprising the target object, wherein when the selection operation corresponds to n target objects and when n is a positive integer, generating the first image file comprises: generating a second image file comprising the n target objects; or generating m image files comprising at least one of the n target objects, wherein m is an integer, and wherein m≤n. ### Claim 11 (depends on 9) The image recognition method of claim 9, wherein before recognizing the target object, the image recognition method further comprises: receiving and storing the image recognition model from a server; or performing image recognition training based on a picture in the terminal to obtain the image recognition model, and wherein after storing the object category information, the image recognition method further comprises sending the first label information to the server to share the first label information on the server. ### Claim 12 (depends on 11) The image recognition method of claim 11, wherein before sending the first label information to the server, the image recognition method further comprises revising the first label information based on a revision operation received from a user for the first label information, and wherein sending the first label information to the server comprises sending revised first label information to the server. ### Claim 13 (depends on 11) The image recognition method of claim 11, further comprising: receiving first label revision information from the server; and updating the first label information with the first label revision information. ### Claim 14 (depends on 9) The image recognition method of claim 9, further comprising: searching a target object detail base for second label information matching the first label information; determining whether the second label information is found; storing the second label information and labeling the target object based on the second label information when the second label information is found; and labeling the target object based on the first label information when the second label information is not found. ### Claim 15 (depends on 9) The image recognition method of claim 9, wherein the object category information comprises text information of the target object. ### Claim 16 (depends on 9) The image recognition method of claim 9, wherein the object category information comprises an object name of the target object. ### Claim 17 (depends on 9) The image recognition method of claim 9, wherein the object category information is stored in an image file as the first label information. ### Claim 18 (depends on 9) The image recognition method of claim 9, wherein the object category information is stored in an image file based on a preset label information format. ### Claim 19 (depends on 9) The image recognition method of claim 9, wherein the first label information is stored in an extension segment of an image file. ### Claim 20 (depends on 9) The image recognition method of claim 9, wherein the first label information is stored in a comment segment of an image file.