Target identification method based on abstract algorithm feature encryption, target matching method and face recognition method

Abstract

The present application belongs to the field of target identification, and particularly relates to a target identification method based on abstract algorithm feature encryption, a target matching method and a face recognition method, aiming at solving the security hidden trouble problem caused by the readability of sample data in the target identification technology. The present application target identification method comprises: obtaining sampling information of a target to be identified; performing feature extraction based on the sampling information of the target to be identified, and obtaining a sampling information feature abstract by processing the extracted features through an abstract algorithm; matching the sampling information feature abstract with a pre-stored sample information feature abstract to obtain label information of the target to be identified; and the label information of the target to be identified is the label information corresponding to the matching successful sample information feature abstract. The present application is based on the non-reversibility of abstract data, so that even if the data is leaked, a third party cannot read the target feature information represented by the abstract data, thereby ensuring the safety of user data.

Description

Technical Field

[0001] This invention belongs to the field of target recognition, and specifically relates to a target recognition method, a target matching method, and a face recognition method based on feature encryption using a digest algorithm. Background Technology

[0002] Target recognition, as the name suggests, is the process of obtaining target labels based on target detection information. Common target recognition methods include facial recognition, voice recognition, fingerprint recognition, and other methods that identify personal tags (such as identity, permissions, etc.) based on human physiological characteristics. It also includes the identification of tags for other objects such as vehicles and plants, as well as tags for articles and events.

[0003] Currently, a common approach in target recognition is to match the sampled target data with pre-stored samples or sample data to obtain the target's label information. However, with the increasing development of the internet and the growing exposure of risks related to the leakage of private and confidential data in sample data—such as the data breaches in the facial recognition field and the resulting user information security issues—it's crucial that malicious actors can use users' facial images to replace images in videos or open accounts, causing irreparable damage. Summary of the Invention

[0004] To address the aforementioned problems in existing technologies, namely the security risks arising from the readability of sample data in target recognition technologies, this invention provides:

[0005] Solution 1: A target recognition method based on feature encryption using a digest algorithm, comprising the following steps: Obtain sampling information of the target to be identified; Feature extraction is performed based on the sampling information of the target to be identified, and the extracted features are processed by a summarization algorithm to obtain a feature summary of the sampling information; The sampled information feature summary is matched with the pre-stored sample information feature summary to obtain the label information of the target to be identified; the label information of the target to be identified is the label information corresponding to the successfully matched sample information feature summary. Each feature summary in the sampling information feature summary includes a feature identifier and summary data; The feature identifier is a category identifier, or an ordered arrangement of multiple category identifiers; the summary data is obtained by processing the feature data corresponding to one category through a summary algorithm, or by processing the feature data corresponding to multiple categories in an ordered arrangement through a summary algorithm.

[0006] Option 2: According to the target recognition method based on feature encryption using a digest algorithm described in Option 1, the method for obtaining the sample information feature digest is as follows: The feature information of the sample is processed by a summarization algorithm to obtain a feature summary of the sample information; The feature information of the sample is obtained by feature extraction based on the sample sampling information, or by input through the human-computer interaction port, or by retrieval from the sample feature information storage unit.

[0007] Scheme 3: According to the target recognition method based on feature encryption of digest algorithm described in Scheme 1, the feature category corresponding to the sample information feature digest includes the feature category corresponding to the sample information feature digest.

[0008] Option 4, based on the target recognition method using feature encryption based on digest algorithm described in Option 3, involves "matching the feature digest of the sampled information with the pre-stored feature digest of the sampled information," the method of which is as follows: The sampling information feature summary is matched individually with one or more sample information feature summaries to obtain matching degree information; If there is only one sample information feature summary, then if the matching degree information is greater than the preset first matching degree threshold, it is determined that the match is successful; if there are multiple sample information feature summaries, then the largest sample information feature summary with matching degree information greater than the preset second matching degree threshold is selected as the successfully matched sample information feature summary.

[0009] Option 5: According to the target recognition method based on digest algorithm feature encryption described in Option 3, the matching degree information is... ; Where A represents the matching degree information, a represents the number of successfully matched feature summaries in the sampled information feature summary, and b represents the number of feature summaries contained in the sampled information feature summary.

[0010] Scheme 6: According to the target recognition method based on feature encryption of digest algorithm described in Scheme 1, the features of the measurement data type extracted from the sampling information are used to obtain the hierarchical information corresponding to the features of the measurement data according to the preset hierarchical principle, and the feature digest is obtained by digest algorithm based on the hierarchical information.

[0011] Scheme 7: According to the target recognition method based on feature encryption using digest algorithm described in Scheme 1, "the extracted features are processed by digest algorithm to obtain a feature digest of the sampled information", and the method is Method 1, Method 2 or Method 3; The first method is as follows: for each feature extracted based on the sampling information, a corresponding feature summary is obtained by processing it through a summarization algorithm, and a sampling information feature summary is constructed. The second method is as follows: according to the feature combination corresponding to one or more feature summaries in the sample information feature summary, the features extracted from the sample information are combined, and after combination, the corresponding feature summary is obtained by the summarization algorithm to construct the sample information feature summary. The third method is as follows: For multiple features extracted based on sampling information, the features are combined according to a preset or randomly generated feature combination scheme, and after combination, the corresponding feature summary is obtained by processing through a summarization algorithm to construct the sampling information feature summary.

[0012] Scheme 8: According to the target recognition method based on feature encryption of digest algorithm described in Scheme 7, when constructing the feature digest of sampled information using the method three, the pre-stored feature digest of sampled information is combined according to the combination scheme in the method three and then matched with the sampled information.

[0013] Scheme 9: The target identification method based on digest algorithm feature encryption as described in Scheme 1, after obtaining the tag information of the target to be identified, further includes: Delete the sampling information of the target to be identified.

[0014] Scheme 10: A target matching method based on feature encryption using a digest algorithm, comprising the following steps: Feature extraction is performed on the sampled information of the target to be identified, and the extracted features are processed by a summarization algorithm to obtain a feature summary of the sampled information; The sampled information feature summary is matched with the pre-stored sample information feature summary to obtain the matching result.

[0015] Solution 11: According to the target matching method based on feature encryption using a digest algorithm described in Solution 10, the method for obtaining the sample information feature digest is as follows: The feature information of the sample is processed by a summarization algorithm to obtain a feature summary of the sample information; The feature information of the sample is obtained by feature extraction based on the sample sampling information, or by input through the human-computer interaction port, or by retrieval from the sample feature information storage unit.

[0016] Scheme 12: According to the target matching method based on digest algorithm feature encryption described in Scheme 10, the feature category corresponding to the sample information feature digest includes the feature category corresponding to the sample information feature digest.

[0017] Scheme 13: According to the target matching method based on digest algorithm feature encryption described in Scheme 12, "the feature digest of the sampled information is matched with the pre-stored feature digest of the sampled information," and the method is as follows: The sampling information feature summary is matched individually with one or more sample information feature summaries to obtain matching degree information; If there is only one sample information feature summary, then if the matching degree information is greater than the preset first matching degree threshold, it is determined that the match is successful; if there are multiple sample information feature summaries, then the largest sample information feature summary with matching degree information greater than the preset second matching degree threshold is selected as the successfully matched sample information feature summary.

[0018] Scheme 14: The target matching method based on digest algorithm feature encryption as described in Scheme 12, wherein the matching degree information is... ; Where A represents the matching degree information, a represents the number of successfully matched feature summaries in the sampled information feature summary, and b represents the number of feature summaries contained in the sampled information feature summary.

[0019] Scheme 15: According to the target matching method based on feature encryption of digest algorithm described in Scheme 10, the features of the measurement data type extracted from the sampling information are used to obtain the hierarchical information corresponding to the features of the measurement data according to the preset hierarchical principle, and the feature digest is obtained by digest algorithm based on the hierarchical information.

[0020] Scheme 16: According to the target matching method based on feature encryption using digest algorithm described in Scheme 10, "the extracted features are processed by digest algorithm to obtain a feature digest of the sampled information", and the method is Method 1, Method 2, or Method 3; The first method is as follows: for each feature extracted based on the sampling information, a corresponding feature summary is obtained by processing it through a summarization algorithm, and a sampling information feature summary is constructed. The second method is as follows: according to the feature combination corresponding to one or more feature summaries in the sample information feature summary, the features extracted from the sample information are combined, and after combination, the corresponding feature summary is obtained by the summarization algorithm to construct the sample information feature summary. The third method is as follows: For multiple features extracted based on sampling information, the features are combined according to a preset or randomly generated feature combination scheme, and after combination, the corresponding feature summary is obtained by processing through a summarization algorithm to construct the sampling information feature summary.

[0021] Scheme 17: According to the target matching method based on feature encryption of digest algorithm described in Scheme 16, when constructing the feature digest of sampled information using the third method, the pre-stored feature digest of sampled information is combined according to the combination scheme in the third method and then the two are matched.

[0022] Scheme 18: According to the target matching method based on feature encryption of digest algorithm described in Scheme 10, each feature digest in the sampled information feature digest includes feature identifier and digest data.

[0023] Scheme 19: According to the target matching method based on digest algorithm feature encryption described in Scheme 18, the feature identifier is a category identifier or an ordered arrangement of multiple category identifiers; The summary data is obtained by processing feature data corresponding to one category using a summary algorithm, or by processing feature data corresponding to multiple categories arranged in an ordered manner using a summary algorithm.

[0024] Solution 20: A face recognition method based on feature encryption using a digest algorithm, the method comprising: Obtain the facial image of the person to be identified; Feature extraction is performed on the facial images of the person to be identified, and the extracted features are processed by a summarization algorithm to obtain a feature summary of the sampled information; The sampled information feature summary is matched with the pre-stored sample information feature summary to obtain the tag information of the person to be identified; the tag information of the person to be identified is the person information corresponding to the successfully matched sample information feature summary.

[0025] Solution 21: A voice recognition method based on feature encryption using a digest algorithm, the method comprising: Obtain sampling information of the sound to be identified; Feature extraction is performed on the sampled sound information to be recognized, and the extracted features are processed by a summarization algorithm to obtain a feature summary of the sampled information; The sampled information feature summary is matched with the pre-stored sample information feature summary to obtain the tag information of the sound to be identified; the tag information of the sound to be identified is the tag information corresponding to the successfully matched sample information feature summary.

[0026] Scheme 22: A pupil print recognition method based on feature encryption using a digest algorithm, the method comprising: Obtain the iris print information of the person to be identified; Feature extraction is performed on the pupil print information of the person to be identified, and the extracted features are processed by a summarization algorithm to obtain a feature summary of the sampled information; The sampled information feature summary is matched with the pre-stored sample information feature summary to obtain the tag information of the person to be identified; the tag information of the person to be identified is the person information corresponding to the successfully matched sample information feature summary.

[0027] Scheme 23: A fingerprint recognition method based on digest algorithm feature encryption, the method comprising: Obtain the fingerprint information of the person to be identified; Feature extraction is performed on the fingerprint information of the person to be identified, and the extracted features are processed by a summarization algorithm to obtain a feature summary of the sampled information; The sampled information feature summary is matched with the pre-stored sample information feature summary to obtain the tag information of the person to be identified; the tag information of the person to be identified is the person information corresponding to the successfully matched sample information feature summary.

[0028] Scheme 24: A vehicle identification method based on feature encryption using a digest algorithm, the method comprising: Obtain sampling information of the vehicle to be identified; Feature extraction is performed on the sampled information of the vehicle to be identified, and the extracted features are processed by a summarization algorithm to obtain a feature summary of the sampled information; The sampled information feature summary is matched with the pre-stored sample information feature summary to obtain the tag information of the vehicle to be identified; the tag information of the vehicle to be identified is the tag information corresponding to the successfully matched sample information feature summary.

[0029] Scheme 25: A plant identification method based on feature encryption using a digest algorithm, the method comprising: Obtain sampling information of the plant to be identified; Feature extraction is performed on the sampling information of the plants to be identified, and the extracted features are processed by a summarization algorithm to obtain a feature summary of the sampling information; The sampled information feature summary is matched with the pre-stored sample information feature summary to obtain the label information of the plant to be identified; the label information of the plant to be identified is the label information corresponding to the successfully matched sample information feature summary.

[0030] Scheme 26: A semantic recognition method based on feature encryption using a digest algorithm, the method comprising: Obtain the semantic features of the information to be identified; The semantic features of the information to be identified are processed by a summarization algorithm to obtain a feature summary of the sampled information. The feature summary of the sampled information is matched with the pre-stored feature summary of the sample information to obtain the label information of the information to be identified; the label information of the information to be identified is the label information corresponding to the successfully matched feature summary of the sample information.

[0031] Scheme 27: A target recognition system based on feature encryption using a digest algorithm, comprising a first unit, a second unit, and a third unit; The first unit is configured to acquire sampling information of the target to be identified; The second unit is configured to extract features based on the sampling information of the target to be identified, and process the extracted features through a summarization algorithm to obtain a feature summary of the sampling information; The third unit is configured to match the feature summary of the sampled information with the pre-stored feature summary of the sampled information to obtain the label information of the target to be identified; the label information of the target to be identified is the label information corresponding to the successfully matched feature summary of the sampled information.

[0032] Scheme 28: The target recognition system based on digest algorithm feature encryption as described in Scheme 27 further includes a fourth unit; The fourth unit is configured to store a sample information feature summary of one or more samples.

[0033] Scheme 29: In the target identification system based on digest algorithm feature encryption as described in Scheme 28, the fourth unit is located in a terminal device and / or a remote server.

[0034] Option 30: A target recognition system based on digest algorithm feature encryption, comprising a server and one or more terminals; the terminals and the server are communicatively connected. The server is configured to extract features based on the sampling information of the target to be identified, and process the extracted features using a summarization algorithm to obtain a feature summary of the sampling information; the feature summary of the sampling information is matched with a pre-stored feature summary of the sample information to obtain the tag information of the target to be identified, and then sent to the terminal. The terminal is configured to acquire sampling information of the target to be identified and upload it to the server; and acquire the tag information of the target to be identified obtained by the server. The label information of the target to be identified is the label information corresponding to the feature summary of the successfully matched sample information.

[0035] Option 31: In the target identification system based on digest algorithm feature encryption as described in Option 30, the terminal is further configured as follows: The sampling information of the sample is obtained, features are extracted, and the extracted features are processed by a summarization algorithm to obtain a sample information feature summary. The obtained sample information feature summary and its tag information are stored locally and / or uploaded to the server.

[0036] Scheme 32: According to the target recognition system based on digest algorithm feature encryption described in Scheme 30, the terminal is further configured to acquire the sampling information of the sample and upload it to the server; The server is further configured to extract features from the sampling information of the samples, process the extracted features using a summarization algorithm to obtain a sample information feature summary, and store the obtained sample information feature summary and its tag information locally and / or send them to the terminal.

[0037] Scheme 33: According to the target recognition system based on feature encryption of digest algorithm described in Scheme 31 or 32, after extracting the feature digest of the sample information, the sampling information of the sample is deleted.

[0038] Solution 34: A server, communicatively connected to one or more terminals, wherein the server is configured to extract features based on sampling information of a target to be identified, and process the extracted features using a summarization algorithm to obtain a sampling information feature summary; match the sampling information feature summary with a pre-stored sample information feature summary to obtain the tag information of the target to be identified, and send it to the terminal; wherein the tag information of the target to be identified is the tag information corresponding to the successfully matched sample information feature summary.

[0039] Scheme 35: A target recognition system based on digest algorithm feature encryption, comprising a server and one or more terminals; the terminals and the server are communicatively connected; The server is configured to acquire a feature summary of the sampled information of the target to be identified, match the feature summary of the sampled information with a pre-stored feature summary of the sampled information to obtain the tag information of the target to be identified, and send it to the terminal. The terminal is configured to acquire sampling information of the target to be identified, extract features based on the sampling information of the target to be identified, process the extracted features through a summarization algorithm to obtain a feature summary of the sampling information, and upload it to the server; and acquire the tag information of the target to be identified obtained by the server. The label information of the target to be identified is the label information corresponding to the feature summary of the successfully matched sample information.

[0040] Solution 36: In the target recognition system based on digest algorithm feature encryption as described in Solution 35, the terminal is further configured as follows: The sampling information of the sample is obtained, features are extracted, and the extracted features are processed by a summarization algorithm to obtain a sample information feature summary. The obtained sample information feature summary and its tag information are stored locally and / or uploaded to the server.

[0041] Scheme 37: According to the target recognition system based on digest algorithm feature encryption as described in Scheme 35, the terminal is further configured to acquire the sampling information of the sample and upload it to the server; The server is further configured to extract features from the sampling information of the samples, process the extracted features using a summarization algorithm to obtain a sample information feature summary, and store the obtained sample information feature summary and its tag information locally and / or send them to the terminal.

[0042] Scheme 38: According to the target recognition system based on feature encryption of digest algorithm described in Scheme 36 or 37, after extracting the feature digest of the sample information, the sampling information of the sample is deleted.

[0043] Solution 39: A server, communicatively connected to one or more terminals, wherein the server is configured to acquire a sample information feature summary of a target to be identified, match the sample information feature summary with a pre-stored sample information feature summary to obtain the tag information of the target to be identified, and send it to the terminal; the tag information of the target to be identified is the tag information corresponding to the successfully matched sample information feature summary.

[0044] Solution 40: A terminal, communicatively connected to a server, wherein the terminal is configured to acquire sampling information of a target to be identified, perform feature extraction based on the sampling information of the target to be identified, process the extracted features through a summarization algorithm to obtain a feature summary of the sampling information, and upload it to the server; acquire the tag information of the target to be identified obtained by the server; wherein the tag information of the target to be identified is the tag information corresponding to the feature summary of the successfully matched sample information.

[0045] Option 41: A target identification system based on digest algorithm feature encryption, comprising a server and one or more terminals; the terminals and the server are communicatively connected. The server is configured to store feature summaries of sample information for multiple samples; The terminal is configured as follows: Obtain a feature summary of sample information for one or more samples from the server; The sampling information of the target to be identified is obtained, features are extracted based on the sampling information of the target to be identified, and the extracted features are processed by a summarization algorithm to obtain a feature summary of the sampling information. The sampled information feature summary is matched with the pre-stored sample information feature summary to obtain the label information of the target to be identified; The label information of the target to be identified is the label information corresponding to the feature summary of the successfully matched sample information.

[0046] Option 42: In the target identification system based on digest algorithm feature encryption as described in Option 41, the terminal is further configured as follows: The sampling information of the sample is obtained, features are extracted, and the extracted features are processed by a summarization algorithm to obtain a sample information feature summary. The obtained sample information feature summary and its tag information are stored locally and / or uploaded to the server.

[0047] Scheme 43: According to the target recognition system based on digest algorithm feature encryption described in Scheme 41, the terminal is further configured to acquire the sampling information of the sample and upload it to the server; The server is further configured to extract features from the sampling information of the samples, process the extracted features using a summarization algorithm to obtain a sample information feature summary, and store the obtained sample information feature summary and its tag information locally and / or send them to the terminal.

[0048] Scheme 44: According to Scheme 42 or 43, the target recognition system based on feature encryption using a digest algorithm deletes the sampling information of the sample after extracting the feature digest of the sample information.

[0049] Option 45: A server, communicatively connected to one or more terminals, the server being configured to store sample information feature summaries of multiple samples.

[0050] Option 46: A terminal, communicatively connected to a server, wherein the terminal is configured as follows: Obtain a feature summary of sample information for one or more samples from the server; The sampling information of the target to be identified is obtained, features are extracted based on the sampling information of the target to be identified, and the extracted features are processed by a summarization algorithm to obtain a feature summary of the sampling information. The sampled information feature summary is matched with the pre-stored sample information feature summary to obtain the label information of the target to be identified; The label information of the target to be identified is the label information corresponding to the feature summary of the successfully matched sample information.

[0051] Solution 47: A target recognition system based on digest algorithm feature encryption, comprising one or more terminals; the terminals are configured as follows: Stores a summary of the features of one or more samples and their corresponding label information; The sampling information of the target to be identified is obtained, features are extracted based on the sampling information of the target to be identified, and the extracted features are processed by a summarization algorithm to obtain a feature summary of the sampling information. The sampled information feature summary is matched with the pre-stored sample information feature summary to obtain the label information of the target to be identified; The label information of the target to be identified is the label information corresponding to the feature summary of the successfully matched sample information.

[0052] Scheme 48: According to the target recognition system based on digest algorithm feature encryption as described in Scheme 47, multiple terminals are connected through a communication link; For any of the terminals, if the feature summary of the sampled information corresponding to the sampled information of the target to be identified fails to match its locally pre-stored feature summary of sampled information, it will sequentially match the feature summaries of sampled information pre-stored in other terminals according to a preset priority rule until a match is successful, thereby obtaining the tag information of the target to be identified.

[0053] Solution 49: According to the target recognition system based on feature encryption using a digest algorithm described in Solution 48, the method for "matching with feature digests of sample information pre-stored in other terminals" is as follows: The sampled information feature summaries are sent to other terminals sequentially according to a preset priority rule for matching; or The system will sequentially retrieve and match the feature summaries of pre-stored sample information from other terminals according to preset priority rules.

[0054] Option 50: The target recognition system based on digest algorithm feature encryption as described in any one of Options 47-49, wherein the terminal is configured as follows: The sampling information of the sample is obtained, features are extracted, and the extracted features are processed by a summarization algorithm to obtain a feature summary of the sample information. The obtained feature summary of the sample information and its label information are stored locally.

[0055] Option 51: The target recognition system based on digest algorithm feature encryption as described in Option 50 further includes a sample acquisition device. The sample collection device is configured to acquire sample collection information and send it to the corresponding terminal.

[0056] Option 52, the target recognition system based on digest algorithm feature encryption according to any one of Options 47-49, further includes a sample acquisition device: The sample collection device is configured as follows: The sampling information of the sample is obtained, features are extracted, and the extracted features are processed by a summarization algorithm to obtain a sample information feature summary. The obtained sample information feature summary and its tag information are sent to the corresponding terminal.

[0057] Scheme 53: According to any one of Schemes 50-52, the target recognition system based on feature encryption using a digest algorithm deletes the sampling information of the sample after extracting the feature digest of the sample information.

[0058] Option 54: A device comprising: At least one processor; and A memory communicatively connected to at least one of the processors; wherein, The memory stores instructions executable by the processor. These instructions are used to implement the target recognition method based on digest algorithm feature encryption as described in any one of Schemes 1-9, or the target matching method based on digest algorithm feature encryption as described in any one of Schemes 10-19, or the face recognition method based on digest algorithm feature encryption as described in Scheme 20, or the voice recognition method based on digest algorithm feature encryption as described in Scheme 21, or the pupil recognition method based on digest algorithm feature encryption as described in Scheme 22, or the fingerprint recognition method based on digest algorithm feature encryption as described in Scheme 23, or the vehicle recognition method based on digest algorithm feature encryption as described in Scheme 24, or the plant recognition method based on digest algorithm feature encryption as described in Scheme 25, or the semantic recognition method based on digest algorithm feature encryption as described in Scheme 26.

[0059] Option 55: A computer-readable storage medium storing computer instructions, the computer instructions being executed by the computer to implement the target recognition method based on digest algorithm feature encryption as described in any one of Options 1-9, or the target matching method based on digest algorithm feature encryption as described in any one of Options 10-19, or the face recognition method based on digest algorithm feature encryption as described in Option 20, or the voice recognition method based on digest algorithm feature encryption as described in Option 21, or the pupil vein recognition method based on digest algorithm feature encryption as described in Option 22, or the fingerprint recognition method based on digest algorithm feature encryption as described in Option 23, or the vehicle recognition method based on digest algorithm feature encryption as described in Option 24, or the plant recognition method based on digest algorithm feature encryption as described in Option 25, or the semantic recognition method based on digest algorithm feature encryption as described in Option 26.

[0060] The beneficial effects of this invention are: (1) This invention processes feature data through a digest algorithm to obtain digest data, and performs target identification based on the digest data matching method. It does not store the original target feature data. Based on the irreversibility of the digest data, even if the data is leaked, third parties cannot read the target feature information represented by the digest data, thereby ensuring the security of user data.

[0061] (2) The present invention uses the combination of summary data and feature type data as the matching object, which improves the accuracy and speed of matching.

[0062] (3) Based on the network architecture of server and terminal, the present invention can send and store the user information corresponding to the terminal according to the settings, which improves the identification efficiency of the terminal and saves the storage space of the terminal. Attached Figure Description

[0063] Other features, objects, and advantages of this application will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings: Figure 1 This is a schematic flowchart of the target recognition method based on feature encryption using a digest algorithm according to Embodiment 1 of the present invention; Figure 2 This is an example diagram of the standardized adjustment based on feature information in Embodiment 3 of the present invention; Figure 3 This is a schematic diagram of the target recognition system framework based on feature encryption using a digest algorithm according to Embodiment 10 of the present invention; Figure 4 This is a schematic diagram of the target recognition system framework based on digest algorithm feature encryption according to Embodiment 11 of the present invention; Figure 5 This is a schematic diagram of the target recognition system framework based on feature encryption using a digest algorithm according to Embodiment 19 of the present invention; Figure 6 This is a schematic diagram of the structure of a computer system used to implement the methods, systems, and devices of this application. Detailed Implementation

[0064] The present application will now be described in further detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the invention. Furthermore, it should be noted that, for ease of description, only the parts relevant to the invention are shown in the accompanying drawings.

[0065] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0066] Example 1: An embodiment of the present invention provides a target identification method based on feature encryption using a digest algorithm, such as... Figure 1 As shown, it includes the following steps: Obtain sampling information of the target to be identified; Feature extraction is performed based on the sampling information of the target to be identified, and the extracted features are processed by a summarization algorithm to obtain a feature summary of the sampling information; The sampled information feature summary is matched with the pre-stored sample information feature summary to obtain the label information of the target to be identified; the label information of the target to be identified is the label information corresponding to the successfully matched sample information feature summary.

[0067] Each feature summary in the sampling information feature summary includes a feature identifier and summary data; the feature identifier is a category identifier, or an ordered arrangement of multiple category identifiers; the summary data is obtained by processing the feature data corresponding to a category through a summary algorithm, or by processing the feature data corresponding to multiple categories in an ordered arrangement through a summary algorithm.

[0068] Furthermore, after obtaining the tag information of the target to be identified, the method also includes: deleting the sampling information of the target to be identified.

[0069] In this embodiment, the target to be identified can be any object whose features can be extracted and identified based on the extracted features, such as a face, voice, pupil pattern, fingerprint, vehicle, plant, or information. The target categories listed here are merely for intuitive illustration of the target categories that the target to be identified can include, and should not be construed as limiting the technical solution of this invention.

[0070] The sampling information is determined based on the category of the target to be identified. For example, if the target to be identified is a face, the sampling information is a face image; if the target to be identified is sound, the sampling information is a sound signal; if the target to be identified is pupil veins, the sampling information is a pupil vein image; if the sampling information is a vehicle, the sampling information can be a vehicle image, or the detected external structure measurement information, or the acquired three-dimensional information; if the target to be identified is information, it can be text information, or speech information from which semantic features can be extracted, etc.

[0071] This invention replaces the intuitive feature data (such as face images, sound clips, etc.) in the prior art with feature summary data. It utilizes the irreversible nature of the summarization algorithm, so it is impossible to reverse the original feature data based on the calculated summary. This can protect data and privacy. Even if the feature summary data is obtained, it is impossible to reverse the feature information of the sample.

[0072] The feature digest of the sampled information and the feature digest of the pre-stored sampled information are obtained using the same digest algorithm, such as any one of SHA-1, SHA256 or MD5, or other digest algorithms, as long as the same plaintext data is input and processed by the same message digest algorithm to obtain the same ciphertext, and the generated ciphertext is unique and irreversible.

[0073] Feature extraction is performed on the sampled information of the target to be identified. Based on the category of the target, existing feature extraction methods can be used, which are well described in existing technologies and will not be elaborated here. Of course, user-defined feature extraction methods can also be used.

[0074] In this embodiment, the objects being matched are the sampling information feature summary and the sample information feature summary.

[0075] The method for obtaining the sample information feature summary is as follows: the sample feature information is processed by a summarization algorithm to obtain the sample information feature summary; the sample feature information is obtained by feature extraction based on the sample sampling information, or by input through a human-computer interaction port, or by retrieval from the sample feature information storage unit.

[0076] The feature summary of the sampled information is obtained through a summarization algorithm based on the extracted features.

[0077] The same feature extraction method can be used to extract sample information features and sampling information features. For some industrial applications, a certain type of target has the same features, such as vehicles, parts, and other industrial products. In this case, feature parameters can be measured manually or automatically, entered through a human-computer interaction port, and then a sample information feature summary can be generated. Alternatively, the sample information feature summary can be generated by retrieving it from a sample information feature storage unit that stores sample feature information. In this embodiment, to avoid the leakage of readable sample feature data, the sample information feature summary can be generated on a physically isolated processing device. The generated sample information feature summary data is then copied out and placed into the system configured with the target recognition method based on the digest algorithm feature encryption of this invention for target recognition matching.

[0078] In this embodiment, the tag information corresponding to the sample information feature summary can be personnel information, personnel classification, vehicle model, plant variety, semantic context association information, etc., depending on the sample category. Of course, it can also be the tag information set for it.

[0079] Generally, the feature category corresponding to the sample information feature summary is greater than or equal to the feature category corresponding to the sample information feature summary, and the feature category corresponding to the sample information feature summary is included in the feature category corresponding to the sample information feature summary. For example, in the field of face recognition, the category corresponding to the sample information feature summary can be N types of features obtained based on facial feature key points, and the category corresponding to the sample information feature summary can be all or part of these N types of features.

[0080] In this embodiment, the extracted features are processed by a summarization algorithm to obtain a feature summary of the sampled information, which can be method one, method two, or method three.

[0081] Method 1: For each feature extracted based on the sampling information, a corresponding feature summary is obtained by processing it with a summarization algorithm to construct a feature summary of the sampling information; Method 2: Combine the features extracted from the sampled information according to the feature combinations corresponding to one or more feature summaries in the sampled information feature summary, and then process the combination through a summary algorithm to obtain the corresponding feature summary, thus constructing the sampled information feature summary; Method 3: Multiple features extracted from the sampling information are combined according to a pre-defined or randomly generated feature combination scheme. After combination, a summarization algorithm is used to obtain the corresponding feature summary, thus constructing the sampling information feature summary. When constructing the sampling information feature summary using this method, the pre-stored sample information feature summaries are combined according to the combination scheme in this method, and then the two are matched.

[0082] To improve matching accuracy and efficiency, each feature summary in the sampling information feature summary includes a feature identifier and summary data. Similarly, each feature summary in the sample information feature summary also includes a feature identifier and summary data.

[0083] The feature identifier is a single category identifier, or an ordered arrangement of multiple category identifiers; the summary data is obtained by processing the feature data corresponding to one category using a summarizing algorithm, or by processing the feature data corresponding to multiple categories in an ordered arrangement using a summarizing algorithm. For greater clarity, this scheme will be illustrated in detail with the following face recognition implementation examples.

[0084] When the features extracted from the sampling information contain features of measurement data type, the hierarchical information corresponding to the features of the measurement data is obtained according to the preset hierarchical principle, and the feature summary is obtained by the summarization algorithm based on the hierarchical information.

[0085] Since the feature data needs to be processed by a summarization algorithm, the measurement accuracy of the data is not necessarily the more precise the better. Instead, the accuracy level is selected based on the current feature extraction technology, such as micrometers or millimeters, in order to eliminate measurement errors and ensure that the corresponding feature data extracted from multiple data collections of the same target are the same.

[0086] Alternatively, a hierarchical approach can be used. Level units are set, and the corresponding level is obtained by dividing the actual measured value by the corresponding level unit and rounding down. This method can be used for features such as length and area. For example, with length units, if each 0.1 mm represents a level, then the length feature 31.15 mm divided by 0.1 equals 311.5, and rounding down gives the corresponding length level as 311. Of course, other level units can also be used, such as 0.2 mm, 0.5 mm, 2 mm, etc.

[0087] The method for matching the sampled information feature summary with the pre-stored sample information feature summary is as follows: The sampling information feature summary is matched individually with one or more sample information feature summaries to obtain matching degree information; If there is only one sample information feature summary, then if the matching degree information is greater than the preset first matching degree threshold, it is determined that the match is successful; if there are multiple sample information feature summaries, then the largest sample information feature summary with matching degree information greater than the preset second matching degree threshold is selected as the successfully matched sample information feature summary.

[0088] Matching information is ; Where A represents the matching degree information, a represents the number of successfully matched feature summaries in the sampled information feature summary, and b represents the number of feature summaries contained in the sampled information feature summary.

[0089] The sampling information feature summary can also be matched with the pre-stored sample information feature summary using the following method: Based on feature category encoding, a set number of permutations and combinations of feature categories are randomly selected as category combination items; Based on the obtained category combination items, the feature information summary data corresponding to each category ranking is obtained by sorting the feature summary data of the sampling information and the sample information respectively, and then processing it based on the summary algorithm to obtain the category ranking feature summary data corresponding to each category group. The category-ranked feature summary data corresponding to the feature summary of the sampled information is compared with the category-ranked feature summary data corresponding to the feature summary of each sample information to obtain matching degree information; Select the sample with the highest matching degree in the feature summary that is greater than the set matching degree threshold as the data that is successfully matched.

[0090] In this method, the feature categories included in each category ranking item are determined by a random function, and of course, a fixed number of ranking items can also be determined.

[0091] Example 2 This embodiment of a target matching method based on digest algorithm feature encryption includes the following steps: Feature extraction is performed on the sampled information of the target to be identified, and the extracted features are processed by a summarization algorithm to obtain a feature summary of the sampled information; The sampled information feature summary is matched with the pre-stored sample information feature summary to obtain the matching result.

[0092] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working process and related explanations of the target matching method based on the feature encryption of the digest algorithm described above can be referred to the corresponding process in the aforementioned Embodiment 1, and will not be repeated here.

[0093] Example 3 This embodiment provides a face recognition method based on digest algorithm feature encryption, the method comprising: Obtain the facial image of the person to be identified; Feature extraction is performed on the facial images of the person to be identified, and the extracted features are processed by a summarization algorithm to obtain a feature summary of the sampled information; The sampled information feature summary is matched with the pre-stored sample information feature summary to obtain the tag information of the person to be identified; the tag information of the person to be identified is the person information corresponding to the successfully matched sample information feature summary.

[0094] The facial features used in this embodiment can be: distance information obtained based on the extracted facial feature points according to pre-defined feature point pairs, or ratio data based on the distance of pre-defined feature point pairs, or other quantifiable descriptive feature information. For example, the size and position of facial features such as the iris, nose, and corners of the mouth, the distance between facial features, and the distance between the two eyes.

[0095] When using this method, since the acquired images cannot be identical to the original sample images, there may be inconsistencies such as skewness and size. Therefore, directly comparing the above data is completely impractical. Thus, it is necessary to retain the standardization adjustment basis feature information while acquiring the facial feature summary data, so as to standardize and adjust the acquired facial images based on this information, thereby obtaining the same facial feature information for the same user. For example... Figure 2 As shown, among the 66 facial feature points identified, the standardization adjustment based on feature information may include the central feature point (which may be the tip of the nose 34), the outer contour features (which may be the left corner of the left eye 46, the right corner of the right eye 37, the left corner of the mouth 55, and the right corner of the mouth 49), the distance from the central feature point to each outer contour feature point (solid line in the figure), and the distance between two adjacent outer contour feature points (dashed line in the figure).

[0096] The facial image of the person to be identified can be acquired through an image acquisition device. If there are multiple faces in the recognition area, the face to be identified can be selected manually or automatically. After the face to be identified is locked, the image of the face to be identified can be preliminarily processed, such as adjusting the color, angle, and brightness.

[0097] The feature extraction process can utilize existing techniques and will not be elaborated upon here. The feature data may include the size and position of facial features such as the iris, nose, and corners of the mouth, as well as the distances between facial features and between the eyes. It is understood that standardization is required when extracting facial features. Since there are no prior reference samples, image standardization can be performed based on the aforementioned preservation and standardization adjustments according to the feature information. Alternatively, the smallest bounding rectangle or circle of the identified feature points can be obtained and then enlarged or reduced to a uniform size to obtain standardized feature data. Alternatively, instead of directly using parameters, the ratio between parameters can be used as the basis for generating summary data. Of course, existing feature extraction methods can also be used to extract feature data.

[0098] The facial features extracted in this embodiment are consistent with the facial features contained in the sample data of the facial database.

[0099] The extracted features can include the distance between two set feature points, or the perimeter or area of ​​a facial organ obtained from the feature points, or other feature data.

[0100] In this embodiment, the method for obtaining summary data through the summarization algorithm is as follows: the facial feature data obtained based on the set facial features is used to obtain corresponding hierarchical information according to the preset hierarchical principle; the obtained hierarchical information is processed by the summarization algorithm to obtain feature summary data; the feature category code of the corresponding facial feature data is obtained according to the preset facial feature category coding table; the feature category code and the feature summary data are concatenated to obtain the summary data of the corresponding facial feature data.

[0101] This method is applied to the acquisition of feature summaries of sampling information and the acquisition of feature summaries of sample information.

[0102] Since the feature data needs to be processed by a summarization algorithm, the measurement accuracy of the data is not necessarily the more precise the better. Instead, the accuracy level is selected based on the current image recognition technology, such as micrometers or millimeters, in order to eliminate measurement errors and ensure that the corresponding feature data extracted from multiple facial images of the same person are the same.

[0103] A tiered approach can also be used, setting level units. The actual measured value is divided by the corresponding level unit and rounded down to obtain the corresponding level. This method can be used for features such as length and area. For example, with length units, each 0.1 mm represents a level. Therefore, the distance between the inner corners of the eyes (23.15 mm) divided by 0.1 equals 231.5, and rounded down, the corresponding length level is 231. Other level units can also be used, such as 0.2 mm, 0.5 mm, and 2 mm.

[0104] Feature data can also be a combination of two or more facial features, such as iris contour + interpupillary distance, iris contour + interpupillary distance + distance from the center point of each pupil to the corresponding corner of the mouth, etc.

[0105] In this embodiment, matching the feature summary of the sampling information with the feature summary of the sample information can be achieved using one of the following two methods: Method 1: Based on feature category encoding, the feature summary of the sampled information is compared with the feature summary of each sample information separately to obtain the matching degree information; the sample information feature summary with the largest matching degree that is greater than the set matching degree threshold is selected as the data that is successfully matched.

[0106] Method 2: Based on feature category encoding, a set number of permutations and combinations of feature categories are randomly selected as category combination items; Based on the obtained category combination items, the feature summary data corresponding to each category combination is obtained from the feature summary of the sampling information and the feature summary of the sample information, and then processed by the summary algorithm to obtain the category permutation combination feature summary data corresponding to each category group. The category permutation and combination feature summary data corresponding to the feature summary of the sampled information is compared with the category permutation and combination feature summary data corresponding to the feature summary of the sampled information in the pre-stored sample information to obtain matching degree information; Select the sample with the highest matching degree in the feature summary that is greater than the set matching degree threshold as the data that is successfully matched.

[0107] In Method 2 above, the number of feature categories contained in each category combination is determined by a random function. Of course, a fixed number of combination items can also be determined.

[0108] In Method 2 above, the representation of category combination items and their corresponding category combination feature summary data can be as follows: For example: A1B1C9|ECE7DC2FE50D337B.

[0109] The character before “|” represents the sequence of codes corresponding to the facial feature arrangement sequence, with two digits representing one feature. In this embodiment, A1 represents the outline length of the left pupil, B1 represents the outline length of the right pupil, and C9 represents the interpupillary distance. The character after “|” represents the encoding of the data corresponding to features A1, B1, and C9 obtained by the digest algorithm.

[0110] Similarly, the feature summaries of sample information pre-stored in the face database can also be composed of a sequence of codes corresponding to a sequence of face features, and the encoding of sample feature data based on a summarization algorithm. This allows for comparison of summaries of the same feature arrangement with sample summary data, reducing the amount of data to be compared and improving recognition efficiency. Alternatively, it can be a combination of a single feature code and its summary data. When combined feature comparison is needed, the data of multiple corresponding single features (feature code + its summary data) are arranged and summary data is generated again. The resulting feature arrangement summary data can then be compared with the feature arrangement summary data of the acquired image. The latter method can significantly reduce the amount of data stored in the face database and increase the flexibility of feature arrangement combinations during feature comparison.

[0111] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working process and related explanations of the face recognition method based on digest algorithm feature encryption described above can be found in the corresponding process in the foregoing embodiments, and will not be repeated here.

[0112] Example 4 This embodiment provides a sound recognition method based on digest algorithm feature encryption, the method comprising: Obtain sampling information of the sound to be identified; Feature extraction is performed on the sampled sound information to be recognized, and the extracted features are processed by a summarization algorithm to obtain a feature summary of the sampled information; The sampled information feature summary is matched with the pre-stored sample information feature summary to obtain the tag information of the sound to be identified; the tag information of the sound to be identified is the tag information corresponding to the successfully matched sample information feature summary.

[0113] Depending on the subject, the sound information to be identified can be human speech, music, animal vocalizations, or sounds from machinery or equipment.

[0114] Example 5 A pupil print recognition method based on feature encryption using a digest algorithm, the method comprising: Obtain the iris print information of the person to be identified; Feature extraction is performed on the pupil print information of the person to be identified, and the extracted features are processed by a summarization algorithm to obtain a feature summary of the sampled information; The sampled information feature summary is matched with the pre-stored sample information feature summary to obtain the tag information of the person to be identified; the tag information of the person to be identified is the person information corresponding to the successfully matched sample information feature summary.

[0115] The pupil pattern information in this embodiment includes image information of the pupil texture, which can be obtained through a pupil and iris recognition device.

[0116] Example 6 A fingerprint recognition method based on digest algorithm feature encryption, the method comprising: Obtain the fingerprint information of the person to be identified; Feature extraction is performed on the fingerprint information of the person to be identified, and the extracted features are processed by a summarization algorithm to obtain a feature summary of the sampled information; The sampled information feature summary is matched with the pre-stored sample information feature summary to obtain the tag information of the person to be identified; the tag information of the person to be identified is the person information corresponding to the successfully matched sample information feature summary.

[0117] Example 7 A vehicle identification method based on feature encryption using a digest algorithm, the method comprising: Obtain sampling information of the vehicle to be identified; Feature extraction is performed on the sampled information of the vehicle to be identified, and the extracted features are processed by a summarization algorithm to obtain a feature summary of the sampled information; The sampled information feature summary is matched with the pre-stored sample information feature summary to obtain the tag information of the vehicle to be identified; the tag information of the vehicle to be identified is the tag information corresponding to the successfully matched sample information feature summary.

[0118] Example 8 A plant identification method based on feature encryption using a digest algorithm, the method comprising: Obtain sampling information of the plant to be identified; Feature extraction is performed on the sampling information of the plants to be identified, and the extracted features are processed by a summarization algorithm to obtain a feature summary of the sampling information; The sampled information feature summary is matched with the pre-stored sample information feature summary to obtain the label information of the plant to be identified; the label information of the plant to be identified is the label information corresponding to the successfully matched sample information feature summary.

[0119] Example 9 A semantic recognition method based on feature encryption using a digest algorithm, the method comprising: Obtain the semantic features of the information to be identified; The semantic features of the information to be identified are processed by a summarization algorithm to obtain a feature summary of the sampled information. The feature summary of the sampled information is matched with the pre-stored feature summary of the sample information to obtain the label information of the information to be identified; the label information of the information to be identified is the label information corresponding to the successfully matched feature summary of the sample information.

[0120] In this embodiment, the tag information can be question-and-answer related statement information, information classification, etc.

[0121] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes and related descriptions of Embodiments 4 to 9 described above can be referred to the corresponding processes in the foregoing embodiments, and will not be repeated here.

[0122] Example 10 A target recognition system based on feature encryption using a digest algorithm, such as Figure 3 As shown, it includes a first unit, a second unit, a third unit, and may also include a fourth unit; The first unit is configured to acquire sampling information of the target to be identified; The second unit is configured to extract features based on the sampling information of the target to be identified, and process the extracted features through a summarization algorithm to obtain a feature summary of the sampling information; The third unit is configured to match the feature summary of the sampled information with the pre-stored feature summary of the sampled information to obtain the label information of the target to be identified; the label information of the target to be identified is the label information corresponding to the successfully matched feature summary of the sampled information.

[0123] The fourth unit is configured to store a sample information feature summary of one or more samples.

[0124] The fourth unit is located on the terminal device and / or a remote server. When the fourth unit is located on a remote server, because the hash algorithm produces a smaller amount of data (e.g., SHA1 results in 160 bits, typically represented by a 40-bit hexadecimal string), the terminal can upload only the hash data to the server, reducing the amount of data uploaded, shortening the comparison time, reducing latency, and improving transmission accuracy. Furthermore, compared to traditional target recognition methods, changing even a single feature in the hash algorithm results in entirely different hash data, thus leading to more accurate recognition results.

[0125] Example 11 A target recognition system based on feature encryption using a digest algorithm, such as Figure 4 As shown, it includes a server and one or more terminals; the terminals and the server are communicatively connected. The server is configured to extract features based on the sampling information of the target to be identified, and process the extracted features using a summarization algorithm to obtain a feature summary of the sampling information; the feature summary of the sampling information is matched with a pre-stored feature summary of the sample information to obtain the tag information of the target to be identified, and then sent to the terminal. The terminal is configured to acquire sampling information of the target to be identified and upload it to the server; and acquire the tag information of the target to be identified obtained by the server. The label information of the target to be identified is the label information corresponding to the feature summary of the successfully matched sample information.

[0126] In this embodiment, the terminal is further configured to: acquire the sampling information of the sample, extract features, process the extracted features through a summarization algorithm to obtain a sample information feature summary, and store the obtained sample information feature summary and its tag information locally and / or upload them to the server.

[0127] Alternatively, the terminal may be configured to acquire sample information and upload it to the server; the server may be configured to extract features from the sample information, process the extracted features using a summarization algorithm to obtain a sample information feature summary, and store the obtained sample information feature summary and its tag information locally and / or send it to the terminal.

[0128] Meanwhile, after extracting the feature summary of the sample information, the sampling information of the sample is deleted, which further improves data security.

[0129] Example 12 A server is communicatively connected to one or more terminals. The server is configured to extract features based on sampling information of a target to be identified, and process the extracted features using a summarization algorithm to obtain a feature summary of the sampling information; match the feature summary of the sampling information with a pre-stored feature summary of sample information to obtain tag information of the target to be identified, and send it to the terminal; the tag information of the target to be identified is the tag information corresponding to the successfully matched feature summary of sample information.

[0130] Example 13 A target recognition system based on feature encryption using a digest algorithm includes a server and one or more terminals; the terminals and the server are communicatively connected. The server is configured to acquire a feature summary of the sampled information of the target to be identified, match the feature summary of the sampled information with a pre-stored feature summary of the sampled information to obtain the tag information of the target to be identified, and send it to the terminal. The terminal is configured to acquire sampling information of the target to be identified, extract features based on the sampling information of the target to be identified, process the extracted features through a summarization algorithm to obtain a feature summary of the sampling information, and upload it to the server; and acquire the tag information of the target to be identified obtained by the server. The label information of the target to be identified is the label information corresponding to the feature summary of the successfully matched sample information.

[0131] In this embodiment, the terminal is further configured to: acquire the sampling information of the sample, extract features, process the extracted features through a summarization algorithm to obtain a sample information feature summary, and store the obtained sample information feature summary and its tag information locally and / or upload them to the server.

[0132] Alternatively, the terminal may be configured to acquire sample information and upload it to the server; the server may be configured to extract features from the sample information, process the extracted features using a summarization algorithm to obtain a sample information feature summary, and store the obtained sample information feature summary and its tag information locally and / or send it to the terminal.

[0133] Meanwhile, after extracting the feature summary of the sample information, the sampling information of the sample is deleted, which further improves data security.

[0134] Example 14 A server is communicatively connected to one or more terminals. The server is configured to acquire a feature summary of sampled information of a target to be identified, match the feature summary of sampled information with a pre-stored feature summary of sampled information to obtain tag information of the target to be identified, and send it to the terminal. The tag information of the target to be identified is the tag information corresponding to the successfully matched feature summary of sampled information.

[0135] Example 15 A terminal, communicatively connected to a server, is configured to acquire sampling information of a target to be identified, extract features based on the sampling information of the target to be identified, process the extracted features using a summarization algorithm to obtain a feature summary of the sampling information, and upload it to the server; acquire tag information of the target to be identified obtained by the server; the tag information of the target to be identified is the tag information corresponding to the feature summary of the successfully matched sample information.

[0136] Example 16 A target recognition system based on feature encryption using a digest algorithm includes a server and one or more terminals; the terminals and the server are communicatively connected. The server is configured to store feature summaries of sample information for multiple samples; The terminal is configured as follows: Obtain a feature summary of sample information for one or more samples from the server; The sampling information of the target to be identified is obtained, features are extracted based on the sampling information of the target to be identified, and the extracted features are processed by a summarization algorithm to obtain a feature summary of the sampling information. The sampled information feature summary is matched with the pre-stored sample information feature summary to obtain the label information of the target to be identified; The label information of the target to be identified is the label information corresponding to the feature summary of the successfully matched sample information.

[0137] In this embodiment, the terminal is further configured to: acquire the sampling information of the sample, extract features, process the extracted features through a summarization algorithm to obtain a sample information feature summary, and store the obtained sample information feature summary and its tag information locally and / or upload them to the server.

[0138] Alternatively, the terminal may be configured to acquire sample information and upload it to the server; the server may be configured to extract features from the sample information, process the extracted features using a summarization algorithm to obtain a sample information feature summary, and store the obtained sample information feature summary and its tag information locally and / or send it to the terminal.

[0139] Meanwhile, after extracting the feature summary of the sample information, the sampling information of the sample is deleted, which further improves data security.

[0140] Example 17 A server, communicatively connected to one or more terminals, the server being configured to store sample information feature summaries of multiple samples.

[0141] Example 18 A terminal, communicatively connected to a server, is configured to: Obtain a feature summary of sample information for one or more samples from the server; The sampling information of the target to be identified is obtained, features are extracted based on the sampling information of the target to be identified, and the extracted features are processed by a summarization algorithm to obtain a feature summary of the sampling information. The sampled information feature summary is matched with the pre-stored sample information feature summary to obtain the label information of the target to be identified; The label information of the target to be identified is the label information corresponding to the feature summary of the successfully matched sample information.

[0142] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes and related descriptions of Embodiments 10 to 18 described above can be referred to the corresponding processes in the foregoing embodiments, and will not be repeated here.

[0143] It should be noted that the above embodiments are merely illustrative examples of the division of functional modules. In practical applications, the functions described above can be assigned to different functional modules as needed, that is, the modules or steps in the embodiments of the present invention can be further decomposed or combined. For example, the modules in the above embodiments can be merged into one module, or further divided into multiple sub-modules to complete all or part of the functions described above. The names of the modules and steps involved in the embodiments of the present invention are merely for distinguishing the various modules or steps and are not considered as an improper limitation of the present invention.

[0144] Example 19 A target recognition system based on feature encryption using a digest algorithm, such as Figure 5 As shown, it includes one or more terminals; the terminals are configured as follows: Stores a summary of the features of one or more samples and their corresponding label information; The sampling information of the target to be identified is obtained, features are extracted based on the sampling information of the target to be identified, and the extracted features are processed by a summarization algorithm to obtain a feature summary of the sampling information. The sampled information feature summary is matched with the pre-stored sample information feature summary to obtain the label information of the target to be identified; The label information of the target to be identified is the label information corresponding to the feature summary of the successfully matched sample information.

[0145] Multiple terminals are connected via a communication link. For any terminal, if the feature summary of the sample information corresponding to the sample information of the target to be identified fails to match its locally pre-stored feature summary of sample information, it will sequentially match the feature summary of sample information pre-stored in other terminals according to a preset priority rule until a match is successful, thereby obtaining the tag information of the target to be identified.

[0146] "Matching with pre-stored sample information feature summaries in other terminals" is achieved by: sending the sample information feature summaries to other terminals sequentially according to a preset priority rule for matching; or obtaining pre-stored sample information feature summaries from other terminals sequentially according to a preset priority rule for matching.

[0147] In this embodiment, the terminal is configured to: acquire the sampling information of the sample, extract features, process the extracted features through a summarization algorithm to obtain a sample information feature summary, and store the obtained sample information feature summary and its tag information locally.

[0148] This embodiment also includes a sample collection device: The sample collection device is configured to acquire sample collection information and send it to the corresponding terminal. The sample acquisition device can also be configured to: acquire sample sampling information, extract features, process the extracted features using a summarization algorithm to obtain a sample information feature summary, and send the obtained sample information feature summary and its tag information to the corresponding terminal.

[0149] After extracting the feature summary of the sample information, the sampling information of the sample is deleted from the sample acquisition device and / or terminal.

[0150] The difference between this embodiment and the aforementioned system embodiments is that the server is eliminated, and a decentralized identification network is implemented by connecting multiple terminals, which greatly reduces the cost of the system architecture.

[0151] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working process and related descriptions of Embodiment Nineteen described above can be referred to the corresponding processes in the foregoing embodiments, and will not be repeated here.

[0152] Example 20 A device according to a third embodiment of the present invention includes: At least one processor; and A memory communicatively connected to at least one of the processors; wherein, The memory stores instructions that can be executed by the processor. These instructions are used to implement the target recognition method based on digest algorithm feature encryption, or the target matching method based on digest algorithm feature encryption, or the face recognition method based on digest algorithm feature encryption, or the voice recognition method based on digest algorithm feature encryption, or the pupil vein recognition method based on digest algorithm feature encryption, or the fingerprint recognition method based on digest algorithm feature encryption, or the vehicle recognition method based on digest algorithm feature encryption, or the plant recognition method based on digest algorithm feature encryption, or the semantic recognition method based on digest algorithm feature encryption.

[0153] A computer-readable storage medium according to a fourth embodiment of the present invention stores computer instructions, which are executed by the computer to implement the above-described target recognition method based on digest algorithm feature encryption, or target matching method based on digest algorithm feature encryption, or face recognition method based on digest algorithm feature encryption, or voice recognition method based on digest algorithm feature encryption, or pupil recognition method based on digest algorithm feature encryption, or fingerprint recognition method based on digest algorithm feature encryption, or vehicle recognition method based on digest algorithm feature encryption, or plant recognition method based on digest algorithm feature encryption, or semantic recognition method based on digest algorithm feature encryption.

[0154] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working process and related descriptions of the above-described device and computer-readable storage medium can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0155] The following is for reference. Figure 6 It shows a schematic diagram of the structure of a computer system for implementing the methods, systems, and devices of this application. Figure 6 The server shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of this application.

[0156] like Figure 6 As shown, the computer system includes a Central Processing Unit (CPU) 601, which can perform various appropriate actions and processes based on programs stored in Read Only Memory (ROM) 602 or programs loaded from storage section 608 into Random Access Memory (RAM) 603. RAM 603 also stores various programs and data required for system operation. The CPU 601, ROM 602, and RAM 603 are interconnected via bus 604. An Input / Output (I / O) interface 605 is also connected to bus 604.

[0157] The following components are connected to I / O interface 605: an input section 606 including a keyboard, mouse, etc.; an output section 607 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and speakers, etc.; a storage section 608 including a hard disk, etc.; and a communication section 609 including a network interface card such as a LAN (Local Area Network) card, modem, etc. The communication section 609 performs communication processing via a network such as the Internet. A drive 610 is also connected to I / O interface 605 as needed. A removable medium 611, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on drive 610 as needed so that computer programs read from it can be installed into storage section 608 as needed.

[0158] Specifically, according to embodiments of this disclosure, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this disclosure include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication section 609, and / or installed from removable medium 611. When the computer program is executed by central processing unit (CPU) 601, it performs the functions defined in the methods of this application. It should be noted that the computer-readable medium described above in this application can be a computer-readable signal medium or a computer-readable storage medium, or any combination of the two. A computer-readable storage medium can be, for example—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this application, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in connection with an instruction execution system, apparatus, or device. In this application, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can also be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on a computer-readable medium can be transmitted using any suitable medium, including but not limited to: radio, optical fiber, RF, etc., or any suitable combination thereof.

[0159] Computer program code for performing the operations of this application can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0160] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0161] The terms “first”, “second”, etc., are used to distinguish similar objects, not to describe or indicate a specific order or sequence.

[0162] The term "comprising" or any other similar term is intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus / device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent in such process, method, article, or apparatus / device.

[0163] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of the present invention.

Claims

1. A target recognition method based on feature encryption using a digest algorithm, characterized in that, Includes the following steps: Obtain sampling information of the target to be identified; Feature extraction is performed based on the sampling information of the target to be identified, and the extracted features are processed by a summarization algorithm to obtain a feature summary of the sampling information; The sampled information feature summary is matched with the pre-stored sample information feature summary to obtain the label information of the target to be identified; the label information of the target to be identified is the label information corresponding to the successfully matched sample information feature summary. Each feature summary in the sampling information feature summary includes a feature identifier and summary data; The feature identifier is a category identifier, or an ordered arrangement of multiple category identifiers; the summary data is obtained by processing the feature data corresponding to one category through a summary algorithm, or by processing the feature data corresponding to multiple categories in an ordered arrangement through a summary algorithm.

2. The target identification method based on digest algorithm feature encryption according to claim 1, characterized in that, "The extracted features are processed using a summarization algorithm to obtain a feature summary of the sampled information." The method is as follows: Based on the feature combination corresponding to one or more feature summaries in the sample information feature summary, the features extracted from the sample information are combined, and after combination, the corresponding feature summary is obtained by the summarization algorithm to construct the sample information feature summary.

3. The target identification method based on digest algorithm feature encryption according to claim 1, characterized in that, "The extracted features are processed using a summarization algorithm to obtain a feature summary of the sampled information." The method is as follows: For multiple features extracted based on sampling information, the features are combined according to a preset or randomly generated feature combination scheme, and the corresponding feature summary is obtained by a summarization algorithm after combination, thus constructing a feature summary of sampling information. When constructing the sample information feature summary, the pre-stored sample information feature summaries are combined according to the combination scheme and then matched.

4. The target identification method based on digest algorithm feature encryption according to claim 1, characterized in that, The method for matching the feature summary of the sampled information with the pre-stored feature summary of the sampled information is as follows: The sampling information feature summary is matched individually with one or more sample information feature summaries to obtain matching degree information; If there is only one sample information feature summary, then if the matching degree information is greater than the preset first matching degree threshold, it is determined that the match is successful; if there are multiple sample information feature summaries, then the largest sample information feature summary with matching degree information greater than the preset second matching degree threshold is selected as the successfully matched sample information feature summary.

5. The target identification method based on digest algorithm feature encryption according to claim 4, characterized in that, The matching degree information is as follows: A=a / b Where A represents the matching degree information, a represents the number of successfully matched feature summaries in the sampled information feature summary, and b represents the number of feature summaries contained in the sampled information feature summary.

6. The target identification method based on digest algorithm feature encryption according to claim 1, characterized in that, The method for obtaining the feature summary of the sample information is as follows: The feature information of the sample is processed by a summarization algorithm to obtain a feature summary of the sample information; The feature information of the sample is obtained by feature extraction based on the sample sampling information, or by input through the human-computer interaction port, or by retrieval from the sample feature information storage unit.

7. The target identification method based on digest algorithm feature encryption according to claim 1, characterized in that, For the features of the measurement data type extracted from the sampling information, the hierarchical information corresponding to the features of the measurement data is obtained according to the preset hierarchical principle, and the feature summary is obtained by the summarization algorithm based on the hierarchical information.

8. The target identification method based on digest algorithm feature encryption according to claim 1, characterized in that, After obtaining the tag information of the target to be identified, the method further includes: Delete the sampling information of the target to be identified.

9. A target matching method based on feature encryption using a digest algorithm, characterized in that, Includes the following steps: Feature extraction is performed on the sampled information of the target to be identified, and the extracted features are processed by a summarization algorithm to obtain a feature summary of the sampled information; The sampled information feature summary is matched with the pre-stored sample information feature summary to obtain the matching result.

10. A face recognition method based on feature encryption using a digest algorithm, characterized in that, The method includes: Obtain the facial image of the person to be identified; Feature extraction is performed on the facial images of the person to be identified, and the extracted features are processed by a summarization algorithm to obtain a feature summary of the sampled information; The sampled information feature summary is matched with the pre-stored sample information feature summary to obtain the tag information of the person to be identified; the tag information of the person to be identified is the person information corresponding to the successfully matched sample information feature summary.

Images