A method for unlocking a wearable camera device

By combining wearable camera devices with biometric recognition and symbol set unlocking methods, the problems of low security and insufficient specificity in existing technologies are solved, achieving a highly secure and specific unlocking process and simplifying user operation.

CN116049788BActive Publication Date: 2026-11-10SICHUAN CHENGTU JIEKE TECH CO LTD
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
CN202310107707.1
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-02-13
Publication Date
2026-11-10
Estimated Expiration
2043-02-13

AI Technical Summary

Technical Problem

Existing unlocking methods suffer from low security, insufficient specificity, and inconvenience of use. In particular, traditional digital passwords are complex, difficult to remember, and easy to forget, while biometric identification is easily stolen and imitated.

Method used

Wearable camera devices are used to extract and classify feature data by inputting one or more sets of symbols, combined with biometric identification information. Unlocking is performed using preset verification rules and template libraries, including biometric identification such as fingerprints, signatures, and palm prints, and verification of symbol information and security level combinations.

Benefits of technology

It improves the security and specificity of unlocking, reduces the risk of password theft, simplifies the unlocking process, adapts to the password strength needs of different users, and prevents viruses and network attacks.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN116049788B_ABST
    Figure CN116049788B_ABST
Patent Text Reader

Abstract

This invention relates to a method for unlocking a wearable recording device, comprising: inputting one or more symbol sets within the recording range allowed by the wearable recording device, the symbol sets including the user's biometric identification information and symbol information; responding to the acquired verification information related to the one or more symbol sets, the unlocking method of this invention can determine whether the unlocking conditions are met based on the degree of difference between the verification information and a preset template; when the verification information meets the preset verification rules, the wearable recording device performs unlocking. Based on the unlocking method of this invention, even if the verification information is known in advance, it can still achieve good confidentiality; even if the content of the verification information is identical, the legitimate user can be identified through a specific template of biometric information. The unlocking method of this invention has the advantages of high security, high specificity, and difficulty in being stolen and imitated.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of cryptography, and more particularly to an unlocking method, and more particularly to an unlocking method for a wearable camera device. Background Technology

[0002] Cryptography is a core technology and fundamental support for ensuring network and information security. Through its two core functions of encryption protection and security authentication, it can fully realize security requirements such as anti-counterfeiting, anti-leakage, anti-tampering, and non-repudiation, playing a crucial role as a "messenger," "guardian," and "gene" in cyberspace. In today's highly developed information, networked, and digital world, cryptography has permeated all aspects of social production and life. Important networks and information systems, critical information infrastructure, and digital platforms all rely on cryptographic protection. New technologies and business models such as 5G, the Internet of Things, cloud computing, big data, artificial intelligence, blockchain, quantum communication, and the digital economy are all closely integrated with cryptography. Cryptography is also closely related to people's daily lives. Identity authentication, consumer payments, online transactions, personal information protection, and property protection all rely on cryptography. The application of cryptography is ubiquitous, effectively maintaining the normal operation of society and transaction order, and protecting the legitimate rights and interests of citizens, legal persons, and social organizations.

[0003] The application of cryptography in security has expanded comprehensively. In sectors such as energy, water conservancy, education, broadcasting, and taxation, cryptography applications have continued to deepen, fully demonstrating its important role in safeguarding the interests of the people and businesses. The application of cryptography in various electronic payment methods, including bank cards, online banking, mobile payments, barcode payments, and non-bank payments, has been continuously deepened. The use of cryptography in identity documents, social security cards, emergency broadcasting, digital television, electronic toll collection (ETC) systems, transportation smart cards, and value-added tax invoice systems has achieved large-scale deployment, realizing the widespread application of cryptography.

[0004] Although various encryption technologies have emerged with the development of science and technology, and the complexity of passwords has been increasing, the excessive complexity of passwords makes it very inconvenient for users to remember and unlock them. The best passwords are those that are difficult for people to remember, which contradicts practicality. Therefore, current cryptographic technology is in a stalemate and contradiction. Furthermore, because the numbers, letters, and symbols used in traditional passwords are widely known, more and more passwords are being broken with the increase in computing power, leading to financial losses and social unrest.

[0005] Chinese patent CN104461336B discloses an electronic device and its unlocking method. This electronic device has a touchscreen, and the unlocking method includes the following steps: When the electronic device is in a locked state, it receives corresponding first type of input and second type of input. It obtains symbol information displayed on the touchscreen based on the first type of input and direction information based on the second type of input. When the symbol information and direction information meet preset unlocking conditions, the electronic device is switched to an unlocked state. When the symbol information and direction information do not meet the preset unlocking conditions, the electronic device remains in a locked state.

[0006] While the aforementioned existing technologies provide an unlocking method based on a combination of symbolic and directional information, in practical applications, if the preset unlocking symbolic or directional information is too complex, it becomes difficult for users to remember, consuming significant time during the unlocking process. Furthermore, errors in inputting the wrong symbolic or directional information during a single operation can lead to unlocking failure, and the probability of such failure is directly proportional to the complexity of the preset symbolic or directional information. Although complex combinations of symbolic or directional information can improve security to some extent, the resulting inconvenience far outweighs the benefits, significantly reducing efficiency and user experience. Conversely, if the preset unlocking symbolic or directional information is too simple, it fails to provide the desired effect. Such simple combinations are easily observed and stolen, compromising personal privacy and property security. Moreover, when the unlocking methods provided by the aforementioned technologies are stolen or misused, the security level can only be increased by changing the combination or complexity of the symbolic or directional information, without fundamentally addressing the inherent drawbacks of the existing passwords or unlocking methods.

[0007] Prior art CN115563596A discloses an unlocking control method and device. The method includes: detecting an unlocking request operation for an unlocking target; acquiring first authentication information; performing first authentication on the first authentication information to obtain a first authentication result; if the first authentication result fails, acquiring second authentication information, the second authentication information including the motion trajectory of a terminal device and the motion trajectory of a wearable product; performing second authentication on the second authentication information to obtain a second authentication result; and if the second authentication result passes, performing an unlocking operation on the unlocking target. This allows for unlocking authentication using the second authentication information, including the motion trajectory of the terminal device and the motion trajectory of the wearable product, even when the first authentication result fails due to the user wearing a mask and being unable to use fingerprints, passwords, gestures, etc. This reduces the number of steps required from the user and enables rapid unlocking of the target while protecting user privacy.

[0008] The aforementioned existing technology achieves rapid unlocking of the target by combining first and second authentication. However, this method, which relies on the failure of the first authentication to proceed with the second, does not adequately protect user privacy. It is easy for malicious actors to skip the first authentication process and directly proceed with the second. Furthermore, the second authentication information relies on the movement trajectories of the terminal device and wearable products to unlock the device. Since movement trajectory design must consider practicality, it cannot be overly complex, making it easily susceptible to theft. Moreover, movement trajectories lack specificity; different users can unlock the same device using the same movement trajectory, posing a significant security risk.

[0009] Therefore, there is an urgent need for a highly secure, specific, and easy-to-use unlocking method. Based on this background, the present invention proposes an unlocking method for wearable recording devices.

[0010] Furthermore, on the one hand, there are differences in understanding among those skilled in the art; on the other hand, the applicant studied a large number of documents and patents when making this invention, but due to space limitations, not all details and contents were listed in detail. However, this does not mean that the present invention does not possess the features of these prior art. On the contrary, the present invention already possesses all the features of the prior art, and the applicant reserves the right to add relevant prior art to the background art. Summary of the Invention

[0011] To address the shortcomings of existing technologies, this invention provides a method for unlocking wearable camera devices.

[0012] Preferably, the present invention provides a method for unlocking a wearable camera device, comprising:

[0013] Within the permitted recording range of the wearable recording device, input one or more sets of symbols;

[0014] In response to the acquired set of one or more symbols, the wearable camera device determines verification information associated with the set of one or more symbols.

[0015] The verification information includes the identity of the user associated with the one or more symbol sets and the matching degree between the one or more symbol sets and at least one preset symbol set associated with the user;

[0016] Specifically, when the verification information meets the preset verification rules, the wearable camera device unlocks.

[0017] Preferably, the one or more symbol sets include the user's biometric identification information and symbol information.

[0018] Preferably, the present invention provides a method for unlocking a wearable camera device that further includes: executing a single verification rule for a single verification item or a combination verification rule for different verification items based on the security level, wherein the security level includes a single verification level and a comprehensive verification level.

[0019] Preferably, the wearable recording device determines verification information related to the one or more symbol sets, including:

[0020] Perform feature data extraction and classification on the one or more symbol sets;

[0021] Different verification weights are assigned to the one or more sets of symbols in response to the degree of specificity carried / contained / represented by different feature data.

[0022] Preferably, the extraction of the feature data includes:

[0023] Extract key frames from the video of the user completing the verification information within a specified time;

[0024] Feature regions of the keyframe or a single image displayed by the user are extracted.

[0025] Preferably, the extracted verification information is compared with a preset template library, and different scores are assigned according to the degree of difference between different verification information and preset templates.

[0026] Preferably, the biometric identification information includes one or more of the following: fingerprint identification information, facial recognition information, skin chip identification information, gait identification information, iris identification information, vein identification information, retinal identification information, palm geometry identification information, voice and signature identification information, hand shape identification information, and signature identification information.

[0027] Preferably, a wearable recording device includes:

[0028] Within the permitted recording range of the wearable recording device, input one or more sets of symbols;

[0029] In response to the acquired set of one or more symbols, the wearable camera device determines verification information associated with the set of one or more symbols.

[0030] The verification information includes the identity of the user associated with the one or more symbol sets and the matching degree between the one or more symbol sets and at least one preset symbol set associated with the user;

[0031] Specifically, when the verification information meets the preset verification rules, the wearable camera device unlocks.

[0032] Preferably, the wearable recording device includes:

[0033] The processor is used to connect the various components or modules of the camera device and run or call programs or protocols stored in the memory to complete various corresponding functions.

[0034] The collector is used to collect the identification information provided by the user and send it to the processor for processing;

[0035] Storage, used to store software programs and data, and to store the program instructions required to complete the unlocking process;

[0036] Sensors are used to detect whether the recording device is being worn by the user.

[0037] Preferably, an unlocking system for a wearable camera device includes:

[0038] The identification unit is used to preprocess one or more sets of symbols input within the allowed range of the wearable camera device, i.e., to extract and classify feature data;

[0039] The analysis unit is used to perform analysis and comparison of the preprocessed feature data to determine the degree of difference between the feature data and the preset template, and to assign a score based on the degree of difference.

[0040] The execution unit is used to compare the unlocking threshold of the preset security level with the score, and determine the execution of unlocking the wearable camera device based on the score.

[0041] The beneficial technical effects of the present invention include: the security of the unlocking method for a wearable camera device provided by the present invention is greatly improved compared with the traditional unlocking methods. The unlocking is completed by collecting a set of symbols through the wearable camera device. The set of symbols includes the user's biometric identification information and symbol information. The unlocking password is no longer limited to numbers, letters, Chinese characters, etc., but includes symbols or biometric information of arbitrary shapes. This increases the randomness (selectable range) of the password from multiple dimensions, greatly increasing the difficulty of password cracking.

[0042] The unlocking method of the present invention has high specificity. The user's unlock password contains the user's own biometric features, which are specific. The user's biometric features can be signature features, handwriting signature features, and palm geometry features, from which unique features are extracted and converted into digital codes, and these codes are further combined to form a feature template. When the user performs identity authentication, the device equipped with the unlocking method of the present invention can identify the user's information, acquire the user's features and compare them with the feature template in the data center to determine whether there is a match, so as to decide to accept or reject the current user. Based on the unique biometric features between each individual, the unlocking method of the present invention that uses biometric feature identification information as verification unlocking information has very high specificity when used by different users. Traditional digital unlocking allows anyone to perform corresponding unlocking through a digital password, but the unlocking method of the present invention can still achieve a good confidentiality effect even when the preset symbol set for verifying the unlocking information is known in advance, that is, the unlocking method of the present invention that uses biometric feature identification information as verification unlocking information can configure a specific verification template according to the user's biometric feature information, and can identify a legal user through the specific template of the biometric feature information even if the content of the verification information is consistent. The unlocking method of the present invention has the advantages of high security, high specificity, and being not easy to be stolen and imitated.

[0043] The unlocking method of the present invention can also realize a desired security protection level and risk probability through free combination of verification items on a symbol set composed of biometric feature identification information and symbol information. Assuming that the risk probability of the unlocking method of the present invention that only compares whether the symbol images are consistent is P1 (0<P1<1), the risk probability of only comparing the signature trajectory is P2 (0<P2<1), and the risk probability of only comparing the signature handwriting is P3 (0<P3<1), then the risk probability P for simultaneously verifying the above three conditions is P=P1*P2*P3, which greatly reduces the risk and improves security. In addition, the user can set a combined password of multiple pieces of identification information as needed, and the number of combination items of the combined password depends on the user's settings, which is not limited by video recording, key information extraction and comparison methods, and can flexibly adapt to the password strength requirements of different users. If n pieces of comparison information are set to form a combined password, the risk probability is Pn, and the security will increase in a geometric progression.

[0044] This invention discloses a wearable recording device that is normally in a closed state. Upon detecting a data access request, it enters a password recording state. Only after the correct password is entered will the data connection be opened, allowing external devices to access the storage content of the recording device. Recording is the only password input method, eliminating the risk of password injection through networks or data interfaces. This avoids forced unlocking due to viruses and network attacks, increasing device security. Both the host computer and the terminal recording device disable all wireless communication methods such as WiFi, Bluetooth, infrared, GPS, and BeiDou, retaining only a USB interface to prevent information leakage wirelessly. The terminal recording device uses built-in storage space and does not have any memory card slot, preventing information leakage caused by other read / write methods. Attached Figure Description

[0045] Figure 1 This is a flowchart illustrating the unlocking method for a wearable camera device according to the present invention. Detailed Implementation

[0046] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below in conjunction with specific embodiments. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Therefore, the following detailed description of the embodiments of the present invention is not intended to limit the scope of the claimed invention, but merely represents selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0047] These and other features and advantages of the invention will be more fully understood through the following description of one or more embodiments of the invention in conjunction with the accompanying drawings.

[0048] In view of the prior art, the present invention proposes a method for unlocking a wearable camera device, comprising: inputting one or more symbol sets within the allowed recording range of the wearable camera device;

[0049] In response to the acquired set of one or more symbols, the wearable camera device determines verification information associated with the set of one or more symbols.

[0050] The verification information includes the identity of the user associated with the one or more symbol sets and the matching degree between the one or more symbol sets and at least one preset symbol set associated with the user;

[0051] Specifically, when the verification information meets the preset verification rules, the wearable camera device unlocks.

[0052] According to a preferred embodiment, one or more sets of symbols include the user's biometric identification information. Because traditional digital passwords are easily forgotten or their predictable patterns make them easy to steal, highly specific and difficult-to-steal biometric identification information is used as one of the verification and unlocking information in this invention. Biometric identification technology refers to the technology of using computers to authenticate an individual's identity based on their inherent physiological characteristics (fingerprints, irises, facial features, DNA, etc.) or behavioral characteristics (gait, keystroke habits, etc.). Each individual has unique measurable or automatically identifiable and verifiable physiological characteristics or behavioral patterns, i.e., biometrics. These can be divided into physiological characteristics (such as fingerprints, facial images, irises, palm prints, etc.) and behavioral characteristics (such as gait, voice, handwriting, etc.). Biometric identification is the process of identifying and authenticating an individual based on their unique biometric characteristics. Biometric identification information includes: fingerprint recognition information, facial recognition information, skin chip recognition information, gait recognition information, iris recognition information, vein recognition information, retinal recognition information, palm geometry recognition information, DNA recognition information, voice and signature recognition information, hand shape recognition information, signature recognition information, etc.

[0053] According to a preferred embodiment, the biometric identification information used in this invention can be signature identification information. Signature identification falls under the category of behavioral identification. Signing is a traditional means of identity authentication. Modern signature identification technology mainly identifies the signer by measuring the shape of the signature and the speed, sequence, and pressure characteristics between different strokes. When a user unlocks the device using the unlocking method of this invention, the device equipped with the unlocking method can record the user's signing process within the allowed recording range, and identify the user's signature by capturing continuous video or extracting key frames. The speed between different strokes can be determined by the signing time and the completion degree of the strokes in the key frames; the signing sequence can be determined by comparing the signature image in the extracted key frames with a preset verification template; and the signing pressure characteristics can be determined by the area of ​​ink splatter at key turning points in the signature. The weight of the judgment indicators can also be adaptively adjusted according to different signature content. For example, when judging more complex signature content, the weight of the user's signing sequence indicator should be adaptively increased to ensure the accuracy of the judgment result.

[0054] According to a preferred embodiment, the biometric identification information used in this invention can be signature identification information. Although a personal handwritten signature is not an outward manifestation of an inherent physiological characteristic, since signing is essentially a free wrist movement, each person has their own characteristics in terms of pen grip, pen stroke habits, and the amount of force applied. Research shows that handwritten signatures are a kind of "ballistic motion," an unconscious habitual action. Handwritten signature authentication compares the current signature with pre-stored signature samples or templates based on the characteristics of the individual's handwriting, judging the authenticity of the signature and correspondingly verifying the identity of the current signer. When a user unlocks the device using the unlocking method of this invention, the device equipped with the unlocking method can record the user's signing process within the allowed recording range, and identify the user's signature by capturing continuous video or extracting keyframes. During the signature recognition process, the unlocking method of this invention can verify the user's identity by comparing the user's signature according to feature points. User signatures can be categorized into start points, end points, curve points, and intersection points based on their feature points. After normalizing the extracted feature points, they are matched against a preset template to obtain the verification result. Furthermore, compared to other biometric authentication methods, the most significant advantage of signature authentication is that traditional handwritten signatures are frequently used in daily life, and handwriting is a universally possessed and used skill, making it easily accepted. With the development of sensor and computer technologies, computers and electronic devices supporting handwriting (writing tablets, personal digital assistants, smartphones, etc.) are becoming increasingly widespread, greatly facilitating the application of handwritten signature authentication.

[0055] According to a preferred embodiment, the biometric identification information used in this invention can be palm geometry identification information. Palm geometry identification involves measuring the physical characteristics of the user's palm and fingers for identification; advanced products can also recognize three-dimensional images. As an established method, palm geometry identification is not only high-performing but also relatively easy to use. It is suitable for situations where there are many users, or where users, although not frequent users, find it easy to accept. If needed, this technology can achieve very high accuracy and its performance can be flexibly adjusted to adapt to a wide range of usage requirements. Handprint readers have a wide range of applications and are easily integrated into other systems, thus becoming the preferred technology in many biometric identification projects. Palmprints refer to the various lines on the surface of the palm; the specific arrangement and combination of these lines form the uniqueness of the palmprint. Palmprint features can be used for identification, such as main lines, creases, papillary ridges, and triangular points. Compared to fingerprints, palmprints can extract more features and contain richer information. Therefore, theoretically, palmprints have a higher recognition rate than fingerprints. Furthermore, the shape of palm prints is innate, primarily determined by genetics. Even if the palm is damaged and the ridges fall off, the newly formed palm ridges retain their original shape, exhibiting high stability. When a user unlocks their hand using the unlocking method of this invention, the device equipped with this unlocking method can record the user's palm within the permitted recording range, and then identify the user using the recorded palm image. When a user unlocks using palm print recognition, the recording device equipped with this unlocking method can capture the user's palm print through recording, extract corresponding features after image preprocessing, and obtain the recognition result by comparing it with templates in the database. Preferably, the palm print ROI extraction uses a reference coordinate system algorithm, establishing a reference coordinate system using the valley points between fingers, and then extracting ROI palm print information according to a certain distance criterion. The algorithms for palm print feature extraction during the unlocking process can be divided into two categories: algorithms based on manual design and algorithms based on feature learning. This invention uses an algorithm based on feature learning to extract features from the user's palm print. Algorithms based on feature learning mainly utilize deep neural networks to learn more complex filter structures from data, achieving automatic extraction of palm print features. Deep neural network models consist of interconnected layers. A preprocessed palmprint image is input into the input layer, and after processing through multiple layers of the neural network, the corresponding feature vector is finally obtained. Feature-based learning algorithms leverage the advantages of big data and the powerful non-linear learning capabilities of deep neural networks to extract richer and more robust palmprint features.

[0056] The unlocking method of this invention can sample biometric features, extract their unique characteristics and convert them into digital codes, and further combine these codes to form a feature template. When a user authenticates their identity, the device including the unlocking method of this invention can identify the user's information and obtain their characteristics, comparing them with the feature template in the database to determine whether there is a match, thereby deciding whether to accept or reject the current user.

[0057] Based on the unique biometric characteristics of each individual, this invention's unlocking method, which uses biometric identification information as verification, exhibits high specificity across different users. While traditional digital unlocking allows anyone to unlock using a digital password, this invention's method maintains good security even when the preset symbol set for verification information is known in advance. That is, this invention's unlocking method, using biometric identification information as verification, can assign a specific verification template based on the user's biometric information. Even if the technical verification information is identical, the specific template of the biometric information can be used to identify the legitimate user. This invention's unlocking method has the advantages of high security, high specificity, and difficulty in being stolen or imitated.

[0058] According to a preferred embodiment, one or more symbol sets include symbolic information. Because traditional digital cryptography suffers from problems such as short and simple passwords being easily cracked, and long and complex passwords being difficult to remember, symbolic information with a wide range, strong randomness, and high specificity is used as one of the verification and unlocking information in this invention. It should be noted that the symbolic information used in this invention should include various mainstream or niche characters, numbers, letters, patterns, codes, chemical symbols, traffic signs, etc., currently used in human society. Since the range of symbol sets used in this invention is much larger than that of traditional digital cryptography, and the combination forms and methods of the symbol sets used in this invention are exponentially greater than those of traditional digital cryptography, the security of the unlocking method of this invention is greatly improved. Furthermore, when unlocking using the unlocking method of this invention, the user only needs to display a pre-set symbol pattern in front of a recording device equipped with the unlocking method of this invention to complete the unlocking process. The user can display the pre-set symbol pattern by waving it in the air, using a pen, or other medium to express the symbol pattern on another medium. The symbol pattern displayed by the user is verified through dynamic capture and image recognition to complete the unlocking process.

[0059] According to a preferred embodiment, the present invention proposes a method for unlocking a wearable camera device, comprising: S100, acquiring at least one set of symbol recognition information of the user on the terminal; specifically, when the terminal is in a sleep state, detecting the user's operation on the terminal (e.g., picking up or moving the terminal), and acquiring at least one set of symbol recognition information input by the user through the camera function. Further, the at least one set of symbol recognition information includes one or a combination of symbol recognition information and biometric recognition information. Even further, symbol recognition information refers to various mainstream or niche characters, numbers, letters, patterns, codes, chemical symbols, traffic signs, etc., currently used in human society; biometric recognition information refers to fingerprint recognition information, facial recognition information, skin chip recognition information, gait recognition information, iris recognition information, vein recognition information, retinal recognition information, palm geometry recognition information, DNA recognition information, voice and signature recognition information, hand shape recognition information, signature recognition information, etc.

[0060] For example, the user's symbol recognition information can be obtained in any of the following ways: one way is to record the process of the user writing according to a preset set of symbols using a recording device, or to capture and record the biometric recognition information (such as palm print) displayed by the user. After the recording of the user's symbol recognition information is completed within a specified time, the user's symbol recognition information is extracted in the terminal recording device by preprocessing the video / image or extracting key frames.

[0061] Furthermore, different information extraction methods should be used for different symbol recognition information provided by the user. Specifically, when dealing with signature recognition information, the unlocking method of this invention can extract the biometric recognition information provided by the user by extracting key frames from the video of the user completing the signature / signature process within a specified time. Depending on the content of the signature / signature, instantaneous images reflecting the speed, order, pressure, start point, end point, bending point, and intersection point of different strokes are extracted and compared with preset templates in the database. In addition, when dealing with palmprint recognition information or a single symbol information provided by the user, the image displayed by the user (including: the user's palm image or the image of the completed symbol) can be captured by a recording device equipped with the unlocking method of this invention, and the captured single image can be compared with preset templates in the database. It should be noted that palmprint information, as biometric recognition information, has sufficient specificity. The palmprint contains a wealth of comparable distinguishing features, meaning that user identification can be achieved through a single captured image, or in some usage scenarios with relatively low security requirements, unlocking can be completed simply by comparing the final completed symbol image.

[0062] According to a preferred embodiment, based on the signature recognition information provided by the user, the unlocking method of the present invention can control whether the recording device records the image displayed by the user through a specific symbol preset by the device owner. That is, recording can only begin, pause, or stop after the recording device recognizes the specific symbol preset by the device owner. This further enhances the security and specificity based on the user's signature recognition information, and the addition of the preset specific symbol makes the unlocking process more reliable and stable, avoiding the phenomenon of insufficient user information being recognized or extracted due to recording ending too early or too late, thus preventing unlocking failure. Specifically, the specific symbol preset by the device owner can trigger a corresponding response from the recording device after recognition. For example, the user-preset "√" symbol indicates the start of recording; the recording device will begin recording when it recognizes the "√" symbol drawn by the user on paper. The user-preset "O" symbol indicates the pause of recording; the recording device will pause recording when it recognizes the "O" symbol drawn by the user on paper. The user-preset "X" symbol indicates the end of recording; the recording device will end recording when it recognizes the "X" symbol drawn by the user on paper. When unlocking with a signature, the user first draws a "√" on paper or other media for the recording device to recognize. After recognizing the start recording symbol, the recording device begins recording and can remind the user to start writing the signature via a configured electronic prompt. During the recording of a user's handwriting, the user can pause recording by drawing an "O" symbol on paper or other media. Recording resumes when the recording device recognizes the "O" symbol again; that is, the pause state is unlocked when the recording device recognizes an even number of "O" symbols, and paused when it recognizes an odd number. Recording stops when the recording device recognizes an "X" symbol displayed by the user on paper or other media. Preferably, the signature recognition content can be either the user's signature or other preset words.

[0063] Furthermore, when using a signature recognition method with preset specific identification symbols, the recording device only needs to detect whether the specific identification symbol displayed by the user is correct to achieve initial screening of the owner's information. That is, if the recording device does not detect the preset specific identification symbol, it assumes that the user is not the owner and will not start the recording function, which improves recognition efficiency and reduces memory usage. Recording can only begin after the recording device detects that the specific identification symbol displayed by the user is a preset symbol, and the signature words and handwriting are compared after the recording is completed. Preferably, the recording device equipped with the unlocking method of the present invention can match a pre-stored model trained on a large number of owner signature recordings in continuous video frames based on artificial intelligence training (when the owner is right-handed, the right eye image is the primary recognition image, and the left eye image is the auxiliary recognition image). When the recording device recognizes that the signature content and handwriting displayed by the user match the pre-stored model, the recording device can unlock the device. Specifically, a user's signature characteristics can be identified through the initial stroke, the path of the handwriting, the final stroke, and the arrangement of different strokes during the signature process. The recording device can identify the initial and final strokes of different users writing the same content. For example, if the signature is in Chinese characters, each character can generate at least one set of initial and final stroke feature information for the unlocking method of this invention. The recognition areas for the initial and final strokes are divided according to the structural features of the Chinese characters. For example, if the Chinese character has a top-bottom structure, it should include at least an upper and lower recognition area; if the Chinese character has a left-right structure, it should include at least a left and right recognition area. By comparing the user's initial and final stroke areas with preset target areas, it can be determined whether the user's writing habits match the device owner's.

[0064] The signature unlocking function of the recording device can respond to the feature recognition of the writing path of different users for the same written content. For example, when the signature content is a symbol, each symbol can generate at least specific path recognition information. That is, when writing a circular symbol, the user's signature feature can be identified based on whether the user uses a clockwise or counterclockwise path. Similarly, when a user writes Chinese characters for signature, different users will have different writing paths for the same Chinese characters, such as from top to bottom, from left to right, or from the middle to both sides. Based on the differences in these writing habits, the signature feature of the device owner can be obtained, and by comparing the signature path, it can be determined whether the user is the device owner.

[0065] Signature unlocking of recording devices can be based on the arrangement and combination of different strokes during the signing process. For example, when the signature content is Chinese characters, due to the richness of strokes in Chinese characters, different users will use different combinations of different strokes when writing Chinese characters. For instance, when writing Chinese characters with "three horizontal and three vertical strokes" (the three horizontal and three vertical strokes are divided into upper horizontal, middle horizontal, and lower horizontal strokes, and the three vertical strokes are divided into left vertical, middle vertical, and right vertical strokes), there are 720 possible arrangements of different strokes without connected strokes, and the arrangements of different strokes are more complex when connected strokes are present. Therefore, it can be seen that identifying user signature information by the arrangement and combination of different strokes has high specificity and good security. The unlocking method of this application can pre-store several stroke arrangements commonly used by the device owner in the database, and compare them with the arrangement and combination of different strokes in the signature content displayed by the user to determine whether the user is the device owner.

[0066] According to a preferred embodiment, the unlocking method of this application can also extract some detection parameters for owner identification and unlocking in environments without a writing surface. Specifically, it can identify whether the user is the owner by recognizing their hand features when signing. These features are unique to the owner and can be extracted from the owner's signature video recording, allowing the owner to verify their signature by writing in the air with a pen even in scenarios without a writing surface. Specifically, the hand features are extracted during the owner's signature using a precise finger movement segment extraction algorithm. First, a pre-stored model trained with a large number of pre-collected owner signature videos is used to collect subtle finger movements during the signature. A data preprocessing algorithm filters noise from the original data to obtain effective acceleration signal data generated by the finger movements. Simultaneously, a data sequence segmentation algorithm is used to extract action segments, dividing the continuous acceleration signal data sequence generated by the finger movements into several data segments of equal length. Acceleration and mean thresholds are used to filter and extract data segments containing complete finger movements, automating the extraction of finger movement segments and allowing the recording device to actively recognize the movements. Secondly, a finger movement acceleration feature value selection method is used. The finger movement fragments extracted by the recording device still retain the original data form at the time of acquisition. Feature value extraction is required to obtain the acceleration data feature values ​​within the action window. A feature value selection algorithm is then used to select features with excellent data representation capabilities from all acceleration data feature values ​​of the finger movement, reducing system overhead. Finally, signature unlocking in scenarios without a writing surface is achieved based on the finger movement features recognized during user signature recognition. In this solution, a finger movement recognition classification model is constructed using a large training set of finger movement samples combined with a classification algorithm, and this classification model is stored in the recording device application. Simultaneously, the selected data feature values ​​are used as input to the finger movement recognition classification model, enabling the recording device to correctly understand and recognize different finger movements, obtain recognition results, and then determine whether to proceed with unlocking.

[0067] S101. Compare the extracted user-provided at least one set of symbol recognition information with a preset template to determine whether the score of the at least one set of symbol recognition information reaches a preset threshold. Specifically, when the score of the at least one set of symbol recognition information reaches the preset threshold, proceed to S102; when the score of the at least one set of symbol recognition information does not reach the preset threshold, the process ends.

[0068] After obtaining comparable information through the above information extraction methods, the comparable information is compared with a preset template in the database, or by extracting key frames, or by extracting predetermined key distinguishing feature points from the symbol recognition information. The symbol recognition information provided by the user is compared with the preset template, and a final comparison score is given. The final comparison score is compared with different thresholds preset according to different security levels to determine whether to unlock.

[0069] According to a preferred embodiment, the method for assigning a score based on symbol recognition information in S101 includes: collecting N sets of user information recognition samples before proceeding to S100; constructing a database of preset templates based on each set of collected information recognition samples; continuously improving the compatibility between the database templates and the users through a feature learning algorithm; establishing standard state parameters and their corresponding standard scores based on the preset template database; reducing the score accordingly based on the deviation between the recognition information and the preset template; and finally, determining whether to unlock based on a comparison of the comprehensive score and a preset threshold.

[0070] Furthermore, in this embodiment, N can be 200. In other embodiments, the value of N can be set according to the desired matching degree between the user and the unlocking method. This invention does not impose any specific limitations here. The user's information identification sample is not limited to a specific identification information, but can include all identification information reflecting the user's characteristics, including: palm geometry identification information, signature identification information, symbol identification information, etc. The comparable information extracted from the user's identification information by the recording device equipped with the unlocking method of this invention is compared with a preset template in the database. Based on the similarity between the extracted comparable information and the preset template, three levels are obtained and different scores are assigned. The three similarity levels are: Level 1 (same), Level 2 (similar), and Level 3 (different). Different scores are assigned according to different levels: Level 1 is 9 points, Level 2 is 5 points, and Level 3 is 1 point. A comparison information matrix is ​​established by comparing different comparison items and the scores obtained after comparing different comparison items with a preset template. Different weights are assigned according to the importance of different comparison items and their ability to reflect the specificity of the user's comparison information. Finally, the sum of the product of the weights and scores of different comparison items is calculated and compared with a preset threshold. If the final total score is greater than or equal to the threshold, unlocking is performed; otherwise, unlocking fails.

[0071] According to a preferred embodiment, taking signature recognition information as an example, in the signature-based unlocking method of this invention, the comparison items reflecting the specific characteristics of the user's signature can be the start point, end point, bend point, and intersection point of the signature. Weights are assigned to these four comparison items based on their importance and ability to reflect the specificity of the user's comparison information: the start point has a comparison weight of 25%, the end point has a comparison weight of 10%, the bend point has a comparison weight of 35%, and the intersection point has a comparison weight of 30%. The final total score is obtained by calculating the specific score values ​​of each comparison item during the comparison process. It should be noted that the level division and the corresponding level scores and weight ratios can be set according to the desired level of security, and this invention does not impose specific limitations on this.

[0072] Furthermore, the unlocking method of the present invention is applicable to unlocking scenarios with different security levels. It executes a single verification rule for a single verification item or a combination of verification rules for different verification items based on different security levels. The security levels include single verification levels and comprehensive verification levels. Depending on the required security level, the unlocking method of the present invention can provide an unlocking method using a single verification rule or an unlocking method using a combination of verification rules. The single verification rule unlocking method and the combination verification rule unlocking method can be implemented through different unlocking steps and preset unlocking thresholds. Specifically, when the security level is low, the unlocking method of the present invention should use a single verification rule. The single verification rule unlocking method involves comparing the user's input identification information with a template in a preset database to obtain a final score, i.e., the difference between the user's identification information score and the standard score. Unlocking is completed when the user's identification information score is greater than or equal to the preset standard score. When the security level is high, the unlocking method of the present invention should use a combination of verification rules. The combined verification rule unlocking method compares at least two sets of user-input identification information with templates in a preset database. Only after the first set of user-provided identification information reaches a preset standard score can the next set of identification information be compared. Unlocking is only performed after all identification information meets the unlocking standard score. Furthermore, with the combined verification rule unlocking method, unlocking will fail if even one set of identification information fails to meet the standard score, thus increasing the unlocking requirements and security level. Different combinations of verification conditions create different strengths, reducing risk.

[0073] Furthermore, users can pre-set combined verification conditions according to their desired security levels, namely the verification order and standard score of different comparison items in the combined verification rule. In the combined verification rule unlocking method, the number of comparison items depends on the user's security requirements and is not limited by video recording, key information extraction, or comparison methods, thus flexibly adapting to the password strength needs of different users. Assuming the probability of unlocking by comparing only one piece of user-provided identification information is P, then in the comprehensive comparison principle unlocking method, the probability of unlocking using n pieces of identification information is Pn, thus the security increases exponentially.

[0074] Based on the above embodiments, this disclosure also provides a wearable recording device equipped with the unlocking method of the present invention. The recording device includes a processor, a data acquisition unit, a storage unit, and sensors. The processor is the control center of the wearable recording device. The processor can connect and integrate various components or modules of the recording device through various interfaces and lines. The processor can run or call programs or protocols stored in the memory to complete various corresponding functions, such as assigning values ​​to user identification information, and implementing corresponding unlocking or locking functions based on the user's identification information score. The processor can call and process data stored in the memory to assign and compare user-provided identification information. The processor controls and monitors the entire recording device through connections with various components or modules. Preferably, the processor includes one or more processing units.

[0075] According to a preferred embodiment, the processor used in the unlocking method of the present invention can be composed of an arithmetic logic unit (ALU) and a control unit. The ALU includes: an arithmetic logic unit (ALU), a floating-point unit (Floating-point ALU), a general-purpose register set, and special-purpose registers. The control unit includes: an instruction controller, a timing controller, a bus controller, and an interrupt controller. The processor used in the present invention incorporates out-of-order execution technology, enabling the processor to execute several instructions out of sequence, thereby achieving full-load operation of the processor's internal circuitry and correspondingly increasing the speed of program execution.

[0076] According to a preferred embodiment, the unlocking method of the present invention uses a memory that can store software programs and data. In this embodiment, the memory can store user-provided identification information and preset comparison templates, and can also store program instructions required to complete the unlocking process. This allows the processor to execute various functional applications and data processing while running the software program stored in the memory. For example, in the unlocking method of a wearable recording device provided in this embodiment, the memory mainly consists of a program storage area and a data storage area. The program storage area can store the operating system, applications (unlocking program, comparison program, etc.), etc. The data storage area can store user-provided comparison identification information and preset identification information template datasets, etc. Furthermore, the recording device uses built-in storage space and does not have any memory card slots to prevent information leakage caused by other read / write methods.

[0077] According to a preferred embodiment, the collector is connected to the processor. The collector may include a video collector, an image collector, etc., for collecting identification information provided by the user and sending it to the processor for processing to generate comparison information that can unlock the device. The processor then determines whether the device can be unlocked by comparing the comparison result with a preset template in the memory.

[0078] According to a preferred embodiment, the sensor may include a light sensor, a motion sensor, and other sensors. The light sensor may include an ambient light sensor, which can detect whether the recording device is being worn by a user based on the ambient light level. The motion sensor is a motion sensing sensor that can detect whether the user needs to unlock the device to use it based on changes in position when the user picks up the recording device.

[0079] According to a preferred embodiment, the unlocking method for a wearable recording device provided by this invention can achieve functions including but not limited to the following: Video capture function: The device has one-click power-on, automatic recording, and one-click power-off functions, enabling automatic recording upon power-on and supporting taking photos during recording. Audio capture function: The terminal device has a separate audio recording function to save storage space. Image capture function: The terminal device has a separate photo-taking function to adapt to temporary scenarios or special purposes. Data encryption function: The terminal device supports multiple encryption algorithms such as AES, SM2, SM4, RSA, and DES, which can flexibly adapt to user needs and achieve encrypted storage of collected information. Identity recognition function: The terminal device has an identity recognition function; only by connecting to a legitimate host computer can a data transmission channel be established. If an illegitimate host computer is connected, a data transmission channel cannot be established, and the device automatically enters a locked state and cannot be operated. Only a legitimate host computer can unlock the locked device. Wireless shielding function: The terminal device supports shielding multiple wireless methods such as WiFi, Bluetooth, infrared, GPS, and Beidou at the system level, supporting user customization to meet user needs while preventing wireless leaks. Alarm function: When the battery is low or the storage space is almost full, the terminal device can remind the user to charge the battery or export files to clear space via voice.

[0080] According to a preferred embodiment, the wearable recording device provided by this invention has a host computer function. That is, the terminal recording device is connected to a host computer and terminal management software is installed to achieve the following functions: Terminal device management function: enabling device naming, numbering, and other functions for convenient and precise device management. Terminal device maintenance function: configurable according to user needs, clearing the terminal device's storage space, including automatic clearing after successful video import or clearing according to a specified time period. File import function: configurable according to user needs, enabling automatic import of newly captured videos or import of videos within a specified time range. File management function: users can quickly find files based on search criteria such as time, device number, and device issuing unit. Direct browsing function for encrypted files: the host computer software has a built-in dedicated audio and video player that can directly play encrypted videos without generating decrypted temporary audio and video files, reducing the risk of leakage; the host computer software also has a built-in dedicated image browser that can directly browse encrypted images without generating decrypted temporary image files, reducing the risk of leakage. Multi-terminal connection function: This device is a 1-to-many USB adapter, enabling a single host computer to connect to and manage multiple terminals simultaneously, improving file transfer efficiency and reducing working time. For example, if a host computer has two spare USB ports, using a 1-to-4 adapter, it can manage eight terminals simultaneously.

[0081] It should be noted that the specific embodiments described above are exemplary. Those skilled in the art can devise various solutions inspired by the disclosure of this invention, and these solutions all fall within the scope of this invention and its protection. Those skilled in the art should understand that this specification and its accompanying drawings are illustrative and not intended to limit the scope of the claims. The scope of protection of this invention is defined by the claims and their equivalents. This specification contains multiple inventive concepts; terms such as "preferredly," "according to a preferred embodiment," or "optionally" indicate that the corresponding paragraph discloses an independent concept. The applicant reserves the right to file divisional applications based on each inventive concept.

Claims

1. A method for unlocking a wearable camera device, characterized in that, include: Within the recording range permitted by the wearable recording device, the user inputs one or more symbol sets by waving in the air or expressing symbolic patterns on a medium; in response to the acquired one or more symbol sets, the wearable recording device determines verification information associated with the one or more symbol sets, wherein the step of determining the verification information includes: The hand features of the account holder during signature are extracted by executing an algorithm that precisely extracts finger movement fragments. Feature collection is performed using a pre-stored model trained on a large number of pre-collected owner signature videos, capturing subtle finger movements when users sign. By filtering out noise from the raw collected data through data preprocessing algorithms, effective acceleration signal data generated by finger movements can be obtained. Meanwhile, a data sequence segmentation algorithm is used to extract action segments, which segments the continuous acceleration signal data sequence generated by finger movements to obtain several data segments of equal length. Data segments containing complete finger movements were extracted by filtering using acceleration and mean thresholds. The finger motion acceleration feature value selection method is used. The extracted finger motion segment has the original data form at the time of acquisition. The acceleration data feature value within the window of the finger motion segment is obtained by feature value extraction. The selected data feature values ​​are used as input to the finger movement recognition and classification model; The verification information includes the identity of the user associated with the one or more symbol sets and the matching degree between the one or more symbol sets and at least one preset symbol set associated with the user; Specifically, when the verification information meets the preset verification rules, the wearable camera device unlocks; The wearable camera device has an identity recognition function, wherein: When the wearable camera device is connected to the host computer, a data transmission channel is established only when the host computer is a legitimate host computer; When the host computer is an illegal host computer, no data transmission channel is established, and the wearable camera device automatically enters a locked state so that it cannot perform any operation; After the wearable recording device is locked, it can only be unlocked through the legitimate host computer. Specifically, feature data extraction and classification are performed on the one or more symbol sets, and different verification weights are assigned to the one or more symbol sets in response to the specificity of different feature data carried / contained / represented.

2. The unlocking method for a wearable recording device according to claim 1, characterized in that, The method further includes: executing a single verification rule for a single verification item or a combination of verification rules for different verification items based on the security level, wherein the security level includes a single verification level and a comprehensive verification level.

3. The unlocking method for a wearable recording device according to claim 1, characterized in that, The extraction of the feature data includes: Extract key frames from the video of the user completing the verification information within a specified time; Feature regions of the keyframe or a single image displayed by the user are extracted.

4. The unlocking method for a wearable recording device according to claim 1, characterized in that, The extracted verification information is compared with the preset template library, and different scores are assigned according to the degree of difference between different verification information and the preset template.

5. A wearable camera device, characterized in that, A method for unlocking a wearable camera device as described in any one of claims 1 to 4, comprising: Within the permitted recording range of the wearable recording device, input one or more sets of symbols; In response to the acquired set of one or more symbols, the wearable camera device determines verification information associated with the set of one or more symbols. The verification information includes the identity of the user associated with the one or more symbol sets and the matching degree between the one or more symbol sets and at least one preset symbol set associated with the user; Specifically, when the verification information meets the preset verification rules, the wearable camera device unlocks.

6. A wearable recording device according to claim 5, characterized in that, The wearable camera device includes: The processor is used to connect the various components or modules of the camera device and run or call programs or protocols stored in the memory to complete various corresponding functions. The collector is used to collect the identification information provided by the user and send it to the processor for processing; Storage, used to store software programs and data, and to store the program instructions required to complete the unlocking process; Sensors are used to detect whether the recording device is being worn by the user.

7. An unlocking system for a wearable camera device, characterized in that, A method for unlocking a wearable camera device as described in any one of claims 1 to 4, comprising: The identification unit is used to preprocess one or more sets of symbols input within the allowed range of the wearable camera device, i.e., to extract and classify feature data; The analysis unit is used to perform analysis and comparison of the preprocessed feature data to determine the degree of difference between the feature data and the preset template, and to assign a score based on the degree of difference. The execution unit is used to compare the unlocking threshold of the preset security level with the score, and determine the execution of unlocking the wearable camera device based on the score.

Citation Information

Patent Citations

  • Electronic device and unlocking method thereof

    CN104461336B

  • Unlocking control method and device

    CN115563596A

  • Man-machine interaction integrated device based on Leap Motion equipment

    CN106774938A

  • Unlocking method and device for wearable equipment, equipment and storage medium

    CN108804006A