A passive optical blood relative feature collection and automatic matching method and system based on a light guide film

Abstract

This invention discloses a passive optical kinship feature acquisition and automatic matching method and system based on a photoconductor film, belonging to the field of optical acquisition and public welfare kinship search technology. This invention utilizes a passive photoconductor film attached to or integrated into a terminal screen to non-contactly acquire optical features such as human skin spectrum, subcutaneous blood vessels, and skin texture under user-initiated triggering. These features are encrypted locally to generate feature codes and then uploaded to the cloud. The cloud uses an AI kinship comparison algorithm to compare the similarity of the feature codes. If a preset threshold is reached, a suspected kinship is identified, and a warning message is pushed. After user verification, assistance can be provided in establishing contact with relatives. This invention reuses existing photoconductor film hardware, requires no dedicated acquisition equipment, and is compatible with various terminals such as mobile phones, tablets, smartwatches, computers, and shopping mall touch screens. It supports multiple kinship comparison algorithms and features non-invasiveness, privacy security, autonomous controllability, high accuracy, and ease of adoption. It effectively solves the problems of high cost, low efficiency, and poor privacy in traditional kinship search methods, combining technical practicality with social welfare value.

Description

Technical Field

[0001] This invention relates to the fields of optical data acquisition, biometric identification, and public welfare family tracing technologies. Specifically, it relates to a passive optical blood relative feature acquisition and automatic matching family tracing method and system based on a photoconductor film. Relying on the applicant's previous patented technology, it eliminates the need for additional dedicated equipment, enabling the active acquisition, encrypted uploading, and AI comparison of passive optical features of the human body with blood relative similarity. This solves the problem of finding relatives of abducted persons in situations where there is no information or contact, balancing public welfare value with technological practicality. The technical solution is adaptable to various terminal devices with screen display functions, compatible with the implementation of the photoconductor film being independently attached or integrated into the terminal device, and compatible with multiple blood relative feature comparison algorithms, effectively avoiding the risk of technical bypass. Citation of prior application

[0002] This application is based on four unpublished prior patent applications filed by the applicant. The filing dates of these prior basic patents are all earlier than this application, and they were not published before the filing date of this application. Therefore, they do not constitute prior art for this application. Their core technical features are incorporated into this application through reference. Specifically, the scope of the reference includes the technical content directly related to the technical solution of this application in each of the prior patent applications (including photoconductive film structure, encryption technology, communication protocol, AI instruction parsing, etc.). All prior patent applications and this application belong to the same applicant and there is no conflict of ownership. The referenced content consists of core technologies independently developed by the applicant. Specific reference information is as follows: Prior patent application 1: Application No. 2026102814971, Application Date 2026-03-10, Invention Title: A method for implementing permissionless AI interaction based on photoconductive tempered glass film; Prior patent application 2: Application No. 2026103109757, Application Date 2026-03-14, Invention Title: An AI Instruction Intelligent Execution System Based on an External Interaction Layer; Prior patent application 3: Application No. 2026103269949, Application Date 2026-03-17, Invention Title: Unauthorized Eye Tracking Touch Terminal Control Method and System Based on Optical Guide Interaction Layer; Prior patent application 4: Application No. 2026103373270, Application Date 2026-03-19, Invention Title: A Distributed Butler System and Method Based on Multi-Terminal Collaboration. Background Technology

[0003] Currently, reunification for abducted individuals still faces numerous challenges: abducted individuals often lack identity and kinship information, relying solely on vague memories and physical characteristics to locate relatives; relatives are also unable to obtain information about the abducted individual's location and current situation, making the search process time-consuming, labor-intensive, and extremely costly. Existing search technologies primarily rely on DNA comparison and facial recognition. DNA comparison requires blood sampling, which is invasive, costly, and difficult to implement on a large scale; facial recognition depends on clear facial images, is easily affected by makeup, age changes, and occlusion, resulting in limited accuracy, and requires abducted individuals and their relatives to actively register and upload information, failing to achieve convenient and efficient coverage for reunification.

[0004] The applicant's previously filed, unpublished patent applications have achieved core technologies such as passive optical acquisition, low-power communication, encrypted storage, and multi-terminal collaboration of photoconductor films. Among them: passive optical acquisition and photoconductor film physical structure are based on prior application patent 1; encrypted storage and AI recognition technology are based on prior application patent 3; multi-terminal communication and data transmission technology are based on prior application patent 4; and cross-terminal adaptation and instruction scheduling technology are based on prior application patent 2. Based on the aforementioned patented technologies, this application reuses existing photoconductor film hardware and core technologies to construct a contactless, highly accurate blood relative feature matching and family tracing system. This system addresses the shortcomings of existing family tracing technologies, allowing abducted individuals and their relatives to actively operate the system, using a photoconductor film attached to the terminal screen or integrated into the terminal device to complete feature collection and uploading, and automatic matching. This significantly improves the efficiency and reduces the cost of family tracing, possessing extremely high social value. Furthermore, addressing the risk of existing technologies being circumvented, this application constructs a comprehensive protection barrier by broadening the terminal compatibility range, covering core algorithm logic, and being compatible with various implementation forms of photoconductor films. This prevents competitors from circumventing protection by changing terminal types, modifying algorithms, or altering the installation method of the photoconductor film. Summary of the Invention

[0005] (a) Purpose of the invention The purpose of this invention is to provide a passive optical kinship feature acquisition and automatic matching method and system for family tracing based on the applicant's prior patent technology. It reuses existing optical film hardware (based on prior patent application 1), eliminating the need for additional dedicated equipment. This achieves active acquisition, encrypted uploading, cloud storage, and AI comparison of passive optical features of the human body, solving the problem of family tracing for abducted individuals in situations where there is no information, no contact, and no cooperation from their relatives. It realizes "active operation, film application participation, and accurate matching," balancing technological feasibility, data security, and social welfare value. Simultaneously, it broadens the scope of technology protection, adapting to various terminal devices with screen display functions, and is compatible with implementations where the optical film is independently attached or integrated into the terminal device. It covers the core comparison algorithm logic, blocking vulnerabilities that competitors might circumvent through algorithm modification, terminal type replacement, or optical film integration.

[0006] (II) Technical Solution To achieve the above objectives, the present invention adopts the following technical solution, which is based on the applicant's prior patent technology, reuses existing photoconductor film hardware (which can be attached to various terminal screens with screen display functions or integrated into the terminal device, based on prior application patent 1), requires no additional dedicated hardware equipment, and is compatible with multiple bloodline feature comparison algorithms. The specific technical solution is as follows: A passive optical kinship feature acquisition and automatic matching method based on a photoconductor film, characterized by the following steps: S1: Establishing optical interaction between the photoconductor film set on the terminal screen and the terminal APP, configuring the user-initiated human optical feature acquisition permission; the photoconductor film can be independently attached to the surface of the terminal screen, or integrated into the terminal device and set as an integral part of the terminal screen; the terminal is an electronic device with screen display function, including mobile phones, tablets, smartwatches, computers, and shopping mall touch screens; S2: Responding to the user's active trigger, acquiring the user's human optical features through the photoconductor film and generating an optical feature code; S3: Encrypting the optical feature code and uploading it to a cloud comparison database in response to the user's active trigger; S4: The cloud comparison database uses a preset... The blood relative feature comparison algorithm compares the uploaded optical feature code with the existing feature codes in the database. When the similarity reaches a preset threshold, it is determined to be a suspected blood relative and an early warning is triggered. The blood relative feature comparison algorithm includes, but is not limited to, convolutional neural network (CNN) algorithm, deep neural network (DNN) algorithm, support vector machine (SVM) algorithm, etc. Any algorithm that can achieve blood relative similarity comparison based on human optical features falls within the protection scope of this method. S5: The early warning information is pushed to the corresponding terminal APP. The user can apply for further verification. After the verification is successful, the two parties will be assisted in establishing contact. S6: The user can manage his own optical feature code through the terminal APP. The cloud comparison database encrypts and stores the feature code actively uploaded by the user and does not store the user's privacy information. Detailed Implementation

[0007] The human optical characteristics described in this invention are defined as skin spectral reflectance characteristics, subcutaneous capillary light absorption patterns, facial skin optical texture, and pupillary response to ambient light. These characteristics can be used for kinship identification and have clear scientific theoretical and research support: publicly available academic research (such as "Research on the Correlation between Iris Genetic Similarity and Kinship" and "Analysis of the Family Genetic Characteristics of Skin Spectral Characteristics") shows that there is a statistically significant correlation between skin spectral reflectance patterns and subcutaneous capillary distribution patterns among direct relatives; pupillary response to ambient light is genetically regulated, with significantly higher similarity among relatives than among non-relatives; facial skin optical texture is influenced by genetic factors, with a high degree of overlap in texture features among direct relatives. The combination of these characteristics can effectively distinguish between blood relatives and non-blood relatives, providing a reliable scientific basis for kinship identification.

[0008] The following details each step and clarifies the compatibility of this solution with different terminal types, light guide film implementation forms, and algorithm types, thus mitigating the risk of competitors bypassing it: S1: Binding and configuring the light guide film device and acquisition permissions. User terminals (including mobile phones, tablets, smartwatches, computers, shopping mall touch screens, and other electronic devices with screen display functions) are equipped with light guide films. The light guide film can be independently attached to the surface of the terminal screen or integrated into the terminal device and set up as an integral part of the terminal screen. Both implementation forms are based on the core technology of the light guide film in the prior application patent 1. Their core optical structure and passive working principle are completely the same, only the installation method is different, and both fall within the protection scope of this invention. The photoconductor film is a passive optical structure. The term "passive" in this invention is specifically defined as follows: the photoconductor film itself is a passive optical structure, containing no chips, circuits, or independent power supplies. The light source required for its operation is provided by the terminal screen or ambient light. Non-sensitive electrical contact (such as electrostatic conduction during bonding, and basic electrical signal conversion of optical signals) is allowed between the photoconductor film and the terminal screen. Such contact is only used to realize optical signal transmission, does not involve the reading of core data of the terminal system, does not depend on the power supply of the terminal system, and does not change the passive core attribute of the photoconductor film. The photoconductive film interacts with the accompanying APP on the terminal (which can reuse the function module of the "Manager APP" in prior patent 4) through optical signals. The terminal APP receives the optical feature signals transmitted by the photoconductive film by emitting light through the screen. No power is required for the photoconductive film, achieving passive interaction. The interaction process is based on the multi-terminal collaborative binding technology of prior patent 4 to complete device binding. The APP disables the "passive optical feature acquisition" permission by default. Users need to actively enable the acquisition permission. After enabling, the acquisition operation can be actively triggered through the APP. The acquisition permission is only used to collect human optical features and does not collect terminal screen content or user privacy information, which meets the core requirement of "permission-free interaction" of prior patent 1 (does not read the terminal system's core data, does not apply for system auxiliary function permissions, only accesses the system APIs required to realize the basic operation of the APP, screen emission, and data encryption transmission, and does not access any system APIs unrelated to acquisition and comparison). At the same time, relying on the cross-terminal adaptation technology of prior patent 2, it ensures that the APP runs normally on different types of terminals and different system terminals, adapting to the screen size and system environment of various terminals such as mobile phones, tablets, smartwatches, computers, and shopping mall touch screens. For small-screen terminals such as smartwatches, the APP interface can be adaptively scaled, and the screen illuminates to guide the user to align their finger or face with the collection area during data collection. For fixed terminals such as shopping mall touch screens, screen prompts can guide users to operate independently, and the data collection process is consistent with that of mobile terminals, ensuring that all types of terminals can stably achieve the feature collection function.

[0009] S2: The passive optical feature active acquisition light guide film for human body is based on the passive optical acquisition technology of the prior patent application 1. It realizes the transmission of light signals through edge light entry and internal total internal reflection, without relying on the power supply of the terminal system. Regardless of whether the light guide film is independently attached or integrated into the terminal, its acquisition principle and core structure are completely the same, and it can realize non-contact acquisition of human body optical features. After the user actively enables data collection permissions, the data collection operation is triggered by clicking the "Collect" button in the terminal APP (or by active triggering methods such as terminal touch operation or voice triggering). At this time, the photoconductor film collects the user's human optical characteristics (skin spectral reflectance characteristics, subcutaneous capillary light absorption patterns, facial skin optical texture, and pupil response to ambient light) through the principles of optical transmission and reflection (which conforms to the working principle of the photoconductor film in prior application patent 1). The data collection process is non-contact and non-invasive, and does not affect the user's normal use of the terminal. The collection frequency is controlled by the user, and each collection requires the user to actively trigger it. The collection time is ≤100ms, and the collected data is an optical feature code (the original image and original optical data are not stored) to avoid privacy leakage. The data collection accuracy of the photoconductor film relies on the micro-nano grating structure (grating period 500nm-2μm) of prior application patent 1 to ensure the accuracy of feature collection. This accuracy standard is applicable to all terminal types and photoconductor film implementation forms.

[0010] S3: Encryption and Active Uploading of Human Optical Feature Data Collected by the Optical Conveyor Film. Using the encryption storage technology (end-to-end dual encryption + AES-128 encryption) of the prior patent 3, the data is first encrypted locally on the terminal to generate a unique "human optical feature code" (irreversible encryption, making it impossible to deduce the original optical data or user identity information from the feature code). The optical feature code temporarily stored locally on the terminal is encrypted using the same SHA-256 hash encryption algorithm as the cloud, with a default storage period of 7 days, which users can set from 1 to 30 days. If the feature code is not uploaded within the storage period, the terminal APP will automatically delete it, and the deletion is irreversible. Users can manually delete locally stored feature codes at any time; deletion operations are logged synchronously to ensure data traceability. The encrypted optical feature code must be actively uploaded to the terminal APP by the user clicking the "Upload" button (or through terminal touch, voice triggering, etc.) before it can be transmitted to the terminal APP. The APP then transmits the data via an encrypted communication link (based on the local area network TCP / IP protocol communication technology of the prior patent 4, using RSA). The encryption algorithm is uploaded to the cloud comparison database. Users can choose to upload immediately or temporarily store the data locally on the terminal and actively trigger the upload later, ensuring that users have independent control over their own feature data. The data transmission process complies with the "unauthorized data reporting" specification of the prior patent application 4, and does not transmit terminal system data or user privacy information. This encrypted transmission logic is applicable to all terminal types and is not affected by terminal size or system differences.

[0011] S4: Cloud-based AI-powered Bloodline Feature Similarity Comparison. The cloud-based comparison library is based on the AI ​​instruction parsing module of prior patent application 3, and incorporates a bloodline optical feature comparison model. This model, based on the AI ​​instruction parsing module of prior patent application 3, can employ various bloodline feature comparison algorithms, such as Convolutional Neural Network (CNN), Deep Neural Network (DNN), and Support Vector Machine (SVM). Any algorithm capable of achieving bloodline similarity comparison based on human optical features and outputting similarity scores falls within the protection scope of this invention, preventing adversaries from circumventing protection by modifying the algorithm type. All of the above algorithms can achieve classification and similarity calculation of bloodline optical features through supervised training. Their core input is the optical feature code generated by this method, and their output is a similarity score. Different algorithms differ in feature extraction layers, network structures, and loss functions, but all can achieve the comparison objective of this invention and are equivalent alternatives to this invention. The model sample types include optical feature samples of close relatives such as parents and children, siblings, etc. (sample size ≥ 100,000 sets, including 60,000 sets of blood relatives and 40,000 sets of non-blood relatives; sample sources cover different ages, genders, and regions, covering all age groups from old to young, and including samples with different physiological states such as normal skin and sensitive skin). The training method is supervised training, which achieves accurate recognition of blood relatives' optical features through sample labeling, feature extraction, and model iterative optimization. The testing method combines blind testing and control testing. In blind testing, the kinship label of the sample is hidden. In control testing, this model is tested simultaneously with a traditional face recognition kinship recognition model. The cloud system regularly performs batch comparisons of all optical feature codes in the database, and can also trigger real-time comparison with all feature codes in the database immediately after a new optical feature code is uploaded. The comparison process only targets optical feature codes and does not involve any user identity information. The comparison result only records the "similarity score" (0-100 points), and the similarity threshold is set to 85 points and above. The basis for setting this threshold is: based on the above 10 In blind tests and control tests with tens of thousands of blood relatives and non-blood relatives, the accuracy of this model can reach over 98.5%, effectively distinguishing between blood relatives and non-blood relatives, balancing accuracy and error tolerance. When the similarity of two optical feature codes reaches a threshold, they are judged as "suspected blood relatives". The error tolerance mechanism of AI comparison refers to the special population adaptation logic of the prior patent application 3 to ensure the accuracy of feature comparison under different ages and physiological states. This error tolerance mechanism is applicable to all comparison algorithms.

[0012] S5: Matching Result Warning and Feedback When the cloud system determines that two optical feature codes are "suspected blood relatives", it immediately triggers a warning mechanism. Through an encrypted communication link (based on the communication technology of prior patent application 4), the warning information is pushed to the terminal APP corresponding to the two feature codes (without pushing any identity information or location information of the other party, only pushing a reminder of "suspected blood relative match detected, please go to the APP to confirm"); regardless of the terminal type, such as mobile phone, tablet, smartwatch, computer or mall touch screen, the warning information can be pushed through the corresponding terminal APP or terminal display interface to ensure that users can receive it in time; after receiving the reminder, users can apply for further verification in the APP (such as uploading a face for auxiliary verification, voluntarily submitting a DNA comparison application, etc.). After the verification is successful, the public welfare family-finding platform (linked with the cloud database) will assist both parties in establishing contact to achieve family reunion; if the user has not enabled the collection permission, has not actively collected and uploaded the feature code, they will not participate in the comparison and will not receive the warning; the push logic of the warning information conforms to the multi-terminal information feedback mechanism of prior patent application 4, ensuring reliable information transmission and adapting to information push scenarios of various terminals.

[0013] S6: Feature Code Management and Privacy Protection. Users can view their own actively collected and uploaded optical feature codes (only encrypted feature codes are displayed, not the original data) through the terminal APP, and can manually delete feature codes (after deletion, the cloud database will also delete the feature code and it will no longer be used for comparison); the cloud comparison database adopts encrypted storage technology (based on the SHA-256 hash encryption algorithm of prior patent application 3 + anti-tampering data self-destruction mechanism), storing only the optical feature codes actively uploaded by users, and not storing any user identity information, terminal information, or location information, ensuring user privacy and security; the feature code storage period can be set by the user (by default, it is stored until the user actively deletes it). If no action is taken after the storage period expires, the user will be automatically reminded to confirm whether to retain it. Temporary data will be executed according to the "clear immediately after execution" mechanism of prior patent application 3; this privacy protection logic is applicable to all terminal types and photoconductor film implementation forms, ensuring user data security in different scenarios.

[0014] Exception handling mechanism Binding failure: When the photoconductor film fails to bind with the terminal APP, the APP will automatically prompt the binding steps (such as re-attaching the photoconductor film, restarting the APP, and checking terminal compatibility) and provide troubleshooting guidance to adapt to binding scenarios of different terminal types; Invalid signature: If the signature becomes invalid due to light interference or terminal screen abnormality during the data collection process, the APP will immediately prompt the user to adjust the lighting environment, check the terminal status, and re-trigger the data collection. Network Interruption: If the network is interrupted during the upload process, the signature is automatically temporarily stored locally on the terminal. The user will be automatically reminded to upload after the network is restored. This feature is compatible with network scenarios such as fixed terminals like computers and shopping mall touch screens, as well as mobile terminals like mobile phones and tablets. Misjudgment during comparison: If a user disagrees with a comparison result below 85, they can apply for a second comparison to collect multiple sets of optical feature codes, thereby improving the comparison accuracy. The second comparison logic applies to all comparison algorithms. Accidental deletion of signature codes: If a user accidentally deletes a signature code, it can be recovered through the APP recycle bin within 72 hours. After 72 hours, it cannot be recovered to ensure data security. Photoconductor film aging / damage: When the photoconductor film ages or wears down due to long-term use, resulting in decreased acquisition accuracy or inability to acquire features, the APP will detect the abnormality through optical signals and immediately prompt the user to check the condition of the photoconductor film (such as whether it is worn or whether the adhesion is tight), and guide the user to replace the photoconductor film (independently attached type) or contact the terminal manufacturer for repair (integrated type). After replacement / repair, the APP will automatically complete rebinding and acquisition calibration to ensure normal acquisition function.

[0015] A passive optical kinship feature acquisition and automatic matching system based on a photoconductor film is characterized by being built upon the applicant's prior patent technology, reusing existing photoconductor film hardware, and requiring no additional dedicated equipment. The system includes a photoconductor film, a terminal APP, an encrypted communication link, a cloud-based comparison database, an AI comparison module, and a privacy protection module. The connection relationships and collaborative logic of each module are as follows: the photoconductor film and the terminal APP establish passive interaction through optical signals; the terminal APP receives the optical feature signals transmitted by the photoconductor film through screen illumination; the encrypted communication link uses RSA encryption based on the TCP / IP protocol; the optical feature data acquired by the photoconductor film is encrypted by the terminal APP and then uploaded to the cloud-based comparison database via the encrypted communication link; the AI ​​comparison module is embedded in the cloud-based comparison database; and the privacy protection module runs through the entire process of terminal, communication, and cloud, achieving full-process data encryption protection. The data flow sequence is clearly defined as: photoconductor film acquires human optical features → transmits to the terminal APP → terminal APP encrypts the feature data to generate an optical feature code → user actively triggers upload → encrypted communication link transmits the feature code to the cloud-based comparison database → AI comparison module… The comparison module calls the comparison algorithm to complete the similarity comparison → if the comparison reaches a threshold, an early warning is triggered → the encrypted communication link pushes the early warning information to the terminal APP → the user applies for verification → after successful verification, assistance is provided to establish contact; the terminal is an electronic device with a screen display function, including mobile phones, tablets, smartwatches, computers, and shopping mall touch screens; the photoconductor film can be independently attached to the surface of the terminal screen, or integrated into the terminal device and set as an integral part of the terminal screen; the AI ​​comparison module can use a variety of blood relative feature comparison algorithms. All algorithms that can achieve blood relative similarity comparison based on human optical features fall within the protection scope of this system. The specific functions of each module are as follows: The photoconductor film (tempered glass photoconductor film) is the core component described in the prior patent application 1. It can be attached to the screen surface of user terminals (including mobile phones, tablets, smartwatches, computers, and shopping mall touch screens) or integrated into the terminal device. Users can install the independently attached photoconductor film themselves, while the integrated photoconductor film can be pre-installed with the terminal device without professional operation. After installation or pre-installation, it automatically establishes optical signal interaction with the terminal APP and completes the initial data acquisition calibration. The photoconductor film is a passive optical structure that does not contain any chips, circuits, or independent power supplies. The light source required for its operation is provided by the terminal screen or ambient light. It achieves optical signal transmission through edge light entry and internal total internal reflection, which conforms to the structural design of the prior patent application 1 (including micro-nano grating array, grating period 500nm-2μm). Through the principle of optical transmission and reflection, it collects the user's human optical characteristics after the user actively triggers the acquisition operation. The acquisition process is non-contact and does not affect the user's normal use of the terminal. It is used to provide core data support for blood kinship feature comparison. The photoconductor film does not need to interact with the terminal system for sensitive data; it only communicates with the terminal APP through optical signals. The transmission of non-sensitive optical feature data, without reading the core data of the terminal system or requesting sensitive system permissions, and with the core functions and optical principles of independently attached and integrated photoconductive films being completely identical, all fall within the scope of this protection.

[0016] Terminal App: The app can reuse the "Manager App" functional module from prior patent application 4, establishing passive interaction with the photoconductor film via optical signals. It is compatible with various terminals such as mobile phones, tablets, smartwatches, computers, and shopping mall touchscreens. The interface and functions can be adaptively adjusted according to the terminal screen size and system type. It is used for photoconductor film device binding, data collection permission management, optical feature code encryption processing, active feature code uploading and storage, comparison and warning reception, and feature code management. The app has a built-in privacy protection module to ensure that the collected optical feature data is not leaked or misused. It supports users manually enabling / disabling data collection permissions, manually triggering data collection and uploading operations, and manually deleting feature codes, ensuring users' independent control over their own data. This complies with the "no-permission operation" requirement of prior patent application 1 (no application for sensitive system permissions, only access to the system APIs required for basic app operation, screen illumination, and encrypted data transmission). Simultaneously, relying on the adaptation parameter library of prior patent application 2, it achieves cross-terminal and cross-system adaptation, and supports users to independently set the optical feature code temporary storage period (1-30 days).

[0017] Encrypted communication link: Based on the local area network TCP / IP protocol communication technology of prior application patent 4, it adopts the RSA encryption algorithm, adapts to the network transmission scenarios of various terminals, and is used for information transmission (optical feature code, early warning information, etc.) between terminal APP and cloud comparison database. It ensures the security of data transmission process, prevents data from being tampered with or leaked, and complies with the cross-terminal data transmission specification of prior application patent 4.

[0018] Cloud-based comparison database: Built upon the encrypted storage technology and AI instruction parsing module of prior patent application 3, it is used to encrypt and store human optical feature codes actively uploaded by users. It has a built-in blood relative optical feature comparison model, is compatible with multiple comparison algorithms, and can realize batch and real-time comparison of optical feature codes, triggering a suspected blood relative warning. It does not store any user privacy information, but only retains the encrypted feature codes and comparison records. The storage mechanism complies with the privacy protection specifications of prior patent application 3, and has the functions of automatic temporary data deletion and anti-tampering data self-destruction. It can adapt to the feature code storage needs of a large number of users with different terminal types.

[0019] AI Comparison Module: Built upon the AI ​​instruction parsing module of prior patent application 3, and embedded with a cloud-based comparison library, this module employs various kinship feature comparison algorithms, including Convolutional Neural Network (CNN), Deep Neural Network (DNN), and Support Vector Machine (SVM). Trained on massive amounts of kinship samples, it can accurately calculate the similarity between two optical feature codes. The similarity threshold can be adjusted between 80 and 90 points according to user needs to determine whether the individuals are suspected kin and trigger an early warning mechanism, ensuring comparison accuracy. The comparison logic references the special population fault-tolerance mechanism of prior patent application 3, improving adaptability in different scenarios. All algorithms capable of kinship optical feature comparison fall within the protection scope of this module, preventing adversaries from circumventing protection by modifying algorithms.

[0020] Privacy Protection Module: This module runs throughout the entire system and includes features such as local data encryption on the terminal (based on the end-to-end dual encryption technology of prior patent 3), encrypted cloud data storage (based on the SHA-256 hash encryption algorithm of prior patent 3), irreversible encryption of feature codes, and user-managed feature codes (enabling / disabling collection, actively triggering collection and uploading, and deleting feature codes). A new data deletion log function has been added to trace all feature code deletion operations, ensuring that users' human optical feature data is not leaked or misused. This complies with relevant privacy protection laws and regulations and the privacy and security design requirements of prior patent 3, and is adaptable to privacy protection scenarios of various terminals. Beneficial effects

[0021] Compared with the prior art, this invention is based on the applicant's previous patent technology, reuses existing photoconductor film hardware, and does not require additional dedicated equipment, thus having the following advantages: Easy to popularize and resistant to circumvention: It reuses existing light guide film hardware (based on prior patent application 1), supports two implementation methods: independent attachment and terminal integration. It is compatible with various terminals with screen display functions, such as mobile phones, tablets, smartwatches, computers, and shopping mall touch screens, preventing competitors from circumventing protection by changing the terminal type or integrating the light guide film into the terminal. Users only need to set the light guide film on the terminal and participate in the search for relatives through active operation. There is no need to purchase additional special equipment, which can achieve large-scale popularization and cover a huge number of people. It is especially suitable for abducted persons, relatives in remote areas, and other groups searching for relatives. The low-cost mass production characteristics of the light guide film (based on the pure optical structure design of prior patent application 1) further lowers the threshold for promotion.

[0022] Autonomous and controllable, contactless and non-invasive: Users can actively enable collection permissions and actively trigger collection and upload operations, fully controlling their own feature data; the photoconductive film collection process is contactless and non-invasive, does not affect the user's normal use of the terminal, and meets the core requirement of "authority-free interaction" of the prior patent application 1. It does not rely on system permissions, avoids being blocked by the system, and solves the defects of existing family-finding technologies (DNA, facial recognition) that require active collection but are complex to operate, have poor controllability, and rely on permissions; at the same time, it is compatible with the collection scenarios of various terminals, improving the ease of use.

[0023] Privacy and security are guaranteed: Relying on the encrypted storage and privacy protection technology of the prior patent application 3, the collected data is only the human optical feature code, which is irreversibly encrypted and cannot be used to deduce the original data and user identity information; the cloud library only stores the feature codes actively uploaded by users and does not store any privacy information. It has the functions of automatic deletion of temporary data and self-destruction of data to prevent disassembly; users can manage the feature codes themselves, and the data temporarily stored locally on the terminal also has a complete encryption and deletion mechanism, which fully complies with relevant privacy protection laws and regulations. The privacy and security far exceed existing family-finding technologies and is suitable for the privacy protection needs of various terminals.

[0024] High accuracy, resistance to algorithm bypass, and high efficiency in family tracing: The AI ​​comparison module is based on the AI ​​instruction parsing technology of prior patent application 3, and is compatible with various blood kinship feature comparison algorithms such as convolutional neural networks (CNN) and deep neural networks (DNN), preventing adversaries from bypassing the protection by modifying the algorithm type; through training with massive blood kinship samples, it can accurately identify the optical feature similarity between blood relatives, unaffected by factors such as age, makeup, or occlusion, and the comparison accuracy is higher than that of traditional face recognition. According to tests, the blood kinship recognition accuracy of the AI ​​comparison module of this invention can reach more than 98.5%, while the blood kinship recognition accuracy of traditional face recognition is about 85%, significantly improving the reliability of recognition; after the feature code is uploaded, the cloud automatically compares it without human intervention, greatly improving the efficiency of family tracing and realizing "proactive participation and rapid early warning"; multi-terminal communication and data transmission rely on the technology of prior patent application 4 to ensure reliable and lossless data transmission between different terminal types.

[0025] Highly reusable and feasible: Based entirely on the core technologies of the applicant's four existing unpublished prior patent applications, no new hardware or R&D of new core technologies is required, enabling rapid implementation and promotion. Key technologies such as the structural design of the photoconductor film, the permissionless interaction mechanism, encrypted storage, and multi-terminal collaboration have all been verified in the prior patent applications, demonstrating high technological maturity. This not only continues the existing functions of the photoconductor film but also adds public welfare value in finding missing persons, achieving a dual enhancement of commercial and social value. At the same time, it is compatible with various implementation forms, comparison algorithms, and terminal types of the photoconductor film, further improving the feasibility and circumvention resistance of the technology.

[0026] Wide applicability and strong resistance to bypassing: Relying on the cross-platform adaptability of the prior patent application 1, the light guide film structure design of patent 1 is compatible with terminal screens of different sizes and systems. Its permissionless interaction mechanism can ensure that the terminal APP runs normally on mainstream systems such as HarmonyOS, Android, and iOS, without the need for separate development for different terminals, achieving cross-terminal universality; It supports two forms of light guide film: independent attachment and integration, and is compatible with multiple comparison algorithms, preventing adversaries from bypassing protection by means of terminal replacement, light guide film integration, algorithm modification, etc.; No matter where the abducted person and their relatives are located, no matter what terminal device they use, as long as both parties set up the light guide film and actively complete the collection and upload operation, feature collection and automatic comparison can be completed, breaking geographical restrictions, information restrictions, and device restrictions, and covering various family reunification scenarios.

Claims

1. A passive optical kinship feature acquisition and automatic matching method based on photoconductive film, characterized in that, Includes the following steps: S1: Through the light guide film set on the terminal screen and the terminal APP Establish optical interaction and configure user-initiated human optical feature acquisition permissions; the light guide film can be independently attached to the terminal screen surface or integrated into the terminal device and set as an integral part of the terminal screen; the terminal is an electronic device with screen display function, including mobile phones, tablets, smartwatches, computers, and shopping mall touch screens; S2: In response to the user's active trigger, the light guide film collects the user's human optical features and generates an optical feature code; S3: The optical feature code is encrypted and uploaded to the cloud comparison database in response to the user's active trigger; S4: The cloud comparison database uses a preset blood relative feature comparison algorithm to compare the uploaded optical feature code with the existing feature codes in the database. When the similarity reaches a preset threshold, it is determined to be a suspected blood relative and an early warning is triggered; the blood relative feature comparison algorithm includes, but is not limited to, convolutional neural network (CNN) algorithm, deep neural network (DNN) algorithm, and support vector machine (SVM) algorithm; S5: The early warning information is pushed to the corresponding terminal APP, and the user can apply for further verification. After verification, assistance is provided to establish contact between the two parties; S6: The user can communicate through the terminal APP. It manages its own optical signature, and the cloud-based comparison library encrypts and stores the signatures actively uploaded by users, without storing users' private information.

2. The method according to claim 1, characterized in that, In step S1, the photoconductor film is a passive optical structure, containing no chips, circuits, or independent power supply. Its operating light source is provided by the terminal screen or ambient light. Non-sensitive electrical contact (such as electrostatic conduction during bonding, or basic electrical signal conversion of optical signals) is permitted between the photoconductor film and the terminal screen. This contact is solely for optical signal transmission and does not involve reading core terminal system data or rely on the terminal system's power supply. The photoconductor film and the terminal APP establish passive interaction through optical signals. The terminal APP receives the optical feature signals transmitted by the photoconductor film by emitting light through its screen, without requiring power from the photoconductor film. The interaction binding process is based on the multi-terminal collaborative binding technology of prior application patent 4 (application number 2026103373270, invention title: "A Distributed Butler System and Method Based on Multi-Terminal Collaboration"). Data collection permissions are disabled by default and must be actively enabled by the user. Once enabled, it is only used to collect human optical features and does not collect terminal screen content or user privacy information, complying with application number 2026102814971 (invention title: "A Permission-Free AI Interaction Implementation Method Based on Photoconductor Tempered Glass Film"). The prior application patent 1 requires "unauthorized interaction" (not reading the core data of the terminal system, not applying for system auxiliary function permissions, only accessing the system APIs required to realize the basic operation of the APP, screen illumination, and data encryption transmission, and not accessing any system APIs unrelated to collection and comparison); at the same time, relying on the cross-terminal adaptation technology of the prior application patent 2 with application number 2026103109757 and invention title "An AI Instruction Intelligent Execution System Based on External Interaction Layer", it ensures that the APP runs normally on different types of terminals and different system terminals.

3. The method according to claim 1, characterized in that, In step S2, the human optical characteristics are defined as skin spectral reflectance characteristics, subcutaneous capillary light absorption patterns, facial skin optical texture, and pupil response to ambient light characteristics. The photoconductor film is based on the passive optical acquisition technology of the prior patent application 1, application number 2026102814971, entitled "A Method for Implementing Unauthorized AI Interaction Based on Photoconductor Tempered Glass Film". It achieves light signal transmission through edge light entry and internal total internal reflection. The acquisition operation requires active triggering by the user. The grating period of the photoconductor film is 500nm-2μm, and the single acquisition time is ≤100ms. The core optical structure and acquisition principle of the independently attached and integrated photoconductor films are completely consistent.

4. The method according to claim 1, characterized in that, In step S3, the optical feature code is irreversibly encrypted. It utilizes the full-link dual encryption + AES-128 encryption technology of the prior application patent 3 (application number 2026103269949, invention title: "Control Method and System for Unauthorized Eye-Tracking Touch Terminal Based on Optical Guide Interaction Layer") to achieve local encryption on the terminal. Cloud upload and transmission are achieved using the RSA encryption algorithm, ensuring data security and preventing the inference of original optical data and user identity information from the feature code. The optical feature code temporarily stored locally on the terminal uses the SHA-256 hash encryption algorithm. The storage period can be set by the user from 1 to 30 days, with a default of 7 days. It is automatically deleted after the expiration date and cannot be recovered. This encrypted transmission logic is applicable to all terminal types.

5. The method according to claim 1, characterized in that, In step S4, the AI ​​comparison module is built based on the AI ​​instruction parsing module of the prior application patent 3, application number 2026103269949, entitled "Control Method and System for Unauthorized Eye-tracking Touch Terminal Based on Optical Guide Interaction Layer". It can use various blood relative feature comparison algorithms such as Convolutional Neural Network (CNN), Deep Neural Network (DNN), and Support Vector Machine (SVM). It is trained in a supervised manner through ≥100,000 sets of optical feature samples of close relatives (60,000 sets of blood relatives and 40,000 sets of non-blood relatives, with sample sources covering different ages, genders, and regions). The preset similarity threshold is 85 points and above. This threshold is set based on the fact that the accuracy rate can reach more than 98.5% after blind testing and control testing of the above 100,000 sets of blood relatives and non-blood relatives. The comparison process only targets the optical feature code and does not involve any user privacy information. The comparison logic includes a special population fault tolerance mechanism and is applicable to all comparison algorithms.

6. The method according to claim 1, characterized in that, In step S5, the warning message only pushes the reminder "A suspected blood relative match has been detected. Please go to the APP to confirm." It does not push the other party's identity information or location information. Users can apply for face-assisted verification or DNA comparison verification through the APP. The warning message push is based on the multi-terminal information feedback mechanism of the prior application patent 4, application number 2026103373270, invention title "A Distributed Butler System and Method Based on Multi-Terminal Collaboration," which is adapted to information push scenarios of various terminals.

7. The method according to claim 1, characterized in that, In step S6, the cloud-based comparison library uses the SHA-256 hash encryption algorithm to store optical feature codes (based on prior application patent 3, application number 2026103269949, invention title "Control Method and System for Unauthorized Eye-tracking Touch Terminal Based on Optical Guide Interaction Layer"). Users can manually delete their own feature codes, which are then simultaneously deleted from the cloud library and no longer used for comparison. Feature codes are stored by default until actively deleted by the user, and temporary data is subject to an "immediately clear after execution" mechanism. If a user accidentally deletes a feature code, it can be recovered through the APP recycle bin within 72 hours; after 72 hours, it cannot be recovered. This privacy protection logic applies to all terminal types and optical guide film implementations.

8. A passive optical kinship feature acquisition and automatic matching kinship tracing system based on a photoconductive film, characterized in that, Based on the applicant's prior patented technology, and reusing existing photoconductor hardware, the system requires no additional dedicated equipment. It includes a photoconductor film, a terminal app, an encrypted communication link, a cloud-based comparison library, an AI comparison module, and a privacy protection module. The connection relationships and collaborative logic of each module are as follows: The photoconductor film and the terminal app establish passive interaction through optical signals. The terminal app receives the optical feature signals transmitted by the photoconductor film through screen illumination. The encrypted communication link uses RSA encryption based on the TCP / IP protocol. The optical feature data collected by the photoconductor film is encrypted by the terminal app and then uploaded to the cloud-based comparison library via the encrypted communication link. The AI ​​comparison module is embedded in the cloud-based comparison library. The privacy protection module runs through the entire process of terminal, communication, and cloud, achieving full-process data encryption protection. The data flow sequence is: photoconductor film collects human optical features → transmits to the terminal app → terminal app encrypts the feature data to generate an optical feature code → user actively triggers upload → encrypted communication link transmits the feature code to the cloud-based comparison library → AI comparison module. The comparison module calls the comparison algorithm to complete the similarity comparison → if the comparison reaches the threshold, an early warning is triggered → the encrypted communication link pushes the early warning information to the terminal APP → the user applies for verification → after the verification is passed, assistance is provided to establish contact; the photoconductor film can be independently attached to the surface of the terminal screen, or integrated into the terminal device and set as an integral part of the terminal screen. It is attached to the user terminal screen or integrated into the terminal device, and establishes passive interaction with the terminal APP through optical signals. It is a pure optical structure (no chip, no circuit, no power supply, the light source required for operation is provided by the terminal screen or ambient light). Based on the prior patent application 1 with application number 2026102814971 and invention title "A Method for Implementing Unauthorized AI Interaction Based on Photoconductor Tempered Glass Film", it is used to collect the user's human optical features after the user actively triggers the collection operation; the terminal is an electronic device with screen display function, including mobile phones, tablets, smartwatches, computers, and shopping mall touch screens; the terminal APP This system is used for photoconductor film bonding, acquisition permission management, optical feature code encryption and active uploading, early warning reception, and feature code management. It supports users to actively enable acquisition permissions and actively trigger acquisition and uploading operations. It can reuse the sub-housekeeper APP function module of the prior application patent 4 (application number 2026103373270, invention title: "A Distributed Housekeeper System and Method Based on Multi-Terminal Collaboration"), and is compatible with various terminal types. The encrypted communication link is used for encrypted data transmission between the terminal APP and the cloud library. It is built based on the local area network TCP / IP protocol communication technology of the prior application patent 4 (application number 2026103373270, invention title: "A Distributed Housekeeper System and Method Based on Multi-Terminal Collaboration"), and is compatible with network transmission scenarios of various terminals.The cloud-based comparison library is used to encrypt and store optical feature codes actively uploaded by users. It is built upon the encryption storage technology of prior application patent 3 (application number 2026103269949, invention title: "Unauthorized Eye-Tracking Touch Terminal Control Method and System Based on Optical Guide Interaction Layer"), and can adapt to the feature code storage needs of a massive number of users with different terminal types. The AI ​​comparison module is used for optical feature code similarity comparison, determining suspected blood relatives and triggering warnings. It is also based on the AI ​​technology of prior application patent 3 (application number 2026103269949, invention title: "Unauthorized Eye-Tracking Touch Terminal Control Method and System Based on Optical Guide Interaction Layer"). The instruction parsing module can employ various kinship feature comparison algorithms, such as Convolutional Neural Network (CNN), Deep Neural Network (DNN), and Support Vector Machine (SVM). The privacy protection module is used to implement data encryption between the terminal and the cloud, user privacy protection, and autonomous management of feature codes. It is built upon the privacy and security technology of prior application patent 3 (application number 2026103269949, invention title: "Control Method and System for Unauthorized Eye-Tracking Touch Terminal Based on Optical Guide Interaction Layer"), supporting user-controlled data collection and uploading operations, and adapting to various terminal privacy protection scenarios.

9. The system according to claim 8, characterized in that, The light guide film does not require sensitive data interaction with the terminal system. It only transmits non-sensitive optical feature data with the terminal APP through optical signals, without reading the core data of the terminal system or requesting sensitive system permissions. The light guide film is a transparent optical film that collects human optical features through edge light entry and internal total internal reflection (based on the prior application patent 1 with application number 2026102814971 and invention title "A Permissionless AI Interaction Implementation Method Based on Light Guide Tempered Glass Film"). The grating period is 500nm-2μm. It does not require data interaction with the terminal system. The independently attached light guide film can be installed and replaced by the user. The integrated light guide film can be pre-installed with the terminal at the factory. After installation or pre-installation, it will automatically bind with the terminal APP and complete the initial acquisition calibration. The acquisition operation needs to be actively triggered by the user.

10. The system according to claim 8, characterized in that, The terminal APP complies with the "permission-free operation" requirement of prior application patent 1 (application number 2026102814971, invention title: "A Permission-Free AI Interaction Implementation Method Based on Light-Guided Tempered Glass Film"), and does not request sensitive system permissions, only accessing the system APIs required for basic APP operation, screen illumination, and encrypted data transmission. The AI ​​comparison module and privacy protection module are both based on the technology of prior application patent 3 (application number 2026103269949, invention title: "Permission-Free Eye-Tracking Touch Terminal Control Method and System Based on Light-Guided Interaction Layer"). The encrypted communication link is based on the technology of prior application patent 4 (application number 2026103373270, invention title: "A Distributed Butler System and Method Based on Multi-Terminal Collaboration"), supporting user-initiated collection and upload operations. The terminal APP allows users to independently set the optical feature code storage period (1-30 days), and the similarity threshold of the AI ​​comparison module can be set between 80-90% according to user needs. The privacy protection module adds a data deletion log function, which can trace all signature deletion operations; the terminal APP can adaptively adjust the interface and functions according to the terminal screen size and system type to adapt to various terminals.