Processing system based on face feature recognition

Abstract

The invention discloses a processing system based on face feature recognition. The processing system comprises an image acquisition module, a feature extraction module, a recognition comparison module, a data storage and management module and a result output and interaction module. A multi-source fusion technology is applied to image acquisition, a quantum neural calculation model is adopted for feature extraction, a chaotic encryption algorithm is used for recognition and comparison, data storage is combined with a block chain and an intelligent compression technology, and result output supports multiple innovative modes. All the modules operate cooperatively, efficient and accurate face feature recognition is achieved, different scene requirements are met, and corresponding innovation methods and related technologies are applied to all the steps to guarantee efficient operation of the system. The human face feature recognition system has multiple advantages. According to the method, high-quality images can be acquired in a complex environment, features can be accurately extracted, rapid and safe comparison can be realized, data can be reliably stored, an innovative output mode is provided, different scene requirements are met, and identification accuracy, safety and user experience are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of face recognition systems, and more specifically, to a processing system based on face feature recognition. Background Art

[0002] With the rapid development of information technology and the increasing demand for social security, facial feature recognition technology plays an increasingly important role in many fields. However, the existing facial feature recognition system still faces many challenges and difficulties.

[0003] In terms of image acquisition, traditional systems are often greatly restricted by environmental factors. For example, in low-light conditions, especially at night or in dim indoor environments, the quality of the collected facial images is seriously degraded, with problems such as blur and high noise, which greatly increases the difficulty of subsequent feature extraction and recognition. Moreover, for complex background interference, such as crowded crowds and scenes with drastic changes in background light, it is difficult to accurately focus and separate facial images, affecting the accuracy and efficiency of acquisition.

[0004] In the feature extraction stage, existing technical methods are insufficient in the comprehensiveness and accuracy of extracting facial features. It is difficult to effectively extract and analyze subtle features of the face, such as slight changes in skin texture and subtle differences in facial expressions. At the same time, the ability to distinguish similar features between different individuals is limited, which can easily lead to misidentification or low recognition accuracy. In addition, the computational complexity of traditional feature extraction algorithms is high, resulting in slow processing speeds, making it difficult to meet application scenarios with high real-time requirements.

[0005] In the recognition and matching process, the existing system shows obvious limitations when faced with large-scale data and complex and changing facial features. The efficiency and accuracy of the matching algorithm need to be improved, especially when dealing with databases of massive facial data. The search and matching time is too long and cannot meet the needs of rapid response. Moreover, the system has poor adaptability to dynamic changes in the face, such as aging, makeup changes, and hairstyle changes, which can easily lead to recognition failure or misrecognition.

[0006] In terms of data storage and management, the security and reliability of traditional systems have hidden dangers. Data storage is vulnerable to hacker attacks and data tampering, resulting in the leakage of facial feature data, posing serious risks to user privacy and security. At the same time, for the storage and management of large-scale data, traditional storage methods are costly and have poor scalability, making it difficult to meet the growing demand for data storage.

[0007] In terms of result output and interaction, existing systems are often single-form and lack flexibility and innovation. They can only provide simple text or image display results, and cannot meet the diverse needs of users and the requirements of complex application scenarios. For example, in some scenarios that require real-time interaction and immersive experience, such as virtual reality and augmented reality applications, existing systems cannot provide effective support. Summary of the invention

[0008] The present invention proposes a processing system based on facial feature recognition, which is used to solve the problems mentioned in the above-mentioned prior art.

[0009] In order to achieve the above objectives, the present invention adopts the following technical solutions: a processing system based on facial feature recognition, including an image acquisition module, a feature extraction module, a recognition and comparison module, a data storage and management module, and a result output and interaction module.

[0010] The image acquisition module uses multi-source fusion imaging technology, combined with high-resolution optical sensors and depth perception sensors, to accurately capture the three-dimensional shape and rich texture information of the face in various complex environments. For example, infrared auxiliary imaging is used in dark environments to ensure clear acquisition of facial images. The image acquisition quality evaluation index formula is: Where Q image Indicates the image acquisition quality, R resolution Indicates the image resolution, D depth_accuracy Indicates the accuracy of depth data, S stability Indicates image stability, F adaptability Indicates environmental adaptability, N noise Represents image noise.

[0011] The feature extraction module uses a quantum neural computing model, which is based on the superposition and entanglement characteristics of quantum bits and can efficiently extract the fusion information of microscopic and macroscopic features of the face, including but not limited to the fine structural features of the facial features, the dynamic change characteristics of facial muscles, and the unique bioelectric signal characteristics. The feature extraction accuracy evaluation index formula is: Among them A extract represents the feature extraction accuracy, N rorrect_features Indicates the number of correctly extracted features, N total_features Indicates the total number of features to be extracted, E quantum_enhancement represents the quantum computing enhancement factor).

[0012] The recognition and comparison module uses a high-speed matching algorithm based on chaotic encryption to quickly and securely compare the extracted facial features with the features in the database. The randomness and complexity of the chaotic system are used to encrypt data transmission and generate dynamic keys during the matching process to ensure data security and recognition accuracy. The recognition and comparison speed evaluation index formula is: Where V recognition Indicates the recognition and comparison speed, N faces_processed Indicates the number of faces processed per unit time, T processing represents the processing time, C chaos_factor Represents the chaotic encryption acceleration factor.

[0013] The data storage and management module combines blockchain distributed storage architecture with intelligent data compression technology. Blockchain ensures that data cannot be tampered with and is traceable, while intelligent data compression technology reduces storage costs and improves storage efficiency while ensuring data integrity. The data storage capacity evaluation indicator formula is: Among them C storage Indicates the data storage capacity, N compressed_data_stored Indicates the amount of compressed data actually stored, S storage_space Indicates the total space of the storage device, D blockchain_efficiency Represents the blockchain storage efficiency factor.

[0014] The result output and interaction module supports a variety of innovative output methods, including but not limited to virtual reality (VR) immersive feedback, augmented reality (AR) visual guidance, and brain-computer interface thought interaction response. For example, in the field of security, AR technology can be used to overlay recognition results in the form of virtual tags on real scenes to show security personnel, or brain-computer interface devices can be used to allow users to interact with the system through thought instructions.

[0015] Preferably, the multi-source fusion imaging technology in the image acquisition module supports automatic environmental perception and adaptive parameter adjustment, and can automatically optimize imaging parameters according to different environmental factors such as lighting, temperature, humidity, etc.

[0016] Preferably, the quantum neural computing model in the feature extraction module uses a quantum annealing algorithm to perform optimization training of the model to improve the performance and efficiency of feature extraction.

[0017] Preferably, the high-speed matching algorithm based on chaotic encryption in the recognition and comparison module supports dynamic key update and multimodal feature fusion comparison to improve the accuracy and security of recognition.

[0018] Preferably, the blockchain distributed storage architecture in the data storage and management module adopts smart contracts to realize automated management of data access rights and secure control of data sharing.

[0019] Preferably, the multiple innovative output modes in the result output and interaction module support user customization and scene adaptive switching to meet the needs of different users and application scenarios.

[0020] A processing method based on facial feature recognition comprises the following steps:

[0021] Image acquisition: The image acquisition module uses multi-source fusion imaging technology and automatic environmental perception technology to collect image information containing faces, and performs preliminary image preprocessing, such as noise reduction and contrast enhancement.

[0022] Feature extraction: The preprocessed image is transferred to the feature extraction module, and the quantum neural computing model and quantum annealing algorithm are used to efficiently extract facial features.

[0023] Identification and matching: The extracted facial features are safely and quickly matched with the features in the database using a high-speed matching algorithm based on chaotic encryption.

[0024] Data storage: The recognition results and facial feature data are securely stored using blockchain distributed storage architecture and intelligent data compression technology.

[0025] Result output: The result output and interaction module presents the recognition results to users or related systems in a variety of innovative output methods and supports user interaction operations.

[0026] Preferably, in the image acquisition step, dynamic tracking technology is used to ensure that the face is always in the best imaging position.

[0027] Preferably, in the feature extraction step, feature enhancement technology is used to highlight and optimize key features.

[0028] Preferably, in the result output step, emotion perception technology is used to adjust the output method and content according to the user's emotional state and attention focus.

[0029] Compared with the prior art, the present invention has the following beneficial effects:

[0030] 1.Excellent image acquisition capability:

[0031] By adopting multi-source fusion imaging technology and automatic environmental perception function, it is possible to collect high-quality facial images in various complex environments. Whether in dark light, strong light, complex background and other harsh conditions, it can ensure the acquisition of clear and accurate facial images, providing a solid foundation for subsequent processing. For example, in nighttime security monitoring, it can clearly capture the facial features of people, greatly improving the effectiveness of monitoring.

[0032] 2. Efficient and accurate feature extraction:

[0033] The quantum neural computing model can comprehensively and accurately extract various features of the face, including the fusion information of microscopic and macroscopic features. It can not only accurately identify the fine structure of the facial features, but also capture the dynamic changes of facial muscles and unique bioelectric signal characteristics. This greatly improves the recognition and differentiation of features, reduces the probability of misidentification, and significantly improves the accuracy of recognition. For example, in identity authentication scenarios, it can accurately distinguish people with similar faces to ensure the security and reliability of authentication.

[0034] 3. Fast and safe identification and comparison:

[0035] The use of a high-speed matching algorithm based on chaotic encryption not only ensures the speed of recognition and comparison, but also improves the security of data. When faced with large-scale face data, it can quickly search and compare, greatly shortening the response time and meeting application scenarios with high real-time requirements. At the same time, chaotic encryption technology ensures the security of data during transmission and comparison, prevents data from being stolen or tampered with, and protects user privacy and data security. For example, in face recognition payment in the financial field, payment verification can be completed quickly while ensuring the security of user funds.

[0036] 4. Reliable data storage and management:

[0037] Combining blockchain distributed storage architecture and intelligent data compression technology, secure storage and efficient management of data are achieved. Blockchain technology ensures the immutability and traceability of data, and improves the security and credibility of data. Intelligent data compression technology reduces storage costs and improves storage efficiency while ensuring data integrity, and also facilitates data backup and recovery. For example, in an enterprise's personnel management system, it can safely store a large number of employees' facial feature data, and conveniently manage and query data.

[0038] 5. Innovative result output and interaction methods:

[0039] It supports a variety of innovative output methods, such as virtual reality immersive feedback, augmented reality visual guidance, and brain-computer interface thought interaction response. This provides users with a richer, more intuitive and convenient interactive experience, meeting the needs of different users and application scenarios. For example, in the field of education and training, virtual reality technology can be used to provide students with an immersive learning experience, and face recognition technology can be used to achieve personalized teaching interaction; in the medical field, augmented reality technology can be used to provide doctors with visual information of patients' facial features to assist in medical diagnosis and treatment. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1This is a schematic block diagram of a processing system based on facial feature recognition proposed by the present invention. DETAILED DESCRIPTION

[0041] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0042] In the description of the present invention, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", "clockwise", "counterclockwise" and the like indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the referred device or element must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as limiting the present invention.

[0043] In addition, the terms "first" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "multiple" is two or more, unless otherwise clearly and specifically defined. In addition, the terms "installed", "connected" and "connected" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, and it can be the internal connection of two elements. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances. The present invention will be further described in detail below in conjunction with the accompanying drawings.

[0044] Reference Figure 1 : A processing system based on facial feature recognition, including an image acquisition module, a feature extraction module, a recognition and comparison module, a data storage and management module, and a result output and interaction module.

[0045] The image acquisition module uses multi-source fusion imaging technology, combined with high-resolution optical sensors and depth perception sensors, to accurately capture the three-dimensional shape and rich texture information of the face in various complex environments. For example, infrared auxiliary imaging is used in dark environments to ensure clear acquisition of facial images. The image acquisition quality evaluation index formula is: Where Q image Indicates the image acquisition quality, R resolution Indicates the image resolution, D depth_accuracy Indicates the accuracy of depth data, S stability Indicates image stability, F adaptability Indicates environmental adaptability, N noise Represents image noise.

[0046] The feature extraction module uses a quantum neural computing model, which is based on the superposition and entanglement characteristics of quantum bits and can efficiently extract the fusion information of microscopic and macroscopic features of the face, including but not limited to the fine structural features of the facial features, the dynamic change characteristics of facial muscles, and the unique bioelectric signal characteristics. The feature extraction accuracy evaluation index formula is: Among them A extract represents the feature extraction accuracy, N corrext_features Indicates the number of correctly extracted features, N total_features Indicates the total number of features to be extracted, E quantum_enhancement represents the quantum computing enhancement factor).

[0047] The recognition and comparison module uses a high-speed matching algorithm based on chaotic encryption to quickly and securely compare the extracted facial features with the features in the database. The randomness and complexity of the chaotic system are used to encrypt data transmission and generate dynamic keys during the matching process to ensure data security and recognition accuracy. The recognition and comparison speed evaluation index formula is: Where V recognition Indicates the recognition and comparison speed, N faces_processed Indicates the number of faces processed per unit time, T processing represents the processing time, C chaos_factor Represents the chaotic encryption acceleration factor.

[0048] The data storage and management module combines blockchain distributed storage architecture with intelligent data compression technology. Blockchain ensures that data cannot be tampered with and is traceable, while intelligent data compression technology reduces storage costs and improves storage efficiency while ensuring data integrity. The data storage capacity evaluation indicator formula is: Among them C storage Indicates the data storage capacity, N compressed_data_stored Indicates the amount of compressed data actually stored, S storage_space Indicates the total space of the storage device, D blockchain_efficiency Represents the blockchain storage efficiency factor.

[0049] The result output and interaction module supports a variety of innovative output methods, including but not limited to virtual reality (VR) immersive feedback, augmented reality (AR) visual guidance, and brain-computer interface thought interaction response. For example, in the field of security, AR technology can be used to overlay recognition results in the form of virtual tags on real scenes to show security personnel, or brain-computer interface devices can be used to allow users to interact with the system through thought instructions.

[0050] In the present invention, the multi-source fusion imaging technology in the image acquisition module supports automatic environmental perception and adaptive parameter adjustment, and can automatically optimize imaging parameters according to different environmental factors such as light, temperature, humidity, etc.

[0051] In the present invention, the quantum neural computing model in the feature extraction module adopts a quantum annealing algorithm to perform optimization training of the model to improve the performance and efficiency of feature extraction.

[0052] In the present invention, the high-speed matching algorithm based on chaotic encryption in the recognition and comparison module supports dynamic key update and multi-modal feature fusion comparison to improve the accuracy and security of recognition.

[0053] In the present invention, the blockchain distributed storage architecture in the data storage and management module gives full play to its unique advantages and innovatively adopts smart contracts to realize the automated management of data access rights and the security control of data sharing.

[0054] In terms of automated management of data access rights, smart contracts will first establish a comprehensive permission system based on preset rules and policies. When new data needs to be stored or existing data needs to be accessed, smart contracts will automatically determine the corresponding access rights based on factors such as the type and source of the data and the user roles involved. For example, for facial feature data involving personal privacy, smart contracts will decide whether to allow access based on the user's identity information and authorization status. If it is the owner of the data, the smart contract will automatically grant full access rights after verifying his or her identity, including operations such as viewing, modifying, and deleting data. For other unauthorized users, smart contracts will strictly limit their access rights or even completely prohibit access.

[0055] Smart contracts can also achieve dynamic permission adjustments. As time goes by and business scenarios change, data access requirements and security levels may change. For example, in a company's human resources management system, when an employee leaves, the smart contract will automatically revoke the employee's access rights to the company's internal facial data. Or when a project is completed, the relevant data access rights will also be adjusted accordingly according to preset rules. Through this dynamic permission management mechanism, it can ensure that data is always accessed and used within a safe and reasonable range.

[0056] Smart contracts play a key role in ensuring the security of data sharing. When data sharing is required, smart contracts will automatically perform relevant operations according to pre-set sharing protocols and security rules. For example, when facial feature data is shared between different departments for collaborative work, smart contracts will encrypt and verify the data transmission process to ensure that the data is not tampered with or stolen during transmission. At the same time, smart contracts will also record detailed information on each data sharing, including the time of sharing, participants, and shared data content, for subsequent auditing and tracing.

[0057] Smart contracts can also achieve fine-grained control over data sharing. It can segment and filter data according to different sharing needs and security levels, and only share the necessary data. For example, when collaborating with a third party to conduct market research, it may only be necessary to share some anonymized facial feature data, such as age distribution, gender ratio and other statistical information, without sharing specific personal identity information and complete facial image data. Through this fine-grained control, it can meet the needs of data sharing and protect data security and privacy to the greatest extent.

[0058] In addition, the execution of smart contracts is based on the distributed ledger technology of blockchain, which is decentralized, tamper-proof and highly transparent. This means that once a smart contract is deployed on the blockchain, its execution process will not be controlled or affected by a single node, and all operation records will be publicly recorded on the blockchain, which can be viewed and verified by any participant at any time. This further enhances the security and credibility of data storage and management, and provides a solid technical guarantee for data access rights management and shared security control.

[0059] In the present invention, the multiple innovative output modes in the result output and interaction module support user personalized customization and scene adaptive switching to meet the needs of different users and application scenarios.

[0060] A processing method based on facial feature recognition comprises the following steps:

[0061] Image acquisition: The image acquisition module uses multi-source fusion imaging technology and automatic environmental perception technology to collect image information containing faces, and performs preliminary image preprocessing, such as noise reduction and contrast enhancement.

[0062] Feature extraction: The preprocessed image is transferred to the feature extraction module, and the quantum neural computing model and quantum annealing algorithm are used to efficiently extract facial features.

[0063] Identification and matching: The extracted facial features are safely and quickly matched with the features in the database using a high-speed matching algorithm based on chaotic encryption.

[0064] Data storage: The recognition results and facial feature data are securely stored using blockchain distributed storage architecture and intelligent data compression technology.

[0065] Result output: The result output and interaction module presents the recognition results to users or related systems in a variety of innovative output methods and supports user interaction operations.

[0066] In the present invention, in the image acquisition step, dynamic tracking technology is used to ensure that the human face is always in the best imaging position.

[0067] In the present invention, feature enhancement technology is innovatively used in the feature extraction step to achieve highlighting and optimization of key features.

[0068] Feature enhancement technology first involves multi-dimensional analysis of the collected facial images. By using advanced image processing algorithms and mathematical models, the system can carefully disassemble and evaluate the various parts and features of the face. For example, for the eyes of the face, the system will analyze its shape, size, contour, and the texture and color distribution of the eyeball. For the nose, it will pay attention to details such as the height of the nose bridge, the width of the nose wings, and the shape of the nostrils. For the mouth, it will study the lip shape, lip color, and visible features of the teeth. Through detailed analysis of the features of these different parts, the system can establish a comprehensive and accurate initial model of facial features.

[0069] After determining the initial feature model, feature enhancement technology uses a series of methods to highlight key features. One of the important methods is contrast enhancement. By adjusting the brightness and color contrast of different areas in the image, key features are made more eye-catching in the image. For example, for the contour edge of a face, by increasing the contrast between it and the background, it is made clearer and more recognizable. For some unique facial features, such as moles, scars, or special textures, by increasing their brightness and contrast in the image, they are made more prominent. In this way, in the subsequent feature extraction process, these key features can be more easily and accurately identified and extracted by the system.

[0070] In addition to contrast enhancement, feature enhancement technology also uses feature amplification and sharpening methods. For some subtle but critical features, such as wrinkles around the eyes and fine lines at the corners of the mouth, the image amplification technology is used to locally amplify the images of these areas so that the system can capture these detailed features more clearly. At the same time, the sharpening algorithm is used to enhance the edges of these features to make them sharper and clearer. For example, when identifying older users, these subtle facial wrinkles may be important features to distinguish different individuals. Through the processing of feature enhancement technology, it can be ensured that these features will not be ignored during the feature extraction process.

[0071] In addition, feature enhancement technology also combines the attention mechanism in deep learning. The system trains an attention model to automatically learn which features are the most critical and discriminative for face recognition. During the feature extraction process, this attention model will automatically focus more computing resources and attention on these key features, thereby further improving the accuracy and reliability of key feature extraction. For example, when facing people with similar faces, the attention model may focus on some unique feature differences on their faces, such as differences in the shape of eyebrows and the contours of ears, etc., and improve recognition accuracy by emphasizing the extraction and analysis of these key features.

[0072] At the same time, feature enhancement technology also has the ability to adjust dynamically. According to different application scenarios and the actual situation of the face, the system can automatically adjust the parameters and methods of feature enhancement. For example, in security monitoring scenarios, for fast-moving face images, the system will use a faster and more efficient feature enhancement algorithm to ensure that key features can be accurately extracted in a short time. In high-precision identity authentication scenarios, the system will perform more detailed and comprehensive feature enhancement processing to meet higher requirements for recognition accuracy.

[0073] In the present invention, advanced emotion perception technology is fully utilized in the result output step to achieve intelligent adjustment of output mode and content according to the user's emotional state and attention focus.

[0074] Emotion perception technology mainly relies on a series of multimodal sensors and intelligent analysis algorithms. First, through high-precision biosignal sensors, such as skin galvanic response sensors and heart rate monitoring sensors, the user's physiological signal data is collected in real time. These physiological signals can reflect the user's emotional changes. For example, when the user is in an excited state, the skin galvanic response will increase and the heart rate will increase accordingly. At the same time, advanced computer vision technology is used to analyze the user's facial expressions and eyes in real time. For example, when the user has obvious facial expression changes such as smiling or frowning, the system can quickly identify and judge their emotional tendencies.

[0075] After determining the user's emotional state, the system will adjust the output method accordingly according to different emotional types. If the user is detected to be in a pleasant emotional state, the interface of the result output can use bright, vivid colors and cheerful sound effects to enhance the user's positive experience. For example, in an entertainment application scenario, when the system recognizes that the user is enjoying an interactive session and is in high spirits, it can pop up some interesting animation effects or play a relaxing music while displaying the recognition results to further enhance the user's entertainment experience. When the system perceives that the user is in a nervous or anxious emotional state, the output method will be more concise and clear to avoid too much complex information interfering with the user. For example, in a face recognition verification scenario for a financial transaction, if the system finds that the user is nervous, it may simplify the display interface of the verification result, highlight the key pass or fail information, and reduce unnecessary text descriptions and decorative elements to help users quickly understand the results and relieve their nervousness.

[0076] To determine the focus of attention, the system will use a combination of eye tracking technology and scene analysis algorithms. Eye tracking technology can accurately monitor the user's eye movement trajectory and gaze point position. For example, when a user is viewing the face recognition results, the system can understand the key areas of the user's attention on the interface through eye tracking data. If the user's attention is mainly focused on a specific detailed information area, such as the identity information part of the recognition result, the system can automatically enlarge the display ratio of the area or provide more relevant detailed information links. At the same time, the scene analysis algorithm will combine the current application scenario and surrounding environment information to further assist in determining the user's focus of attention. For example, in a smart shopping scenario, if the system detects that the user is standing in front of a product display area for face recognition, and his eyes are mainly focused on the price tag of the product, the system can automatically push the product's discount information and related purchase suggestions while outputting the face recognition results.

[0077] In addition, the system also has the ability to adjust dynamically. As the user's emotional state and focus of attention change, the output method and content will be optimized and updated in real time. For example, in a long human-computer interaction process, the user's emotions may change due to various factors, or their focus of attention may shift from one area to another. The system will continue to monitor these changes and adjust the output strategy in a timely manner. For example, in an online education face recognition sign-in scenario, if the user's attention is focused on the face recognition prompt information on the sign-in interface at the beginning, the system will guide the user to complete the face recognition operation with clear text and animation. Once the sign-in is successful, if the system finds that the user's attention has shifted to the course content display area, it can display the successful sign-in information in a concise pop-up window without interfering with the user, and provide a link to quickly enter the course learning.

[0078] Through this intelligent output adjustment method based on emotion perception technology, the system of the present invention can provide users with a more personalized, comfortable and efficient interactive experience.

[0079] The above are only preferred specific implementation modes of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical solutions and inventive concepts of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A processing system based on facial feature recognition, including an image acquisition module, a feature extraction module, a recognition and comparison module, a data storage and management module, and a result output and interaction module; The image acquisition module uses multi-source fusion imaging technology, combined with high-resolution optical sensors and depth perception sensors, to capture the three-dimensional morphology and texture information of the face, and uses infrared-assisted imaging to acquire face images in a dark environment. The image acquisition quality evaluation index formula is: Where Q image Indicates the image acquisition quality, R resolution Indicates the image resolution, D depth_Accuracy Indicates the accuracy of depth data, S stability Indicates image stability, F adaptability Indicates environmental adaptability, N noise represents image noise; The feature extraction module uses a quantum neural computing model, which is based on the superposition and entanglement characteristics of quantum bits to extract the fusion information of microscopic and macroscopic features of the face, including the structural features of the facial features, the dynamic change characteristics of facial muscles, and the characteristics of bioelectric signals. The accuracy evaluation index formula is: Among them A extract represents the feature extraction accuracy, N correct_features Indicates the number of correctly extracted features, N total_features Indicates the total number of features to be extracted, E quantum_enhancement represents the quantum computing enhancement factor; The recognition and comparison module uses a matching algorithm based on chaotic encryption to compare the extracted facial features with the features in the database, and uses the randomness and complexity of the chaotic system to encrypt data transmission and generate dynamic keys during the matching process. The recognition and comparison speed evaluation index formula is: Where V recognition Indicates the recognition and comparison speed, N faces_processed Indicates the number of faces processed per unit time, T processing represents the processing time, C chaos_factor represents the chaotic encryption acceleration factor; The data storage and management module combines blockchain distributed storage architecture with intelligent data compression technology. The data storage capacity evaluation index formula is: Among them C storage Indicates the data storage capacity, N compressed_data_stored Indicates the amount of compressed data actually stored, S storage_space Indicates the total space of the storage device, D blockchain_efficiency Represents the blockchain storage efficiency factor; The result output and interaction module supports virtual reality feedback, augmented reality visual guidance, and brain-computer interface thought interaction response. Through AR technology, the recognition results are superimposed on the real scene in the form of virtual labels to be displayed to security personnel, or through brain-computer interface devices, users can interact with the system through thought commands.

2. A processing system based on facial feature recognition according to claim 1, characterized in that: The multi-source fusion imaging technology in the image acquisition module supports automatic environmental perception and adaptive parameter adjustment, and automatically optimizes imaging parameters according to different environmental factors such as light, temperature and humidity.

3. A processing system based on facial feature recognition according to claim 1, characterized in that: The quantum neural computing model in the feature extraction module uses a quantum annealing algorithm to perform optimization training on the model.

4. The processing system based on facial feature recognition according to claim 1, characterized in that: The high-speed matching algorithm based on chaotic encryption in the identification and comparison module supports dynamic key update and multi-modal feature fusion comparison.

5. The processing system based on facial feature recognition according to claim 1, characterized in that: The blockchain distributed storage architecture in the data storage and management module adopts smart contracts to realize the automated management of data access rights and the security control of data sharing.

6. A processing system based on facial feature recognition according to claim 1, characterized in that: The output mode in the result output and interaction module supports user personalized customization and scene adaptive switching.

7. A processing method based on facial feature recognition, used to implement a processing system based on facial feature recognition as described in any one of claims 1 to 6, characterized in that: The following steps are involved: Image acquisition: The image acquisition module uses multi-source fusion imaging technology and automatic environment perception technology to collect image information containing faces, and performs preliminary image preprocessing, such as noise reduction and contrast enhancement; Feature extraction: The preprocessed image is transferred to the feature extraction module, and the quantum neural computing model and quantum annealing algorithm are used to efficiently extract facial features; Identification and comparison: The extracted facial features are compared safely and quickly with the features in the database using a high-speed matching algorithm based on chaotic encryption; Data storage: The recognition results and facial feature data are securely stored using blockchain distributed storage architecture and intelligent data compression technology; Result output: The result output and interaction module presents the recognition results to users or related systems in a variety of innovative output methods and supports user interaction operations.

8. The processing method based on facial feature recognition according to claim 6, characterized in that: In the image acquisition step, dynamic tracking technology is used to ensure that the face is always in the best imaging position.

9. The processing method based on facial feature recognition according to claim 6, characterized in that: In the feature extraction step, feature enhancement technology is used to highlight and optimize key features.

10. The processing method based on facial feature recognition according to claim 6, characterized in that: In the result output step, emotion perception technology is used to adjust the output method and content according to the user's emotional state and attention focus.

Images