Information recognition module and data processing method

Abstract

This application relates to an information recognition module and a data processing method. The information recognition module includes: an image recognition module, including a first image sensor, for acquiring first image data for panoramic monitoring; a biometric recognition module, including a second image sensor, for acquiring second image data for biometric recognition; a single SoC chip, the single SoC chip having at least two interfaces, the at least two interfaces being used for electrical connection with the first image sensor and the second image sensor respectively; the single SoC chip having multiple parallel data channels, the multiple parallel data channels being used for isolation and decoupling of the first image data and the second image data; the single SoC chip is also used for unified processing of the first image data and the second image data. This application saves hardware costs and physical space through high integration.

Description

Technical Field

[0001] This application relates to the field of smart device technology, and in particular to an information identification module and data processing method. Background Technology

[0002] With the increasing popularity of smart homes, smart door locks have become an important part of many home security systems. Smart door locks not only have basic unlocking functions but also integrate a growing number of intelligent features, such as real-time monitoring of the external environment via panoramic peepholes and identity authentication using biometric technologies like facial recognition and palm vein recognition. To achieve these functions, traditional smart door locks generally adopt a hardware architecture of "multi-chip + discrete modules." However, this discrete architecture suffers from problems such as hardware stacking and high power consumption. Summary of the Invention

[0003] Therefore, it is necessary to provide an information recognition module and data processing method that can optimize and reduce system power consumption and hardware costs to address the above-mentioned technical problems.

[0004] In a first aspect, this application provides an information recognition module, comprising: an image recognition module including a first image sensor for acquiring first image data for panoramic monitoring; a biometric recognition module including a second image sensor for acquiring second image data for biometric recognition; a single SoC chip having at least two interfaces for electrically connecting to the first image sensor and the second image sensor respectively; the single SoC chip having multiple parallel data channels for isolating and decoupling the first image data and the second image data; and the single SoC chip also being used for unified processing of the first image data and the second image data.

[0005] In one embodiment, the information recognition module further includes: a first supplementary light source for emitting infrared light in a second band; and a second supplementary light source integrated into the biometric module for emitting infrared light in a third band; the image recognition module further includes a filter configured to transmit visible light in the first band and infrared light in the second band emitted by the first supplementary light source, and to block infrared light in the third band emitted by the second supplementary light source, so as to prevent infrared light in the third band from interfering with the panoramic monitoring image when the biometric module is working.

[0006] In one embodiment, the infrared light emitted by the first supplementary light source has a wavelength of 940 nm; the infrared light emitted by the second supplementary light source has a wavelength of 850 nm; and the filter has a cutoff effect on the 850 nm wavelength light.

[0007] In one embodiment, the image recognition module includes a panoramic lens with a field of view greater than 180 degrees, the imaging circle of which falls within the effective light-sensitive area of ​​the first image sensor.

[0008] In one embodiment, the biometric module includes an infrared lens, which is a lens with a diagonal field of view greater than 85 degrees.

[0009] In one embodiment, the image recognition module and the biometric module are located on a first printed circuit board assembly and a second printed circuit board assembly, respectively, with the first printed circuit board assembly connected to the second printed circuit board assembly via a flexible flat cable (FFC).

[0010] Secondly, this application also provides a data processing method applied to an information recognition module including an image recognition module, a biometric recognition module, and a single SoC chip. The image recognition module includes a first image sensor, and the biometric recognition module includes a second image sensor. The single SoC chip has at least two interfaces, which are respectively used for electrical connection with the first image sensor and the second image sensor. The method includes:

[0011] First image data for panoramic monitoring is acquired by a first image sensor;

[0012] Second image data for biometric identification is acquired by a second image sensor;

[0013] The first image data and the second image data are isolated and decoupled through multiple parallel data channels within a single SoC chip, and the first image data and the second image data are processed uniformly through a single SoC chip.

[0014] In one embodiment, the first image data and the second image data are isolated and decoupled through multiple parallel data channels within a single SoC chip, including:

[0015] A multi-channel data routing strategy is adopted to route multiple data streams of the first image data and multiple data streams of the second image data to multiple parallel data channels respectively; the multi-channel data routing strategy indicates that the multiple parallel data channels correspond to different business needs.

[0016] In one embodiment, the resolution information of the multiple parallel data channels is different.

[0017] In one embodiment, the multiple data streams of the second image data include a first data stream for palm vein feature extraction and a second data stream for face recognition;

[0018] Multiple parallel data channels include a first data path corresponding to a first data stream and a second data path corresponding to a second data stream, wherein the resolution of the first data path is greater than the resolution of the second data path.

[0019] The aforementioned information recognition module and data processing method include an information recognition module whose single SoC chip has at least two interfaces, which are respectively used for electrical connection with a first image sensor and a second image sensor; the single SoC chip has multiple parallel data channels, which are used for isolation and decoupling of the first image data and the second image data; the single SoC chip is also used for unified processing of the first image data and the second image data. This application simultaneously realizes panoramic visual monitoring and biometric recognition within a single SoC chip. Through high integration, it saves hardware costs and physical space. Based on hardware-level independent control logic and parallel data processing strategies, it achieves physical isolation and resource decoupling of different video stream processing tasks, ensuring the real-time performance, stability, and non-interference of data processing in multiple service scenarios. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 This is a schematic diagram of the structure of the information recognition module in one embodiment;

[0022] Figure 2 This is a schematic diagram of the spectrum of a filter in one embodiment;

[0023] Figure 3 This is a schematic diagram illustrating the relationship between the effective acquisition area of ​​the image sensor and the imaging circle of the lens in one embodiment;

[0024] Figure 4 This is a block diagram of the internal structure of the information recognition module in one embodiment;

[0025] Figure 5 This is a flowchart illustrating a data processing method in one embodiment;

[0026] Figure 6 This is a schematic diagram illustrating the application of a multi-channel data routing strategy in one embodiment;

[0027] Figure 7 This is a schematic diagram illustrating the application of a multi-channel data routing strategy in another embodiment. Detailed Implementation

[0028] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0029] It should be noted that the terms "first," "second," etc., used in this application can be used to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish the first element from the second element. The terms "comprising" and "having," and any variations thereof, used in this application, are intended to cover non-exclusive inclusion. The term "multiple" used in this application refers to two or more. The term "and / or" used in this application refers to one of the embodiments, or any combination of multiple embodiments.

[0030] It is understood that the term "connection" in the following embodiments should be interpreted as "electrical connection," "communication connection," etc., if the connected circuits, modules, units, etc., transmit electrical signals or data to each other. When used herein, the singular forms "a," "an," and "the" may also include the plural forms, unless the context clearly indicates otherwise. "At least one" means one or more. "At least a portion of an element" means part or all of an element.

[0031] In traditional technology, smart door locks often adopt a "multi-chip + discrete module" architecture. For example, panoramic peepholes require independent video processing chips and camera modules, while facial and palm vein recognition relies on dedicated AI (Artificial Intelligence) acceleration modules. Traditional designs suffer from at least the following drawbacks: ① Hardware redundancy: The coexistence of multiple chips results in a large total PCB (Printed Circuit Board) area and complex structure, hindering miniaturization design; ② Computing silos: Each functional module operates independently, preventing data sharing and leading to high system response latency; ③ Excessive power consumption: Multiple chips working simultaneously result in high overall power consumption, impacting battery life; ④ High cost: The need to purchase multiple modules leads to high debugging and assembly costs.

[0032] To address these issues, the industry has attempted functional integration, using a single camera to simultaneously perform peephole monitoring and facial vein recognition. However, this solution only supports wide-angle peepholes with a field of view of less than 180 degrees. Furthermore, due to the significant edge distortion of wide-angle lenses, the biometric recognition area is forced to be limited to the less distorted central region, restricting the recognition angle and thus placing higher demands on algorithm compatibility. In conclusion, the multi-chip architecture in traditional technologies easily leads to problems such as low integration, high power consumption, and slow response.

[0033] This application relates to a module that integrates a panoramic peephole and a facial palm vein recognition system, which can be applied to smart door locks. This module, through a highly integrated SoC (System on Chip) chip, achieves integrated processing of panoramic monitoring, multimodal biometrics, and edge computing, solving problems such as hardware stacking, distributed computing power, and high power consumption in traditional door locks.

[0034] In this embodiment, a module is designed using a single video processing chip, namely a single-chip panoramic cat-eye and face palm vein recognition hardware module, thereby realizing the single-chip fusion of "panoramic vision + dual-modal biometrics". For example, the chip can support two video inputs: one is connected to an image sensor for cat-eye function, which, together with a panoramic lens with dual-pass 650nm and 940nm wavelength filters and a 940nm wavelength fill light, can realize day and night monitoring with a field of view greater than 180 degrees; the other is connected to an image sensor for face palm vein recognition, which, together with an infrared lens with a diagonal field of view (FOV) greater than 85 degrees and an 850nm wavelength fill light, can realize distortion-free biometrics.

[0035] This application enables unified processing of panoramic video streams, face and palm vein image data within a single SoC chip (hereinafter referred to as SoC chip), optimizing and reducing system power consumption and hardware costs, improving system real-time performance, and avoiding the risk of communication delays between multiple chips. It should be noted that the beneficial effects or technical problems solved by the embodiments of this application are not limited to this one, but may also include other implicit or related problems, which can be referred to in the following description of the embodiments.

[0036] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0037] In one exemplary embodiment, such as Figure 1 As shown, an information recognition module is provided, including: an image recognition module 22, including a first image sensor 202, for acquiring first image data for panoramic monitoring; a biometric recognition module 24, including a second image sensor 204, for acquiring second image data for biometric recognition; a single SoC chip 26, the single SoC chip 26 having at least two interfaces, the at least two interfaces being used for electrical connection with the first image sensor 202 and the second image sensor 204 respectively; the single SoC chip 26 having multiple parallel data channels, the multiple parallel data channels being used for isolation and decoupling of the first image data and the second image data; the single SoC chip 26 is also used for unified processing of the first image data and the second image data.

[0038] Specifically, the single SoC chip 26 in this embodiment can be an integrated circuit capable of performing data processing, such as a system-on-chip (SoC); further, the single SoC chip 26 is configured to execute artificial intelligence algorithms, such as face recognition or palm vein feature extraction. The single SoC chip 26 has at least two independent image sensor interfaces (hereinafter referred to as interfaces) for simultaneously receiving video streams from different image sensors, for example, electrically connected to the first image sensor 202 and the second image sensor 204 respectively, to receive the image data acquired by them; the single SoC chip 26 has multiple parallel data channels. Exemplarily, parallel data channels can refer to parallel hardware data paths integrated within the single SoC chip 26, such as multiple image signal processing (ISP) channels, capable of concurrent processing of multiple data streams.

[0039] In this embodiment, multiple parallel data channels serve as independent image processing pipelines within a single SoC chip 26, enabling concurrent and non-interfering processing of data streams from one or more image sensors, thus achieving hardware-level resource isolation. Furthermore, this embodiment can utilize a single SoC chip to process the first and second image data uniformly. Both image data streams enter the same SoC for decoding, image enhancement, target detection, and biometric comparison, reducing hardware costs, avoiding large-scale image data transmission between multiple chips, and eliminating the overhead of high-speed data interaction between chips. Additionally, a single SoC chip can first analyze the panoramic monitoring image, detect the target area, and then simultaneously acquire the corresponding biometric image, completing face capture and comparison within the chip, achieving integrated panoramic large-area monitoring and local high-precision biometric recognition.

[0040] Taking a single SoC chip 26 as an example, which supports dual-channel video input and has AI computing power, compared with the traditional multi-chip discrete architecture of cat eye and face palm vein recognition modules, this application can save a video processing chip and its supporting power supply and logic control circuit, effectively reducing hardware costs and power consumption, and enhancing the real-time performance of system communication.

[0041] For example, the image recognition module 22 may include, but is not limited to, a panoramic peephole module, which may be a component for acquiring panoramic images with a field of view greater than 180 degrees; the image recognition module 22 includes a first image sensor 202; optionally, the image recognition module 22 in this embodiment may also include a panoramic lens and a filter.

[0042] The biometric module 30 refers to a component used to collect images of a user's face and / or palm veins and other biometric features. The biometric module 30 includes a second image sensor 204. Exemplarily, the biometric module 30 in this application embodiment may also include an infrared lens and a supplementary light source for active illumination.

[0043] The aforementioned information recognition module integrates panoramic monitoring and biometric recognition into a single SoC chip, leveraging its multi-interface and parallel processing capabilities to fundamentally solve the problems of hardware redundancy and computing power silos. Specifically, by employing a highly integrated single SoC chip that supports multiple video inputs and possesses artificial intelligence (AI) computing power, combined with a specific physical structure layout, it achieves both panoramic monitoring and multimodal biometric recognition functions within a single module. Simultaneously, it resolves the mutual interference issues when these two functions operate concurrently, thus providing a high-performance, low-power, and highly integrated smart door lock front-end vision solution.

[0044] In one embodiment, the information recognition module may further include: a first supplementary light source for emitting infrared light in a second band; and a second supplementary light source integrated into the biometric module for emitting infrared light in a third band; the image recognition module further includes a filter configured to transmit visible light in the first band and infrared light in the second band emitted by the first supplementary light source, and to block infrared light in the third band emitted by the second supplementary light source, so as to prevent infrared light in the third band from interfering with the panoramic monitoring image when the biometric module is working.

[0045] Specifically, the information recognition module may also include two supplementary light sources: a first supplementary light source and a second supplementary light source. The first supplementary light source can be used to provide night vision illumination for the image recognition module in low-light environments, and it can emit second-band infrared light. The second supplementary light source can be integrated into the biometric module to provide active illumination during biometric recognition, and it can emit third-band infrared light. In practical applications, the second-band and third-band infrared light are usually different to serve different functional requirements.

[0046] Furthermore, the filter, as an optical element in the embodiments of this application, is located in the optical path of the image recognition module and can selectively transmit light of a specific wavelength band while blocking or reflecting light of other wavelength bands. For example, the filter is configured to allow visible light of the first wavelength band (for daytime imaging) and infrared light of the second wavelength band (for nighttime imaging) emitted by the first supplementary light source to pass through, but at the same time can effectively block or block infrared light of the third wavelength band emitted by the second supplementary light source.

[0047] When the biometric module is activated and the second supplementary light source illuminates the user's biometric features, the emitted third-band infrared light inevitably radiates into the surrounding environment. Without special processing, this infrared light enters the lens of the image recognition module (e.g., a panoramic peephole module), causing the first image sensor to receive a large amount of useless interference signals, resulting in severe red exposure, overexposure, or color distortion in the monitoring image. To address this, this application proposes introducing a filter to filter the interference light from the second supplementary light source, thereby ensuring that even at the moment the biometric function is activated, the image recognition module can still capture a clear, interference-free monitoring image (i.e., a panoramic monitoring image). Thus, this application embodiment achieves concurrent and interference-free operation of both panoramic monitoring and biometric functions within a single integrated module.

[0048] In summary, this application solves the optical interference of the biometric module on the image recognition module during operation through optical design, and realizes the coexistence of the two major functions in a single module.

[0049] In some embodiments, the infrared light emitted by the first supplementary light source has a wavelength of 940 nm; the infrared light emitted by the second supplementary light source has a wavelength of 850 nm; and the filter has a cutoff effect on light with a wavelength of 850 nm. Specifically, regarding the wavelength of the supplementary light source and the spectral characteristics of the filter, this application embodiment proposes that the infrared light emitted by the first supplementary light source can be 940 nm, while the infrared light emitted by the second supplementary light source can be 850 nm. For scenarios where panoramic monitoring and biometric recognition work concurrently, when the 850 nm supplementary light is turned on, the 850 nm wavelength is cut off by the panoramic cat-eye lens filter to prevent the image sensor from collecting light in this band, which would cause red exposure in the monitoring image.

[0050] Taking the panoramic cat-eye module as an example, such as the image recognition module, Figure 2 The spectrum of the filter shown is illustrated. For scenarios where panoramic monitoring and biometric identification operate concurrently, this application proposes using a panoramic cat-eye lens with dual-pass filters for visible light (400nm~650nm) and infrared light (900~1000nm) to solve the red-exposure problem in the image. In daytime mode, the filter primarily transmits 400nm~650nm visible light for sensor imaging; in nighttime mode, it primarily transmits 900nm~1000nm infrared light for sensor imaging. When the 850nm supplementary light is turned on, the filter has a cutoff effect on this wavelength, effectively preventing interference from biometric identification supplementary light on the quality of the panoramic monitoring image.

[0051] In one embodiment, the image recognition module includes a panoramic lens with a field of view (FOV) greater than 180 degrees, and the imaging circle of the panoramic lens falls within the effective light-sensitive area of ​​the first image sensor. Specifically, for the optical performance of the image recognition module, in order to achieve blind-spot-free monitoring, it may include a panoramic lens with a field of view (FOV) greater than 180 degrees. Further, to ensure that the complete panoramic field of view can be acquired, this embodiment proposes that the imaging circle of the panoramic lens completely falls within the effective light-sensitive area of ​​the first image sensor. If the imaging circle is larger than the light-sensitive area, information at the edge of the field of view will be lost; if the imaging circle is too small, the pixel area of ​​the sensor will be wasted. This application achieves high-quality panoramic imaging by precisely matching the lens imaging circle with the sensor size.

[0052] Taking a single video input from a SoC chip connected to a 2-megapixel (4:3 aspect ratio) image sensor as an example, such as Figure 3 As shown, the first image sensor can be a 4:3 aspect ratio sensor with an effective light-sensitive area size of 2814μm × 2114μm (i.e., H=2814μm, W=2114μm), and the panoramic lens has an imaging circle diameter of φ2mm (i.e., φ=2mm). Taking the panoramic cat-eye module as an example, the panoramic cat-eye module uses a 4:3 aspect ratio image sensor with an effective light-sensitive area size of 2814μm × 2114μm. Combined with a panoramic lens with an imaging circle diameter of φ2mm and a field of view (FOV) greater than 180 degrees, it ensures that the lens's imaging circle falls within the sensor's effective area, thereby achieving 180-degree panoramic monitoring.

[0053] It should be noted that the biometric identification with a field of view greater than 180 degrees and without distortion in the embodiments of this application can all be achieved based on optical lenses and optical sensors (CIS).

[0054] In some embodiments, the biometric module includes an infrared lens with a diagonal field of view greater than 85 degrees. Specifically, to improve user experience, the optical design of the biometric module has also been optimized. The biometric module may include an infrared lens with a diagonal field of view greater than 85 degrees; a wider field of view means a larger recognition range.

[0055] Under preset installation conditions (e.g., camera installation height 1.2m, tilted downwards 15 degrees), using a lens with a field of view greater than 85 degrees, the relationship between the recognition height range and recognition distance covered by the biometric module is shown in Table 1. This lens configuration ensures that the recognition angle covers ±25 degrees. This embodiment of the application enables family members of different heights (including children and taller adults) to have their faces or hands reliably captured in natural standing positions and at appropriate distances, thereby greatly improving the convenience and success rate of biometric identification.

[0056] Table 1

[0057]

[0058] In one embodiment, the image recognition module and the biometric recognition module are located on a first printed circuit board assembly and a second printed circuit board assembly, respectively, with the first printed circuit board assembly connected to the second printed circuit board assembly via a connector. Specifically, this application optimizes the physical structure of the information recognition module to adapt to the compact and irregularly shaped structural requirements of devices such as smart door locks.

[0059] In this application, the image recognition module and the biometric module are located on a first printed circuit board assembly and a second printed circuit board assembly, respectively, and the first and second printed circuit board assemblies are physically separated. A single SoC chip can be disposed on the second printed circuit board assembly containing the biometric module. The first printed circuit board assembly can be connected to the second printed circuit board assembly via a connector (e.g., a flexible flat cable, FFC). In practical applications, panoramic peephole lenses typically need to protrude from the door lock panel to obtain a wide field of view, while the main processing circuit can be installed flush inside the door lock. This separate design allows the module to flexibly adapt to the industrial designs of various door locks, achieving structural decoupling and ease of installation.

[0060] Furthermore, in the embodiments of this application, the connector may be a flexible electrical connector, such as a flexible flat cable (FFC), a flexible cable used to reliably transmit electrical signals and power between two separate printed circuit board assemblies to accommodate compact or non-planar assembly requirements.

[0061] Taking the image recognition module as an example (the panoramic peephole module, peephole hardware) used to implement the panoramic peephole function, and the biometric module as an example (face and palm vein recognition module, face and palm vein recognition hardware) used to implement the recognition function, and using a single SoC chip as an example, given that the panoramic peephole structure needs to protrude from the door lock plane, such as... Figure 4 As shown in the embodiment of this application, the cat-eye hardware and the face palm vein recognition hardware are separated into two independent PCBAs, and the cat-eye PCBA is connected to the SoC chip via an FFC cable. Using an FFC cable to connect the panoramic cat-eye PCBA and the recognition PCBA can reduce cable costs and adapt to installation in confined spaces.

[0062] This application embodiment employs a SoC chip that supports dual-channel video input and has AI computing power, and it is placed on the recognition function PCBA (Printed Circuit Board Assembly). One of the video inputs is connected to a 2-megapixel (4:3 aspect ratio) image sensor, combined with an infrared lens with a diagonal field of view (FOV) greater than 85 degrees and an 850nm fill light, thereby reducing hardware cost and power consumption.

[0063] The aforementioned information recognition module features high integration and space saving. Replacing multiple traditional chips with a single SoC chip reduces PCB area by over 60%, facilitating thinner and lighter door lock designs. Shared computing power ensures faster response, with a unified memory architecture and high-speed bus design enabling efficient data transfer from multiple sensors within the chip. Power consumption is significantly reduced by minimizing power-intensive components, thus extending battery life. A clear cost advantage is achieved by reducing the number of external components, lowering BOM (Bill of Materials) costs, and reducing production and debugging complexity. This application realizes ultimate integration and performance optimization of the smart door lock front-end module.

[0064] Based on the same inventive concept, in an exemplary embodiment, such as Figure 5 As shown, this application embodiment also provides a data processing method for smart door locks. Taking the application of this method to the above-mentioned hardware system as an example, it includes the following steps 302 to 306.

[0065] Step 302: First image data for panoramic monitoring is acquired by the first image sensor.

[0066] Specifically, a single SoC chip can receive first image data acquired by the first image sensor for panoramic monitoring through an interface for electrical connection with the first image sensor.

[0067] Step 304: The second image sensor acquires second image data for biometric identification.

[0068] Specifically, a single SoC chip can receive second image data acquired by the second image sensor for biometric identification through an interface for electrical connection with the second image sensor.

[0069] Step 306: Isolate and decouple the first image data and the second image data through multiple parallel data channels within a single SoC chip, and perform unified processing on the first image data and the second image data through a single SoC chip.

[0070] Specifically, the image recognition module transmits the acquired first image data to multiple parallel data channels within a single SoC chip for processing via a first image sensor, while the biometric recognition module transmits the acquired second image data to multiple parallel data channels within a single SoC chip for processing via a second image sensor, thereby isolating and decoupling the first and second image data. Furthermore, the first and second image data can also be processed uniformly within a single SoC chip.

[0071] Taking a single SoC chip as an example, this application embodiment utilizes the SoC chip's multi-sensor interface and multi-ISP (Image Signal Processor) channel capabilities to achieve concurrent processing of ultra-wide-angle video, face, and palm vein data within a single chip. Shared computing power results in faster response, and a unified memory architecture and high-speed bus design enable efficient flow of multi-sensor data within the chip.

[0072] The aforementioned data processing method utilizes the multi-sensor interface and multiple parallel data channels of a single SoC chip to achieve concurrent processing of panoramic monitoring data and biometric data within that chip. This improves system performance and response speed. The unified computing platform and memory architecture enable high-speed flow and sharing of panoramic video streams and biometric data streams within the chip, avoiding cross-chip communication delays and significantly enhancing the system's real-time response speed.

[0073] Furthermore, this application achieves physical isolation and resource decoupling of different video stream processing tasks through hardware-level independent control logic and data routing strategies, ensuring the real-time performance, stability, and non-interference of data processing in multi-service scenarios. In one embodiment, the first image data and the second image data are isolated and decoupled through multiple parallel data channels within a single SoC chip, including: employing a multi-channel data routing strategy to route multiple data streams of the first image data and multiple data streams of the second image data to multiple parallel data channels respectively; the multi-channel data routing strategy indicates that the multiple parallel data channels correspond to different service requirements.

[0074] Specifically, a single SoC chip employs a multi-channel data routing strategy, with each channel's data operating independently to support different business requirements. Specifically, this strategy can route multiple data streams of the first image data and multiple data streams of the second image data to multiple parallel data channels.

[0075] In one embodiment, the resolution information of multiple parallel data channels is different. Specifically, in the embodiments of this application, the resolution information of different parallel data channels is different. For example, the parallel data channel corresponding to the data stream of the second image data has a larger resolution and can be dedicated to biometric feature extraction to meet the biometric feature resolution requirements.

[0076] In one embodiment, the multiple data streams of the second image data include a first data stream for palm vein feature extraction and a second data stream for face recognition; the multiple parallel data channels include a first data path corresponding to the first data stream and a second data path corresponding to the second data stream, wherein the resolution of the first data path is greater than the resolution of the second data path.

[0077] Specifically, for multiple data streams of the second image data, such as a first data stream for palm vein feature extraction and a second data stream for face recognition, the resolution of the first data path corresponding to the first data stream is greater than the resolution of the second data path corresponding to the second data stream.

[0078] The first data path for palm vein recognition and the second data path for face recognition each have their own independent memory regions, direct memory access (DMA) channels, and processing logic units at the hardware level. This hardware-level physical isolation avoids resource contention, preemption, and blocking issues that may arise from multi-task scheduling at the software level, ensuring stable and latency-free computing resources for both palm vein recognition and face recognition tasks. This is suitable for smart locks requiring high security and fast response. This application achieves efficient extraction of two biometric features from a single image through hardware parallel processing.

[0079] Taking an image recognition module using an ultra-wide-angle camera and a biometric recognition module using an infrared (IR) camera as an example, the ultra-wide-angle camera can be configured as an all-weather video acquisition source, responsible for daytime and nighttime video stream input, while the infrared (IR) camera can be configured as a biometric feature acquisition source. Figure 6 As shown, for ultra-wide-angle cameras, the main data path has the highest resolution, while the secondary data path has a lower resolution. For example... Figure 7 As shown, for IR cameras, a multi-channel data routing strategy is adopted, with each channel's data independent to support different business needs: the main data path has the highest resolution and is dedicated to palm vein feature extraction, meeting the resolution requirements of the palm vein algorithm. The secondary channel (sub / third path) has the next lowest resolution and is allocated to face recognition services to meet the resolution requirements of the face model in face recognition services; the third channel has the lowest resolution and is allocated to face recognition services to meet the resolution requirements of the face liveness model in face recognition services.

[0080] This application uses dual-lens multi-channel data routing to achieve physical isolation and resource decoupling between ultra-wide-angle video streams and biometric data streams, ensuring the real-time performance, stability, and non-interference of data processing in multiple business scenarios.

[0081] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages in other steps. It is understood that the steps in different embodiments can be freely combined as needed, and all non-contradictory solutions formed by such combinations are within the scope of protection of this application.

[0082] Based on the same inventive concept, this application also provides a data processing apparatus for implementing the data processing method described above. The solution provided by this apparatus is similar to the implementation scheme described in the above method; therefore, the specific limitations in one or more data processing apparatus embodiments provided below can be found in the limitations of the data processing method described above, and will not be repeated here.

[0083] In one exemplary embodiment, a data processing apparatus is provided, applied to an information recognition module including an image recognition module, a biometric recognition module, and a single SoC chip. The image recognition module includes a first image sensor, and the biometric recognition module includes a second image sensor. The single SoC chip has at least two interfaces, which are respectively used for electrical connection to the first image sensor and the second image sensor. The apparatus includes:

[0084] The first image data acquisition module is used to acquire first image data for panoramic monitoring by the first image sensor.

[0085] The second image data acquisition module is used to acquire second image data for biometric identification by the second image sensor.

[0086] The parallel processing module is used to isolate and decouple the first image data and the second image data through multiple parallel data channels within a single SoC chip, and to perform unified processing of the first image data and the second image data through a single SoC chip.

[0087] In one embodiment, the parallel processing module is used to employ a multi-channel data routing strategy to route multiple data streams of the first image data and multiple data streams of the second image data to multiple parallel data channels respectively; the multi-channel data routing strategy indicates that the multiple parallel data channels correspond to different business requirements.

[0088] In one embodiment, the resolution information of the multiple parallel data channels is different.

[0089] In one embodiment, the multiple data streams of the second image data include a first data stream for palm vein feature extraction and a second data stream for face recognition; the multiple parallel data channels include a first data path corresponding to the first data stream and a second data path corresponding to the second data stream, wherein the resolution of the first data path is greater than the resolution of the second data path.

[0090] Each module in the aforementioned data processing device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the operations corresponding to each module.

[0091] In an exemplary embodiment, the information identification module in this application can be applied to a smart door lock. The smart door lock also includes a lock body and a main control board. The information identification module communicates with the main control board through a universal serial bus interface or a universal asynchronous transceiver interface.

[0092] Specifically, the smart lock integrates the aforementioned image recognition module and biometric module as its core front-end sensing and recognition components onto the outer panel of the lock. By adopting the integrated module (information recognition module) of this application embodiment, the smart lock can achieve both high-definition panoramic monitoring and fast and accurate multimodal biometric unlocking functions with a smaller size, lower power consumption, and lower cost.

[0093] Furthermore, in smart locks, communication between the image recognition module and the biometric module and the main control board of the lock (responsible for handling lock body control, network communication, and other functions) can be achieved through standard interfaces. Specifically, data is exchanged with other modules of the smart lock through the USB and UART interfaces on the PCBA.

[0094] For example, the information recognition module can interact with the main control board for data and commands via a Universal Serial Bus (USB) interface or a Universal Asynchronous Receiver / Transmitter (UART) interface. For instance, after completing biometric identification, the biometric module sends the identification result (such as a user ID or authentication success / failure flag) to the main control board via the UART interface, which then determines whether to drive the motor to unlock the door. This interface design ensures the module has good compatibility and portability.

[0095] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon that, when executed by a processor, implements the steps in the method embodiments described above.

[0096] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the method embodiments described above.

[0097] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0098] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0099] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. An information recognition module, characterized in that, include: The image recognition module includes a first image sensor that acquires first image data for panoramic monitoring; The biometric module includes a second image sensor that acquires second image data for biometric identification; A single SoC chip, the single SoC chip having at least two interfaces, the at least two interfaces being used for electrical connection to the first image sensor and the second image sensor respectively; The single SoC chip has multiple parallel data channels, which are used for the isolation and decoupling of the first image data and the second image data; The single SoC chip is also used to perform unified processing on the first image data and the second image data.

2. The information identification module according to claim 1, characterized in that, The information recognition module also includes: The first supplementary light source is used to emit infrared light in the second band; and The second supplementary light source, integrated into the biometric module, is used to emit infrared light in the third band; The image recognition module further includes a filter configured to transmit visible light of the first band and infrared light of the second band emitted by the first supplementary light source, and to block infrared light of the third band emitted by the second supplementary light source, so as to prevent the infrared light of the third band from interfering with the panoramic monitoring image when the biometric module is working.

3. The information identification module according to claim 2, characterized in that, The first supplementary light source emits infrared light with a wavelength of 940nm; the second supplementary light source emits infrared light with a wavelength of 850nm; and the filter has a cutoff effect on light with a wavelength of 850nm.

4. The information identification module according to claim 2, characterized in that, The image recognition module includes a panoramic lens with a field of view greater than 180 degrees, and the imaging circle of the panoramic lens falls within the effective light-sensitive area of ​​the first image sensor.

5. The information identification module according to claim 2, characterized in that, The biometric module includes an infrared lens, which has a diagonal field of view greater than 85 degrees.

6. The information identification module according to any one of claims 1 to 5, characterized in that, The image recognition module and the biometric module are located on the first printed circuit board assembly and the second printed circuit board assembly, respectively. The first printed circuit board assembly is connected to the second printed circuit board assembly via a flexible flat cable (FFC).

7. A data processing method, characterized in that, An information recognition module is applied to include an image recognition module, a biometric recognition module, and a single SoC chip. The image recognition module includes a first image sensor, and the biometric recognition module includes a second image sensor. The single SoC chip has at least two interfaces, which are respectively used for electrical connection to the first image sensor and the second image sensor. The method includes: First image data for panoramic monitoring is acquired by the first image sensor; The second image sensor acquires second image data for biometric identification; The first image data and the second image data are isolated and decoupled through multiple parallel data channels within the single SoC chip, and the first image data and the second image data are processed uniformly through the single SoC chip.

8. The method according to claim 7, characterized in that, The first image data and the second image data are isolated and decoupled through multiple parallel data channels within the single SoC chip, including: A multi-channel data routing strategy is adopted to route multiple data streams of the first image data and multiple data streams of the second image data to the multiple parallel data channels respectively; the multi-channel data routing strategy indicates that the multiple parallel data channels correspond to different service requirements.

9. The method according to claim 8, characterized in that, The resolution information of the multiple parallel data channels is different.

10. The method according to claim 9, characterized in that, The second image data includes a first data stream for palm vein feature extraction and a second data stream for face recognition. The plurality of parallel data channels include a first data path corresponding to the first data stream and a second data path corresponding to the second data stream, wherein the resolution of the first data path is greater than the resolution of the second data path.

Images