CN202041967U - Palm Vein and Face Dual Mode Biometric Recognition Instrument - Google Patents

Abstract

Palm vein and human face dual-mode biometric recognition instrument, the utility model belongs to the field of optical instruments and pattern recognition, uses a plurality of ARM processors, memories, touch screens, optical filters, etc., and finally forms a high-precision multi-biological Feature recognition equipment. The connection control mode of the ARM processor (8) (9) (10) and the ARM processor (11) can not only share the processing tasks of the main control unit ARM processor (11), but also make the task division of the recognition process more efficient. Clear, which greatly speeds up the recognition speed, and solves the problem of long recognition time caused by the large amount of calculation for palm vein and face recognition. The equipment is easy to use and has strong scalability, and can be widely used in personnel registration, access control, banking and other occasions.

Description

Palm Vein and Face Dual Mode Biometric Recognition Instrument

(1) Technical field

The utility model belongs to the field of optical instruments and pattern recognition, and is mainly a dual-mode biometric feature acquisition device based on near-infrared light, which uses a plurality of ARM processors, memories, touch screens, filters, etc., and finally forms a comprehensive biometric identification devices.

(2) Background technology

Biometrics technology is a computer security technology that has developed rapidly around the world in recent years. It uses image processing and pattern recognition to identify or verify identities based on human biological characteristics such as faces, fingerprints, voices, and veins. As a technology that uses human-specific physiological characteristics (such as fingerprints, faces, irises, veins, retinas, etc.) or behavioral characteristics (such as signatures, voices, gait, etc.) Forgery, uniqueness, and ease of use are gradually replacing traditional identification methods.

Although various human biometric features have their own advantages, today, with increasing requirements for the accuracy and security of the biometric identification system, it is impossible for each biometric feature to be universal in a true sense. Some inherent limitations in the system affect the accuracy, such as the fingerprint characteristics of certain people or groups of people (such as gymnasts) become less due to wear and tear, and the face recognition system is less sensitive to human appearance characteristics: hair, accessories, aging and Distinguishing twins is difficult. However, these inherent limitations are difficult to use algorithms to improve its accuracy, so it is difficult to meet practical applications only by a single method or a single biometric feature. Therefore, it is becoming a research and development trend to integrate multiple biometric features together for identity identification.

Due to the large amount of calculation of biometrics, usually biometric devices can only recognize one biometric feature. There are few existing multi-modal biometric identification systems. In China, there are face and hand shape recognition from Northern Jiaotong University. Most foreign devices also use fingerprints, voice, hand shape, etc., which are prone to missing features, poor anti-interference, or poor recognition accuracy. The biometrics of advanced problems cannot achieve the desired accuracy. However, the problem of multi-biometric feature fusion is that the amount of calculation is large, and the real-time performance of recognition becomes poor.

(3) Contents of the invention

The technical problem solved by the utility model is to provide a multi-modal high-precision biometric feature recognition device. The idea adopted is to select two biometric features of the palm vein and the face for recognition to meet the high-precision requirements. The multi-processor parallel structure to achieve the purpose of shortening the recognition time.

In order to achieve the above-mentioned purpose, the technical solution adopted by the utility model mainly includes the following modules: a face image acquisition module for obtaining facial biological characteristics; a palm vein image acquisition module for obtaining palm vein characteristics; for driving a touch screen Auxiliary processing modules for voice, display and recognition interfaces; main control unit modules for coordinating and managing module operations. Since the first three modules are mutually independent and parallel to each other, and each module has carried out certain preprocessing on the collected data, the operation speed of the whole system has been greatly accelerated.

First, the user needs to place the palm on the palm vein acquisition module and face the face acquisition module. At this time, the palm vein acquisition module will automatically capture the approaching palm. After confirmation, the binary palm vein image obtained by preprocessing will be sent to The main control unit module, the main control unit will open the face acquisition module, and then the face acquisition module will work, search for the position of the face on the screen, and send the position of the face and the captured facial image to the main control unit module; finally The main control unit recognizes the obtained palm vein features and face features, and sends the recognition results to the auxiliary processing module, and then the auxiliary processing module calls the personnel information stored in the information database and displays them on the touch screen. When registering personnel information, it is convenient and easy to use for human-computer interaction through the prompts on the touch screen.

(4) Description of drawings

Fig. 1 is a system structure diagram of the utility model.

Figure 2 is the light absorption characteristics of hemoglobin.

The marks 1, 2, and 3 in Figure 1 represent ordinary webcams, 4, 5, 6, and 7 represent synchronous dynamic memory, 8, 9, 10, and 11 represent ARM processors, 12 represent touch screens, 13 represent audio interfaces, and 14 represent Network interface, 15 represents flash memory Flash, wherein 1, 4, 8 constitute the face image acquisition module, 2, 5, 9 constitute the palm vein image acquisition module, 3, 6, 10, 12, 13 constitute the auxiliary processing module, 7, 11 , 14, 15 constitute the main control unit module.

Below in conjunction with accompanying drawing, the utility model is described further:

(5) Specific implementation methods

Camera 1 needs to add an 850nm high-pass filter to the lens, and arrange an array of 850nm infrared light emitting tubes around it to obtain face images under near-infrared light; camera 2 needs to add a 760nm narrow-band filter to the lens, A 760nm infrared light emitting tube array is arranged around it to obtain palm vein images under near-infrared light. The principle can be obtained from 2. The difference in light absorption characteristics makes the veins stand out; the camera 3 is used to obtain the background image of the face under normal light without adding a filter. SDRAM4, 5, 6, and 7 are used as the memory unit of the corresponding ARM processor to store some temporary data.

When the user registers information, at first click the registration key of touch screen, the ARM processor 10 in the auxiliary processing module sends a signal to the ARM processor 11 in the main control unit module, then the ARM processor 11 controls the ARM processor 9 to start working, by The camera 2 collects the palm vein image, calls the preprocessing algorithm, performs positioning and refinement operations, obtains a binarized palm vein image, and returns the result to the ARM processor 11. At this time, the ARM processor 11 is stored in the flash memory 15 Create a folder in the middle and store the palm vein information, and control the ARM processor 10 to display the information that the palm vein acquisition is completed on the touch screen 12, prompting to prepare for face acquisition, and then the ARM processor 11 controls the ARM processor 8 to start working, and the ARM The processor 8 obtains the face image under the near-infrared light through the camera 1, and after intercepting the face image, sends the image information to the ARM processor 11, and then the ARM processor 11 stores the image information into the file created during palm vein collection folder, and control the ARM processor 10 to intercept the face image under the normal light source and store it in the folder established by the palm vein collection, and display on the touch screen 12 that the face image collection is completed, prompting to input the name of the registered person and other information , and finally the registration is complete.

When performing identification, the person to be identified first puts the palm on the palm vein collection camera 2, and the ARM processor 9 automatically detects the palm. Processor 11 sends the processed palm vein image, and when ARM processor 11 receives the palm vein image, it will control ARM processor 8 to start running, and ARM processor 8 will call into the face detection algorithm to detect people's faces. Normal then the facial image that intercepts is sent to processor 11, then controls ARM processor 10 by ARM processor 11 and prompts on touch screen 12 and please face the camera when detection is not normal, until a human face is detected, the facial image that is intercepted is sent to processor 11. After the collection of face information is completed, the ARM processor 11 invokes the recognition algorithm, compares the information obtained with the information stored in the information library flash memory 15, and if they match, the ARM processor 10 is controlled to display that the recognition is complete, and simultaneously send an audio signal , if it does not match, the stranger will be prompted, and the alarm signal will be prompted by voice. After the recognition is completed, the recognition result can be sent through the network interface 14.

The main idea of the hardware device used in the utility model is parallel processing, which not only has short recognition time and high precision, but also is easy to use and has strong expansibility, and can be widely used in personnel registration, access control, banks and other occasions.

Claims (4)

1. Palm vein and face dual-mode biometric recognition instrument, composed of USB drive-free camera (1) (2) (3), synchronous dynamic memory SDRAM (4) (5) (6) (7), ARM processor ( 8) (9) (10) (11), touch screen (12), audio interface (13), network interface (14) and flash memory F1ash (15) form, it is characterized in that, ARM processor (8) (9) ( 10) connected to the ARM processor (11), all controlled by the ARM processor (11). 2. palm vein as claimed in claim 1 and people's face dual-mode biometric identification instrument, it is characterized in that: camera (1) is a USB driver-free camera, adds 850nm high-pass filter on the lens, and is arranged around the camera 850nm wavelength ring light source. 3. palm vein as claimed in claim 1 and human face dual-mode biometric identification instrument, it is characterized in that: camera (2) is a USB driver-free camera, adds 760nm narrow-band filter on the lens, and is arranged around the camera Ring light source with 760nm wavelength. 4. The palm vein and human face dual-mode biometric identification device as claimed in claim 1, characterized in that: the camera (3) is a USB drive-free camera, the lens does not add an optical filter, and no additional light source is arranged around it

Images