Finger vein recognition technology architecture
1. Medical basis
Finger vein is a new biometric recognition technology, which originates from the advanced development project of human brain function activity management in the field of medical science and technology. In this development, near-infrared ray is used to observe the increase of blood flow. When the near-infrared ray passes through human tissues, the hemoglobin in vein blood vessels has obvious absorption effect on near-infrared ray, so as to make venous blood The tube distribution features are characterized by different gray values in the image. Because of the randomness of the elongation direction, the distribution of finger vein vessels presents different structures. Therefore, it can be determined whether the two vein images come from the same finger. At the same time, it is found that this technology is also suitable for finger vein image acquisition.
2. Technical architecture
Finger vein recognition uses light propagation technology to compare and identify finger vein. When the near-infrared ray passes through a human finger, part of the ray will be absorbed by the hemoglobin in the blood vessel, thus capturing the unique finger vein pattern, and then comparing it with the pre registered finger vein pattern to identify the individual. The technical framework of finger vein recognition is shown in Figure 2. It has been proved by medical practice that the shape of finger vein is unique and stable, that is, each person’s finger vein image is different, and the vein image of different fingers of the same person is also different. The shape of veins in healthy adults is no longer changed, which provides medical basis for digital veins. Therefore, the recognition technology based on the principle of finger vein image may become a reality. So we can use the image of vein distribution in the finger to identify.
3. Finger vein system identification process
In vein scanning, LED emits near-infrared light on one side of the finger, transmits the finger, and on the other side of the finger, the camera takes the vein image. From the extraction to the finger vein image, this process is mainly to capture the finger vein image, and then extract the finger vein feature information, and compared with the previously registered finger vein feature, so as to confirm the identity of the user. In order to compare the scanned pattern with the original registered pattern, the relevant identification results can be obtained (that is, the relevant identification result can be obtained by comparing the scanned pattern with the original registered pattern).
The finger vein recognition algorithm uses the texture tracking method to extract the line structure information in the finger vein, and carries out individual identification by matching the line structure. It can run on PC environment and embedded DSP platform.
Outstanding characteristics of finger vein recognition technology
Compared with other biometric technologies, the finger vein recognition technology has the following major advantages.
·Because the finger vein is hidden in the body, there is no risk of imitation or theft, and the skin condition on the surface of human hand will not affect the recognition work;
·Non invasive and non-contact imaging technology can ensure the convenience and cleanliness of users. Compared with some biometric technology, finger vein recognition technology is non-contact, and it will be more hygienic in public places;
·Because of the relative stability of finger vein shape and the clarity of captured image, the image data captured by low resolution camera can be processed by small simple data image technology;
·High security. Because the vein is hidden in the finger, it is very difficult to copy and steal. Compared with other authentication technologies based on human external characteristics, this method is safer. At the same time, the vein authentication can sense the blood flow and blood pressure in the finger, and can detect in vivo at the same time.
·High accuracy. The sample is in the human body, so the interference from the outside is very small in the matching process. Put your finger gently to trigger high accuracy recognition. According to strict medical proof and mathematical statistics, FRR (false rejection rate) was 0.01%, far (false recognition rate) was 0.0001%, and FTE (registration failure rate) was 0%.
Therefore, compared with other biometric identification technologies, finger vein identification technology is much more difficult to copy and steal than face shape, palm shape, fingerprint and iris; its recognition accuracy and portability are much higher than those of face shape, palm shape, fingerprint and iris. Especially compared with fingerprint technology, it is more convenient, hygienic and more humanized. Finger vein identification can effectively avoid the shortcomings of traditional access control system, and open up a new way to protect personal and property safety for environmental protection and energy-saving biometric identification technology and products.
Editor in charge: CT