A typical fingerprint identification system should include: fingerprint identification sensor feature extraction/matching module feature template library application software. Fingerprint matching can be divided into two steps. The first step is to extract the characteristics of the fingerprint to be verified, and then compare it with the template fingerprint in the fingerprint template library to determine whether the two fingerprint images come from the same finger.
We can see that fingerprint analysis collected by fingerprint identification core is accurate and efficient. The development of fingerprint identification and collection technology can be roughly divided into three ways: optical identification, capacitive sensor and biological radio frequency.
1, optical recognition
Optical identification is an early fingerprint identification technology. Based on the light emitted by the light emitter, it hits the finger and then reflects back to the machine to obtain data, and compares it with the database to see if it is consistent. Optical recognition can only reach the epidermis of the skin, but not the dermis, and it is greatly influenced by whether the finger surface is clean or not.
2. Capacitance sensor
Capacitance sensor identification is to detect fingerprint data by turning on or off the current of two capacitors installed at a certain interval when fingers slide over the fingerprint detection instrument. Capacitive sensors require high cleanliness of fingers, and the sensor surface is made of silicon, which is easy to be damaged. From a technical point of view, the suppliers of capacitive fingerprint identification technology include Authentec, Validity, FingerPrintCardsAB(FPC) and so on. Authentec was acquired by Apple, and Validity was also acquired by Synaptics. Capacitive fingerprint sensor is also the most widely used technology at present.
3, biological radio frequency
The RF sensor emits a small amount of RF signals through the sensor, penetrates the finger epidermis, and obtains the lines of the inner layer, thus obtaining information. This method requires less cleanliness of fingers. Radio frequency is a relatively new technical scheme at present, and it is also a kind of capacitance mode, but it is limited by patent issues. Radio frequency is the future development direction.
[Page]
Fingerprint identification acquisition mode
No matter what acquisition technology is adopted, there are two input methods from the user's point of view: pressing and sliding.
1, sliding
When you swipe your finger across the sensor, the system can get the fingerprint of the whole finger. If you press your finger up, you can't collect a complete image at one time. When collecting, you need to slide your finger on the collection surface, take snapshots of each fingerprint image collected when your finger slides, and then stitch these snapshots together to form a complete fingerprint image.
Sliding mode has the advantages of low cost, easy integration, large-area image acquisition and application of traditional feature point algorithm. However, the disadvantage is that customers need a consistent and standardized action to collect images, and the experience effect is poor, which was not very successful in the previous application and promotion.
2. By type
Put your finger flat on the device to get a fingerprint image. Generally, in order to get the fingerprint of the whole finger, it is necessary to use a sensor larger than the finger and press the whole finger on the sensor at the same time.