Deep learning is a subset of machine learning. It abstracts a mathematical model by simulating the way information flows spread between neurons in brain science. It can extract the essential features of data from different levels of abstraction through end-to-end training, so as to complete complex classification or regression tasks. Eye Technology began to explore the application of deep learning in face recognition in 2012, and launched the first generation of face recognition algorithm based on deep learning in 2014.
Technological strength
Automatically learn features from big data, rather than using hand-designed featuresDepth models are more expressive and efficient
01
Technical
It is a biometric identification technology based on facial feature information. Use a camera or camera to collect the image or video stream containing a face, and automatically detect and track the face in the image, and then carry out a series of related technologies for the detected face, usually called portrait recognition, facial recognition.
Technological strength
The largest floor storage exceeds 90 million yuanGigabit data second responseIt has achieved large-scale application in dozens of institutions such as Agricultural Bank of China, Minsheng, Bank of Communications, Everbright, Postal Savings and Qianhai Life Insurance
02
Technical
A fingerprint is an uneven line of skin on the front end of a finger. This grain contains a lot of characteristic information, such as grain shape, pattern area, core point, triangle point and grain number and other general characteristics; Detail features, such as endpoint, breakpoint, bifurcation point, triangle point and core point, are called "feature points". Fingerprint identification relies on general features and detailed features to confirm a person's identity.
Technological strength
The Ministry of Public Security recommended the id card fingerprint application algorithm providerwon the second prize of National Technological Invention Award and the second prize of Beijing Science and Technology AwardFingerprint related products have passed the FBI of the United States, India STQC testing certification
03
Technical
The principle of the finger vein system is to match the pattern of the veins inside the finger. Because the veins are hidden inside the fingers, they are extremely difficult to copy and steal, making them safer than other techniques that use features outside the body. At the same time, because the principle of vein imaging is the system of red blood cells to infrared light, the image of finger vein is obtained through the image sensor, and the algorithm is used for image analysis and in vivo identification detection.
Technological strength
Meet the requirements of GA/T 939-2012 security digital vein technical standardsSupport 1:1 digital vein verification and 1: N digital vein identificationSupport different security levels, 1-5 level adjustable, standard feature format, strong compatibility
04
Technical
The structure of the human eye is made up of the sclera, the iris, and the pupil. The iris is a circular part between the black pupil and the white sclera. It contains many interlacing details, such as spots, filaments, coronal shapes, stripes, and recess, which can identify a person. Iris recognition technology is used to identify people by comparing the similarity of iris image features.
Technological strength
Iris related equipment has passed the Indian STQC certification,Iris image follows international standard ISO/IEC 19794-6The single-core comparison speed can be up to 1.5 million times/second
05
Technical
organization
organization
Beijing R&D Center
Algorithm and Software R&D Center: Located in Beijing, it mainly develops algorithms, SDK and other basic core technologies.
Shenzhen R&D Center
Intelligent hardware R & D Center: located in Shenzhen, relying on the technical advantages and entrepreneurial atmosphere of Shenzhen Software Park, the center conducts rapid design and implementation of intelligent hardware products.
Jinan R&D Center
Software and Implementation R&D Center: Jinan, based on ABIS biometric unified platform, with implementation delivery and application software development as the core.
Shenyang R&D Center
Algorithm RESEARCH and development Center: Shenyang, based on deep learning, with algorithm research and development and extended exploration as the core.