Generalized face recognition actually includes a series of related technologies to build a face recognition system, including face image acquisition, face location, face recognition preprocessing, identity confirmation and identity search, etc. The narrow sense of face recognition refers to the technology or system for identity confirmation or identity search through the face.

Face recognition technology is also an image analysis technology, the factors affecting the accuracy of face recognition technology include light, posture, expression and other factors, to make face recognition truly productized, light is the first to solve a major problem. Near-infrared light imaging has the advantages of strong anti-interference, anti-counterfeiting, anti-fraud, etc., which makes near-outside image face recognition can make up for the defects and deficiencies of other optical face recognition technology to a large extent, near-infrared face recognition algorithm is to solve the lighting problem in face recognition, and proposed a solution, which includes two parts: Active near infrared face imaging equipment and corresponding illumination independent face recognition algorithm. Using the active near-infrared light source with a higher intensity than the ambient light, with the optical filter of the corresponding band, an environment-independent face image can be obtained, and the face image will only change monotonically with the change of the distance between the person and the camera. By using some special feature extraction methods, the monotony of the image can be further eliminated and the feature expression of the image can be completely light independent.

Face recognition operation process:

Face detection and positioning: Detect the face and binocular face recognition feature comparison in the current video frame or picture through the positioning of the camera of the device: obtain the face contour through the camera head of the device, extract the feature to generate the feature database, compare the real-time generated feature data with the preset feature database, calculate the similarity, and obtain the corresponding recognition result.

Application: This technology is mainly used in products that require the active cooperation of users and short-distance accurate identification, and is widely used in attendance access control industry, building intercom smart home industry, luggage lock industry and so on.