The main application directions of face recognition are face control and face verification (ie, person-in-one).
Face control refers to obtaining the face in the video through the device and comparing it with the control library (blacklist). When the similarity is higher than a certain value, the system prompts and records. The main application areas are the national security level, and the scene of the city's public security arrest.
Face verification is a 1:1 comparison of a face or ID photo with a live video; or a 1:N screening comparison from a self-built face database. The main application areas are public safety, information security, and financial security. For example: conference delegate identification, confidentiality control, user authentication, etc.
Telpo TPS980 have cooperated with Baidu face recognition system, which belongs to offline face recognition. The strict algorithm control technology perfectly solves the problems of light, low light and backlight, and motion blur, which can realize instant capture in various environments.
High-quality faces and high-precision comparisons.
Similarly, the whole process of comparison operation is completed in the local server, and the closed-loop management of the self-built face database is truly realized without passing through the external network and generating no face matching cost.
Applicable scene recommendation:
1. Airport Customs Border Inspection: When the personnel are inspected, they should first swipe the ID card and then take the photo of the site to verify the identity of the person. This is a significant improvement in the accuracy and speed of the human card comparison. And the photos captured on the spot are archived and later checked.
2. Train station control: pure face recognition (no identification part). Identify the face in the tracking camera video and compare it with the library (blacklist). When the similarity is higher than a certain value, the system prompts and records. This kind of dynamic recognition can obtain better quality pictures due to lighting environment guarantee, travel path limitation and the like. Most of the fugitives caught by face recognition are in this way.
3. Bank withdrawal: If face recognition is to ensure the security of the bank account and does not need to memorize various passwords, then face recognition and live detection can be performed simultaneously by swiping the card, verifying the password or fingerprint, and verifying. Then you can increase the security without increasing the user's withdrawal time and operation steps.
4. Conference security: face recognition is to ensure that the personnel entering the venue are consistent with the documents held, so that avoid counterfeit documents, displace, etc. are eliminated, then the conference card can be used first (verification is true and valid),through Face recognition (Catch the scene portrait and the photo stored in the chip of the conference card is the same person) and then the person is allowed to enter. It should be noted here that the “conference certificate” issued by the active party is not the “identity card”, which can effectively manage the actual participants. The passage gates here usually adopt the barrier-free passage mode, and the certificate verification + face verification can be completed during the progress of the participants, without waiting for the queue and improving the conference effectiveness.
5. Control zone security: The traffic of the control zone is usually for internal personnel, such as the airport control zone. Different people have their own access rights and need to abide by strict rules and regulations. You can use the first pass pass to verify the validity period and access rights, and then confirm the fingerprint or face recognition as the same person. The barrier-free access + face recognition terminal + gate channel ensures the integrity of the person in the journey, while ensuring strict access control. If the comparison is successful, the gate will automatically open, and if the comparison fails, the sound and light alarm will be given.
6. Unit card: The employee card issued by the unit often has multiple functions: attendance, access control, consumption, etc. Joining face recognition technology will effectively reduce the occurrence of fraudulent use of cards. Swipe the card first, then verify the face, and make sure the card holder is the ID card.
7. Candidate identification: If face recognition is to ensure that the candidate is the person, to eliminate the "gunner" behavior, then you can first swipe the ID card or admission ticket, and then fingerprint or face recognition to ensure that the candidate is the person. It is possible to increase the accuracy and speed of comparison without increasing the cost.
8. Verification of identity verification: When making a real name card, in order to verify the identity of the person, you can first swipe the ID card, and then take a photo on the spot and compare it with the ID card photo. This will increase accuracy and speed, while also allowing live photo archives for later review.
Create more face recognition scenes with Telpo TPS980.
Click to know more about Telpo face detection
Or Google search Telpo
