Search

CN-116721452-B - Face recognition system, method, device, electronic equipment and program product

CN116721452BCN 116721452 BCN116721452 BCN 116721452BCN-116721452-B

Abstract

The disclosure belongs to the technical field of face recognition, and in particular relates to a face recognition system, a face recognition method, a face recognition device, electronic equipment and a computer program product. The face recognition method comprises the steps of obtaining the similarity between each face sample in a base and a face to be recognized based on each face recognition algorithm in a plurality of face recognition algorithms, wherein the base comprises a plurality of face samples preset for each person, calculating the first similarity sum of each person in the base corresponding to the face recognition algorithm, obtaining the weight of the face recognition algorithm, and determining the person corresponding to the face to be recognized based on the weight of the face recognition algorithm and the first similarity sum of each person. The method and the device can reduce the false recognition rate of face recognition on the face data acquired in different application scenes.

Inventors

  • DENG YAONING

Assignees

  • 光控特斯联(上海)信息科技有限公司

Dates

Publication Date
20260512
Application Date
20230613

Claims (7)

  1. 1. A face recognition system comprising a base and an identification module, wherein: the bottom library comprises a plurality of face samples with preset numbers for each person; The recognition module comprises a plurality of face recognition algorithms; The face recognition system acquires the similarity between each face sample in the base and the face to be recognized by using each face recognition algorithm, and determines the person corresponding to the face to be recognized in the base based on the similarity and the weight of each face recognition algorithm; The step of determining the person corresponding to the face to be identified in the base based on the similarity and the weight of each face recognition algorithm comprises the steps of calculating a first similarity sum of each person in the base corresponding to the face recognition algorithm based on the similarity, calculating a second similarity sum of each person based on the weight of the face recognition algorithm and the first similarity sum of each person, and the person with the highest second similarity sum is the person corresponding to the face to be identified.
  2. 2. The system according to claim 1, wherein: Different face samples of the same person are obtained by the person in different preset scenes, wherein one face sample corresponds to one preset scene.
  3. 3. A face recognition method applied to the face recognition system of claim 1, comprising: The method comprises the steps of respectively obtaining the similarity between each face sample in a base and a face to be recognized based on each face recognition algorithm in a plurality of face recognition algorithms, wherein the base comprises a plurality of face samples with preset numbers for each person; Calculating a first similarity sum of each person in the bottom library corresponding to the face recognition algorithm; acquiring the weight of the face recognition algorithm, and determining the person corresponding to the face to be recognized based on the weight of the face recognition algorithm and the first similarity sum of each person; The step of calculating the first similarity sum of each person in the base corresponding to the face recognition algorithm comprises the steps of adding the similarity of each face sample of the person in the base obtained by the face recognition algorithm to obtain the first similarity sum of the person corresponding to the face recognition algorithm; the step of determining the person corresponding to the face to be identified based on the weight of the face recognition algorithm and the first similarity sum of each person comprises the step of adding the product of the weight of each face recognition algorithm and the first similarity sum of the person corresponding to the face recognition algorithm to obtain a second similarity sum of the person, wherein the person with the highest second similarity sum is the person corresponding to the face to be identified.
  4. 4. A method according to claim 3, wherein said determining a person corresponding to the face to be identified based on the weight of the face recognition algorithm and the first similarity sum of each of the persons comprises: Determining the first N persons before the first similarity sum corresponding to each face recognition algorithm, wherein N is a natural number; Forming a candidate person set by the N persons in front of the first similarity sum corresponding to each face recognition algorithm; And for each person in the candidate person set, adding the product of the weight of each face recognition algorithm and the first similarity sum of the persons corresponding to the face recognition algorithm to obtain a second similarity sum of the persons, wherein the person with the highest second similarity sum is the person corresponding to the face to be recognized.
  5. 5. A face recognition apparatus applied to the face recognition system of claim 1, comprising: The acquisition module is used for acquiring the similarity between each face sample in the base and the face to be identified based on each face recognition algorithm in a plurality of face recognition algorithms, wherein the base comprises a plurality of face samples with preset numbers for each person; the computing module is used for computing a first similarity sum of each person in the bottom library corresponding to the face recognition algorithm; The determining module is used for acquiring the weight of the face recognition algorithm and determining the person corresponding to the face to be recognized based on the weight of the face recognition algorithm and the first similarity sum of each person; the computing module is specifically configured to add the similarity of each face sample of the person in the base obtained by using the face recognition algorithm to obtain a first similarity sum of the person corresponding to the face recognition algorithm; The determining module is specifically configured to add the product of the weight of each face recognition algorithm and the first similarity sum of the people corresponding to the face recognition algorithm to obtain a second similarity sum of the people, where the person with the highest second similarity sum is the person corresponding to the face to be recognized.
  6. 6. An electronic device includes a memory and a processor, The memory is used for storing a computer program; the processor being adapted to implement the method of claim 3 or 4 when executing the computer program.
  7. 7. A computer program product comprising a computer program, instructions which, when executed by a processor, implement the method of claim 3 or 4.

Description

Face recognition system, method, device, electronic equipment and program product Technical Field The disclosure belongs to the technical field of face recognition, and in particular relates to a face recognition system, a face recognition method, a face recognition device, electronic equipment and a computer program product. Background Face recognition is a biological recognition technology for acquiring facial feature information of a person and comparing the feature information with face data pre-recorded in a base, so as to perform personal identification. The probability of recognizing the nail in the bottom library based on the facial feature information of the nail in the face recognition algorithm is called recognition rate, and the probability of recognizing other people not in the bottom library based on the facial feature information of the nail is called false recognition rate. The face recognition algorithm can only determine the degree of similarity of two faces, and judge whether the two faces are the same person or not, and also needs to judge from a service level. But different face recognition algorithms are trained based on different business models of different application scenes, and the face data pre-recorded in the base is often acquired based on a specific application scene. Therefore, each face recognition algorithm has high false recognition rate of recognizing faces acquired from other application scenes. Disclosure of Invention The embodiment of the disclosure provides a face recognition scheme to solve the problem that the face recognition error rate of the existing face recognition scheme for face data from different application scenes is high. A first aspect of an embodiment of the present disclosure provides a face recognition system, including a base and an identification module, wherein: the bottom library comprises a plurality of face samples with preset numbers for each person; The recognition module comprises a plurality of face recognition algorithms; And the face recognition system acquires the similarity between each face sample in the base and the face to be recognized by using each face recognition algorithm, and determines the person corresponding to the face to be recognized in the base based on the similarity and the weight of each face recognition algorithm. In some embodiments, different face samples of the same person are obtained by the person in different preset scenes, wherein one face sample corresponds to one preset scene. In some embodiments, the determining, based on the similarity and the weight of each face recognition algorithm, the person in the base corresponding to the face to be recognized includes: Calculating a first similarity sum of each person in the bottom library corresponding to the face recognition algorithm based on the similarity; Calculating a second similarity sum for each person based on the weight of the face recognition algorithm and the first similarity sum for each person; And the person with the highest second similarity sum is the person corresponding to the face to be identified. A second aspect of an embodiment of the present disclosure provides a face recognition method, which is applied to the face recognition system described in the first aspect of the present disclosure, including: The method comprises the steps of respectively obtaining the similarity between each face sample in a base and a face to be recognized based on each face recognition algorithm in a plurality of face recognition algorithms, wherein the base comprises a plurality of face samples with preset numbers for each person; Calculating a first similarity sum of each person in the bottom library corresponding to the face recognition algorithm; and acquiring the weight of the face recognition algorithm, and determining the person corresponding to the face to be recognized based on the weight of the face recognition algorithm and the first similarity sum of each person. In some embodiments, said computing a first similarity sum for each of said people in said base corresponding to said face recognition algorithm comprises: And adding the similarity of each face sample of the person in the base acquired by the face recognition algorithm to obtain a first similarity sum of the person corresponding to the face recognition algorithm. In some embodiments, the determining a person corresponding to the face to be identified based on the weight of the face recognition algorithm and the first similarity sum of each person includes: Adding the product of the weight of each face recognition algorithm and the first similarity sum of the personnel corresponding to the face recognition algorithm to obtain a second similarity sum of the personnel; And the person with the highest second similarity sum is the person corresponding to the face to be identified. In some embodiments, the determining a person corresponding to the face to be identified based on the weight of the