Documentation for Axxon One 2.0. Documentation for other versions of Axxon One is [available too].

Previous page 얼굴 검출 도구  얼굴 검출 도구의 요구 사항 Next page

The Face detection basic object and the Face detection (VL) object are triggered every time a face is captured in the frame. Basic object is enough to perform a search for faces in the archive (see 유사한 얼굴 검색).

In addition, the following types of detection tools based on metadata from the face detection (see 메타데이터에 대한 일반 정보) are available in Axxon One:

  1. Appearance in area – a detection tool is triggered by the appearance of an object and subsequent face capture in FOV.
  2. Loitering in area – a detection tool triggered by the lengthy presence of an object and its face capture in FOV.

  3. Mask detection – a detection tool is triggered by the face captured with or without a mask.

The Axxon One database stores all faces in binary form:

  1. All captured face images are vectorized* and stored in the t_face_vector table, and their corresponding capture events are stored in the t_json_event table.
  2. Reference images (see Faces) are stored in the t_face_listed table.

* Face vector is the mathematical representation of a face image created upon face capture.

Note

These detection tools require Add-on Face Recognition Pack to be installed (see DetectorPack 애드온 설치).

Attention!

With an increase in the number of faces in the database, the statistical error increases: the more faces in the database, the more often similar faces will be recognized when searching in the archive. Accordingly, the degree of similarity when comparing the reference face with the captured face will decrease.

This statistical error is relevant if:

  1. The 얼굴 검출 도구의 요구 사항 are met.
  2. The database contains over a million faces.

An example of the error calculation results:

  1. Face detection, mugshot dataset (good quality photo), 12 million faces in database, and false matching probability is 0.003%. With these initial data, the researchers obtained an identification error of 0.76%.
  2. Face detection (VL), the initial data are the same. The identification error is 0.81%.
  • No labels