CN-116631032-B - Human body induction detection method, human body induction detection device and human face recognition device

CN116631032BCN 116631032 BCN116631032 BCN 116631032BCN-116631032-B

Abstract

The application provides a human body induction detection method, a human body induction detection device and a human face recognition device. The method comprises the steps of respectively determining whether face information of target objects in a first image and a second image is complete based on difference values of coordinate values of a plurality of target feature points in the first image and coordinate values of the target feature points in the second image, determining a ratio of the area of a first area in the first image to the area of a second area in the second image under the condition that the face information in the first image and the face information in the second image are complete, and determining that the target objects in a moving state exist in the target area under the condition that the ratio is within a preset range, wherein the target area is an area reachable by a binocular camera, so that the problem that the error recognition rate of a human body induction detection method in the prior art is high is solved.

Inventors

YIN XIANGPENG
CHEN YANYU
DU YANG
ZHU PENGFEI
YU YIJUN
Xie Yuechen

Assignees

珠海格力电器股份有限公司
珠海联云科技有限公司

Dates

Publication Date: 20260505
Application Date: 20230526

Claims (9)

1. A human body sensing detection method, comprising: Determining whether face information of a target object in a first image and a second image is complete or not based on difference values of coordinate values of a plurality of target feature points in the first image and coordinate values in the second image, wherein shooting time of the second image is later than that of the first image, and the first image and the second image are shot by an infrared camera in a binocular camera; Under the condition that face information in the first image and face information in the second image are complete, determining a ratio of an area of a first area in the first image to an area of a second area in the second image, wherein the first area is obtained by sequentially connecting a plurality of target feature points in the first image, and the second area is obtained by sequentially connecting a plurality of target feature points in the second image; Under the condition that the ratio is in a preset range, determining that the target object in a moving state exists in a target area, wherein the target area is an area reachable by the binocular camera, The plurality of target feature points include a first feature point, a second feature point and a third feature point, the coordinate values include an abscissa value and an ordinate value, and determining whether face information of a target object in a first image and a second image is complete based on differences between coordinate values of the plurality of target feature points in the first image and coordinate values in the second image, respectively, includes: Determining the absolute value of the difference value between the abscissa value of the first feature point in the first image and the abscissa value of the second image to obtain a first abscissa difference value, determining the absolute value of the difference value between the abscissa value of the second feature point in the first image and the abscissa value of the second image to obtain a second abscissa difference value, and determining the absolute value of the difference value between the abscissa value of the third feature point in the first image and the abscissa value of the second image to obtain a third abscissa difference value; Determining the absolute value of the difference value between the ordinate value of the first feature point in the first image and the ordinate value of the second feature point in the second image to obtain a first ordinate difference value, determining the absolute value of the difference value between the ordinate value of the second feature point in the first image and the ordinate value of the second feature point in the second image to obtain a second ordinate difference value, and determining the absolute value of the difference value between the ordinate value of the third feature point in the first image and the ordinate value of the third feature point in the second image to obtain a third ordinate difference value; And based on the first horizontal coordinate difference value, the second horizontal coordinate difference value and the third horizontal coordinate difference value, determining whether the face information of the target object in the first image and the second image is complete or not respectively.
2. The human body sensing method according to claim 1, wherein determining whether face information of the target object in the first image and the second image is complete based on the first horizontal coordinate difference value, the second horizontal coordinate difference value, and the third horizontal coordinate difference value, respectively, includes: And determining that face information in the first image and the second image is complete under the condition that the first abscissa difference value is smaller than a first abscissa reference value, the second abscissa difference value is smaller than a second abscissa reference value, the third abscissa difference value is smaller than a third abscissa reference value, the first ordinate difference value is smaller than a first ordinate reference value, and the second ordinate difference value is smaller than a second ordinate reference value and the third ordinate difference value is smaller than a third ordinate reference value.
3. The method of claim 1, wherein the plurality of target feature points includes a first feature point, a second feature point, and a third feature point, the first region and the second region are each triangular-shaped regions, Determining a ratio of an area of a first region in the first image to an area of a second region in the second image, comprising: Determining an absolute value of the area of the first region in the first image by adopting a triangle area calculation formula to obtain a first area; determining an absolute value of the area of the second region in the second image by adopting an area calculation formula of the triangle to obtain a second area; And determining the ratio of the second area to the first area to obtain the ratio.
4. The human body sensing detection method according to any one of claims 1 to 3, wherein after determining that the target object in a moving state exists in a target area, the human body sensing detection method further comprises: And starting a visible light camera in the binocular camera to face the target object according to the real-time color image shot by the visible light camera.
5. A human body sensing detection method according to any one of claims 1 to 3, further comprising: and under the condition that the ratio is not in the preset range, the visible light cameras in the binocular camera are kept not to be started.
6. The human body sensing detection method according to any one of claims 1 to 3, wherein before determining whether face information of a target object in a first image and a second image is complete based on differences in coordinate values of a plurality of target feature points in the first image and coordinate values in the second image, respectively, the human body sensing detection method further comprises: And acquiring a real-time target image shot by the infrared light camera, and detecting a human body based on the real-time target image to determine whether the target object exists in the target area.
7. A human body sensing and detecting device, comprising: A first determining unit, configured to determine whether face information of a target object in a first image and a second image is complete based on differences between coordinate values of a plurality of target feature points in the first image and coordinate values in the second image, where a shooting time of the second image is later than a shooting time of the first image, and the first image and the second image are both shot by an infrared camera in a binocular camera; The second determining unit is used for determining the ratio of the area of a first area in the first image to the area of a second area in the second image under the condition that the face information in the first image and the face information in the second image are complete, wherein the first area is obtained by sequentially connecting a plurality of target feature points in the first image, and the second area is obtained by sequentially connecting a plurality of target feature points in the second image; A third determining unit, configured to determine that, when the ratio is within a preset range, the target object in a moving state exists in a target area, where the target area is an area reachable by the binocular camera, The plurality of target feature points comprise a first feature point, a second feature point and a third feature point, the coordinate values comprise an abscissa value and an ordinate value, the first determining unit comprises a first determining module, a second determining module and a third determining module, wherein the first determining module is used for determining the absolute value of the difference value of the abscissa value of the first feature point in the first image and the abscissa value in the second image to obtain a first abscissa difference value, determining the absolute value of the difference value of the abscissa value of the second feature point in the first image and the abscissa value in the second image to obtain a second abscissa difference value, and determining the absolute value of the difference value of the abscissa value of the third feature point in the first image and the abscissa value in the second image to obtain a third abscissa difference value; The second determining module is configured to determine an absolute value of a difference between an ordinate value of the first feature point in the first image and an ordinate value of the second feature point in the first image, obtain a first ordinate difference value, determine an absolute value of a difference between an ordinate value of the second feature point in the first image and an ordinate value of the second feature point in the second image, obtain a second ordinate difference value, and determine an absolute value of a difference between an ordinate value of the third feature point in the first image and an ordinate value of the third feature point in the second image, so as to obtain a third ordinate difference value; The third determining module is configured to determine whether face information of the target object in the first image and the second image is complete based on the first horizontal coordinate difference value, the second horizontal coordinate difference value, and the third horizontal coordinate difference value, and the first vertical coordinate difference value, the second vertical coordinate difference value, and the third vertical coordinate difference value, respectively.
8. A computer-readable storage medium, characterized in that the computer-readable storage medium comprises a stored program, wherein the program, when run, controls a device in which the computer-readable storage medium is located to perform the human body induction detection method according to any one of claims 1 to 6.
9. A face recognition device, comprising: a human body sensing detection device for performing the human body sensing detection method of any one of claims 1 to 6; the binocular camera is used for collecting images in a target area, and the target area is an area which can be reached by the binocular camera.

Description

Human body induction detection method, human body induction detection device and human face recognition device Technical Field The application relates to the field of human body induction detection, in particular to a human body induction detection method, a human body induction detection device, a computer readable storage medium and a human face recognition device. Background With the wide application of face recognition technology, the human body detection application scenes are more and more. For example, daily attendance data processing, access control systems, self-study rooms, library inductive switch lights, and the like. The existing human body sensing technology generally adopts a passive human body detection technology when human body sensing detection is carried out. For example, pyroelectric sensors are provided in face recognition devices. For the human body induction detection method adopting the pyroelectric sensor, the specific principle is that the pyroelectric sensor emits infrared rays to detect the existence and movement of a human body. However, such human body sensing detection methods are susceptible to various heat sources, especially when the ambient temperature is close to the human body temperature, the sensitivity of the pyroelectric sensor is reduced, so that the false touch rate and the false recognition rate of the pyroelectric sensor are high. Therefore, a method for performing human body sensing detection more accurately is needed. Disclosure of Invention The application mainly aims to provide a human body induction detection method, a human body induction detection device, a computer readable storage medium and a human face recognition device, so as to at least solve the problem of high false recognition rate of the human body induction detection method in the prior art. In order to achieve the above object, according to one aspect of the present application, there is provided a human body sensing detection method, including determining whether face information of a target object in a first image and a second image is complete based on difference values of coordinate values of a plurality of target feature points in the first image and coordinate values in the second image, respectively, wherein a shooting time of the second image is later than a shooting time of the first image, the first image and the second image are both infrared light cameras in a binocular camera, determining a ratio of an area of a first region in the first image to an area of a second region in the second image in a case that the face information in the first image and the second image is complete, the first region being a sequentially connected by a plurality of target feature points in the first image, the second region being a sequentially connected by a plurality of target feature points in the second image, and determining that the target object is in a moving state in a preset range. Optionally, the plurality of target feature points include a first feature point, a second feature point and a third feature point, the coordinate values include an abscissa value and an ordinate value, the difference between the coordinate values of the plurality of target feature points in the first image and the coordinate values in the second image is determined based on the difference between the coordinate values of the plurality of target feature points in the first image and the coordinate values of the plurality of target feature points in the second image, the method includes determining whether face information of the target object in the first image and the second image is complete, determining an absolute value of a difference between the abscissa value of the first feature point in the first image and the abscissa value in the second image, obtaining a first abscissa value, determining an absolute value of a difference between the abscissa value of the second feature point in the first image and the coordinate values in the second image, obtaining a second abscissa value, determining an absolute value of the third feature point in the first image and the coordinate values in the second image, obtaining a third coordinate value based on the absolute value of the difference between the abscissa value of the first feature point in the first image and the coordinate value in the second image, determining an absolute value of the difference between the first feature point in the first image and the coordinate value in the second image, and determining whether the face information of the target object in the first image and the second image is complete or not respectively. Optionally, determining whether the face information of the target object in the first image and the second image is complete based on the first abscissa difference value, the second abscissa difference value and the third abscissa difference value, wherein the first ordinate difference value, the second ordinate difference value and th