CN-116704621-B - In-vehicle living body prompting method, device, equipment and storage medium

CN116704621BCN 116704621 BCN116704621 BCN 116704621BCN-116704621-B

Abstract

The application provides a method, a device, equipment and a storage medium for prompting living bodies in a vehicle, and belongs to the technical field of intelligent automobiles. The method comprises the steps of responding to the start of a detection flow, continuously obtaining video images and audio data in a vehicle, determining a living body characteristic value according to images of frames in the video images, determining a volume parameter according to the audio data, calculating a comprehensive parameter according to the living body characteristic value and the volume parameter, controlling a loudspeaker to emit preset sound if the comprehensive parameter is larger than a first preset value, executing the step of determining the living body characteristic value again to obtain a new living body characteristic value, and sending corresponding living body prompt information to a user terminal according to the new living body characteristic value and the comprehensive parameter. The method solves the problem of low accuracy of in-vehicle living body judgment.

Inventors

LI YANG
CHU MINGYANG
DU SIJUN
YU SHUOJUN
ZHU YANG

Assignees

浙江吉利控股集团有限公司
吉利汽车研究院（宁波）有限公司

Dates

Publication Date: 20260512
Application Date: 20230518

Claims (12)

1. An in-vehicle living body prompting method, characterized by comprising the following steps: Responding to the start of the detection flow, and continuously acquiring video images and audio data in the vehicle; intercepting a reflection image of a preset area in an image of each frame in the video image, wherein the preset area is an area where a reflective object in each frame is located; Acquiring the pixel number of the preset area and the brightness of each frame; Determining a living body parameter corresponding to a target frame according to a reflected image of the target frame, a reflected image of an adjacent frame and the pixel number of the preset area, wherein the target frame is any frame of all frames, and the adjacent frame is a frame adjacent to the target frame; Inputting the reflected image of the target frame into a preset living body classification model to obtain a target confidence coefficient; determining a brightness parameter of the target frame according to the brightness of the target frame and the brightness of the adjacent frames; Calculating a living body characteristic value according to the living body parameter, the brightness parameter and the target confidence coefficient of the target frame; determining a volume parameter according to the audio data; calculating comprehensive parameters according to the living body characteristic values and the volume parameters; If the comprehensive parameter is larger than a first preset value, controlling the loudspeaker to emit preset sound, and executing the step of determining the living body characteristic value again to obtain a new living body characteristic value; and sending corresponding living body prompt information to the user terminal according to the new living body characteristic value and the comprehensive parameter.
2. The method according to claim 1, wherein determining the living body parameter corresponding to the target frame according to the reflected image of the target frame, the reflected image of the adjacent frame, and the pixel number of the preset area includes: extracting picture characteristics of the reflected image in each frame by adopting a preset model; dividing the picture characteristics of each frame by the pixel number to obtain unit picture characteristics of each frame; the unit picture characteristics of the target frame and the unit picture characteristics of the adjacent frames are subjected to difference to obtain unit characteristic differences; And if the unit characteristic difference is larger than a preset difference value, setting the living parameter corresponding to the target frame to be 1, otherwise, setting the living parameter corresponding to the target frame to be 0.
3. The method of claim 1, wherein determining the luminance parameter of the target frame based on the luminance of the target frame and the luminance of the neighboring frame comprises: adopting the difference between the brightness of the target frame and the brightness of the adjacent frame to obtain a brightness difference value; if the brightness difference value is larger than a preset brightness difference threshold value, determining the brightness parameter of the target frame as 0, otherwise, determining the brightness parameter of the target frame as 1.
4. The method of claim 1, wherein the calculating the living body characteristic value from the living body parameter, the luminance parameter, and the target confidence level of the target frame comprises: inputting the living body parameter, the brightness parameter and the target confidence coefficient of the target frame into the following formula to obtain a living body characteristic value: In the formula, The living body characteristic value is represented by a value, The brightness parameter is represented by a value of the brightness parameter, The parameter of the living body is represented by, Representing the confidence level of the object in question, And Is a preset weight coefficient.
5. The method according to any one of claims 1 to 4, wherein the sending, to the user terminal, the corresponding post-arranged living body prompt message according to the new living body feature value and the comprehensive parameter includes: if the new living body characteristic value is larger than or equal to a preset living body threshold value, sending a preset high-reliability living body prompt message to the user terminal; If the new living body characteristic value is smaller than a preset living body threshold value and the comprehensive parameter is larger than or equal to a second preset value, sending a preset medium reliability living body prompt message to the user terminal, wherein the second preset value is larger than the first preset value; And if the new living body characteristic value is smaller than a preset living body threshold value and the comprehensive parameter is smaller than the second preset value, sending a preset low-reliability living body prompt message to the user terminal.
6. The method according to any one of claims 1 to 4, wherein said determining a volume parameter from said audio data comprises: extracting volume characteristics of the audio data; if the volume characteristic is larger than a preset volume threshold, the volume parameter is determined to be 1, otherwise, the volume parameter is determined to be 0.
7. The method of any one of claims 1 to 4, further comprising, prior to said continuing to acquire video images and audio data within the vehicle in response to initiation of the detection process: acquiring an in-vehicle image in response to the vehicle locking the door; Intercepting a reflection image of a preset area in the in-car image; Extracting actual features in the reflected image; and calculating target distances between the actual feature and a plurality of prestored existing features, and starting a detection flow if each target distance is greater than a preset distance threshold.
8. The method according to any one of claims 1 to 4, wherein calculating a composite parameter from the living body characteristic value and the volume parameter includes: Inputting the living body characteristic value and the volume parameter into the following formula to obtain a comprehensive parameter: In the formula, The parameter of the combination of the parameters is indicated, The living body characteristic value is represented by a value, Which is indicative of the volume parameter in question, And Indicating a preset constant.
9. An in-vehicle living body presenting device, comprising: the data acquisition module is used for responding to the start of the detection flow and continuously acquiring video images and audio data in the vehicle; the characteristic value determining module is used for intercepting a reflection image of a preset area in an image of each frame in the video image, wherein the preset area is an area where a reflection object in each frame is located; Acquiring the pixel number of the preset area and the brightness of each frame; Determining a living body parameter corresponding to a target frame according to a reflected image of the target frame, a reflected image of an adjacent frame and the pixel number of the preset area, wherein the target frame is any frame of all frames, and the adjacent frame is a frame adjacent to the target frame; Inputting the reflected image of the target frame into a preset living body classification model to obtain a target confidence coefficient; determining a brightness parameter of the target frame according to the brightness of the target frame and the brightness of the adjacent frames; Calculating a living body characteristic value according to the living body parameter, the brightness parameter and the target confidence coefficient of the target frame; the volume determining module is used for determining volume parameters according to the audio data; the parameter calculation module is used for calculating comprehensive parameters according to the living body characteristic value and the volume parameter; The characteristic value obtaining module is used for controlling the loudspeaker to emit preset sound if the comprehensive parameter is larger than a first preset value, and executing the step of determining the living characteristic value again to obtain a new living characteristic value; And the prompt sending module is used for sending corresponding living body prompt information to the user terminal according to the new living body characteristic value and the comprehensive parameter.
10. An electronic device comprising a processor and a memory communicatively coupled to the processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory, so that the processor executes the in-vehicle living body presenting method according to any one of claims 1 to 8.
11. A computer-readable storage medium, wherein computer-executable instructions are stored in the computer-readable storage medium, which when executed by a processor, are configured to implement the in-vehicle living body presenting method according to any one of claims 1 to 8.
12. A computer program product, characterized in that the computer program product comprises a computer program for implementing the in-vehicle living body presentation method as claimed in any one of claims 1 to 8 when being executed by a processor.

Description

In-vehicle living body prompting method, device, equipment and storage medium Technical Field The application relates to the technical field of intelligent automobiles, in particular to an in-vehicle living body prompting method, an in-vehicle living body prompting device, in-vehicle living body prompting equipment and a storage medium. Background With the continuous development of industrial technology, automobiles become an important transportation means for the masses to travel. In order to avoid the high temperature heatstroke condition caused by the infant being locked in the car, it is necessary to ensure that no infant is in the car after the car is locked. Currently, in the prior art, no infant in a car is usually determined through a face recognition scheme. However, the inventor finds that the prior art has at least the following technical problems that the accuracy of face recognition judgment is poor. Disclosure of Invention The application provides a method, a device, equipment and a storage medium for prompting living bodies in a vehicle, which are used for solving the problem of poor accuracy of judging living bodies through face recognition. In a first aspect, the application provides an in-vehicle living body prompting method, which comprises the steps of responding to the start of a detection process and continuously acquiring video images and audio data in a vehicle. A living body characteristic value is determined from images of frames in the video image. A volume parameter is determined from the audio data. And calculating the comprehensive parameters according to the living body characteristic values and the volume parameters. If the comprehensive parameter is larger than the first preset value, the loudspeaker is controlled to make a preset sound, and the step of determining the living body characteristic value is executed again to obtain a new living body characteristic value. And sending corresponding living body prompt information to the user terminal according to the new living body characteristic value and the comprehensive parameter. In one possible implementation, determining the living body characteristic value from the images of each frame in the video image includes capturing a reflection image of a preset area in each frame, where the preset area is an area where a reflective object in each frame is located. The pixel number of the preset area and the brightness of each frame are obtained. And determining a living body parameter corresponding to the target frame according to the reflected image of the target frame, the reflected image of the adjacent frame and the pixel number of the preset area, wherein the target frame is any frame in all frames, and the adjacent frame is a frame adjacent to the target frame. And inputting the reflected image of the target frame into a preset living body classification model to obtain the target confidence. And determining the brightness parameter of the target frame according to the brightness of the target frame and the brightness of the adjacent frames. And calculating a living body characteristic value according to the living body parameter, the brightness parameter and the target confidence coefficient of the target frame. In one possible implementation, determining the living parameters corresponding to the target frame according to the reflected image of the target frame, the reflected image of the adjacent frame and the number of pixels of the preset area includes extracting the picture features of the reflected image in each frame by using a preset model. Dividing the picture characteristic of each frame by the pixel number to obtain the unit picture characteristic of each frame. And carrying out difference on the unit picture characteristics of the target frame and the unit picture characteristics of the adjacent frames to obtain unit characteristic differences. If the unit characteristic difference is larger than the preset difference, setting the living parameter corresponding to the target frame to be 1, otherwise, setting the living parameter corresponding to the target frame to be 0. In one possible implementation, determining the luminance parameter of the target frame based on the luminance of the target frame and the luminance of the neighboring frame includes using the difference between the luminance of the target frame and the luminance of the neighboring frame to obtain a luminance difference value. If the brightness difference value is larger than the preset brightness difference threshold value, the brightness parameter of the target frame is determined to be 0, otherwise, the brightness parameter of the target frame is determined to be 1. In one possible implementation, the calculation of the living body characteristic value according to the living body parameter, the brightness parameter and the target confidence coefficient of the target frame comprises inputting the living body parameter, the brightness paramet