WO2018086056A1

WO2018086056A1 - Combined sound system for automatically capturing positioning of human face

Info

Publication number: WO2018086056A1
Application number: PCT/CN2016/105427
Authority: WO
Inventors: 吴新龙
Original assignee: 吴新龙
Priority date: 2016-11-11
Filing date: 2016-11-11
Publication date: 2018-05-17

Abstract

A combined sound system for automatically capturing positioning of a human face consists of a sensor module (1), a processing control module, a sound production module (2), and a direction control module. The sound production module (2) consists of a directional sound production unit (3) and an omnidirectional sound production unit (4). The processing control module controls the positioning direction of the sound production module (2) by computing data inputted by the sensor module (1), and in a directional sound production mode, a hearer can hear clear directional sound source information without being limited to a specific region. The combined sound system combines a sound frequency positioning technology with a conventional sound technology, breaks through the disadvantage of undiversified sound production, and can resolve the problems such as noise pollution and private information leakage.

Description

Combined sound system for automatically recognizing and capturing face positioning

Technical field

The utility model relates to an acoustic system for automatically recognizing and capturing face positioning, relating to the field of electroacoustics, in particular to a combined acoustic product for automatically recognizing and capturing face positioning.

Background technique

At present, there are many kinds of audio products on the market, and there are many functions. However, these audio products using traditional technology have the same characteristics, that is, the sound emitted can be heard in the whole space, which is specific. Some problems may arise in the application scenario, such as noise problems, private information leakage problems, interference problems between multiple prompts, etc. These problems have also been paid more and more attention and attention.

The audio directional technology is also called parametric acoustic array technology. The main theoretical basis is that two columns of finite amplitude but different frequency sound waves meet in the air, and the nonlinear interaction of air produces audible signals that can be directionally propagated. This is different from the omnidirectional sound waves emitted by conventional speakers, which control the audible sound to directionally propagate in an area of space, and only a small or no sound can be heard outside the area. Therefore, this feature of the directional speaker can well solve problems such as noise pollution and leakage of private information.

The combined sound system of the patented invention combines the audio orientation technology and the traditional sound technology, and combines the traditional omnidirectional speaker and the directional sounding unit in the same sound system, and the user can select the omnidirectional sound source playing mode according to different occasions. Or directional audio playback mode, so such products can meet the diverse needs of consumers.

The directional speaker can be heard because its sound is limited in space. Currently, the directional speakers on the market are fixedly installed, and the listener must enter the audible area covered by the directional speaker to hear the sound. The intelligent sound system and the humanization of the product, the combined sound system of the invention is added Automatic face recognition technology, the directional sound module can automatically align the face position locked to the audience target, so that the listener can hear the audio information played by the directional sound unit without having to stay in the specified small area.

Summary of the invention

The purpose of this patent is to provide a combined sound that can automatically recognize and capture the face positioning, which has both a directional sounding function and an omnidirectional sounding function, and the automatic face recognition positioning function of the product realizes product intelligent control.

In order to achieve the above objectives, the patent is realized by the following technical solutions:

The sensor module is configured to detect a target object entering the target area, and collect information such as a facial feature, a height contour, and a position of the listener. The information is transmitted to the information processing control module, and the sensor module may be one or more of an image recognition sensor, an ultrasonic sensor, an infrared sensor, and a thermal energy sensor.

The sound module is composed of a directional sounding unit module and a traditional omnidirectional speaker module. The directional sounding unit emits a sound beam with a certain angle, and the audible area is small, while the traditional omnidirectional speaker emits an omnidirectional sound. It can be heard in the entire space. The combined sound system can select the directional sounding unit sound or the traditional omnidirectional sound according to different applications, and input the corresponding command signal by inputting the corresponding command signal to the signal processing control module.

The core processing control module determines the angle of rotation of the sounding module through the core algorithm to realize the function of the combined sound system to face the face. The algorithms that can be selected vary depending on the sensor module used. Based on the image recognition sensor, the algorithms that can be used are gray-scale template matching method based on the overall face, artificial neural network method, skin color detection method and the like. Another function of the core processing control module is to control the output of the audio signal, and input the sound source mode, volume adjustment, playback pause, etc. through the button module of the audio system. Control signals to control the output of the audio.

The direction control module is a motion executing component that realizes face positioning, and is composed of a horizontal rotation connecting device and a vertical rotation connecting device. When the detected target enters the detection range of the sensor, the processing control module calculates the angle values of the horizontal rotation and the vertical rotation through the core algorithm, outputs the motor drive module to the horizontal and vertical rotation device, and tracks the target listener who has entered the sensor detection range in real time. .

The combined sound system disclosed in the patent combines the audio orientation technology with the traditional sound technology, breaks the shortcoming of the single sound of the traditional sound, makes the product more entertaining, and solves the problems of noise pollution and private information leakage. The combined audio system integrates the face recognition technology to track and locate the listener's face area in real time. The listener can hear the directional sound source without being limited to staying in the specified small area, and truly realize the product intelligence.

DRAWINGS

In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the embodiments will be briefly described below. It should be understood that the following drawings show only certain embodiments of the present invention, It should not be seen as limiting the scope, and those of ordinary skill in the art can also obtain other related drawings based on these drawings without the inventive work.

Figure 1 is an embodiment of a combined audio system architecture.

2 is an embodiment of a combination of an omnidirectional speaker unit and a directional sounding unit.

Figure 3 is a block diagram of the core signal processing and control structure of the system.

4 is a schematic diagram of the automatic face recognition operation of the integrated sound system.

detailed description

The invention aims to design a combined sound system for automatically recognizing and capturing face positioning, and attaching The embodiment is described in detail with reference to the embodiments, which are only intended to illustrate the technical solution, and the scope of the patent is not limited thereby.

FIG. 1 shows an embodiment of an audio system structure for automatically recognizing a face-to-face location proposed by the present patent. The sensor module 1 is configured to collect external data, and the sounding module 2 includes a directional sounding unit 3 and a conventional omnidirectional speaker 4. The support base 9 of the system provides support and fixation for the entire system, the horizontal rotary connection 8 is used for 360 degree horizontal rotation of the sound module 2, and the vertical rotary connection 7 is used for adjustment and fixation of the sound module 2 in the vertical direction. The system core control circuit 11 is used for overall system function control. The button module 5 of the system is used for system audio source play mode switching, volume adjustment, and play pause adjustment.

Figure 2 shows an embodiment of a combination of an omnidirectional speaker unit and a directional sounding unit, which combines the directional sounding unit 3 and the conventional omnidirectional horn 4 to provide both omnidirectional sound and a certain An angular range of audible sounds. In Fig. 2, only the traditional omnidirectional horn and the directional sounding unit are combined on the same side. The traditional omnidirectional horn and the directional sounding unit can also be installed on different planes, which is more conducive to the perfect combination of sound.

3 is a block diagram of the core signal processing and control structure of the system. The system core circuit 11 receives the external data collected by the sensor module 1 on the one hand, and processes the control signal input by the button module 5 on the other hand, and outputs the control signal to the power amplifier drive. And the motor drive, the driving directional sounding unit 3 and the conventional omnidirectional horn 4 play the user-specified audio, and the driving motor motor rotates to make the sounding module 2 point to the face part.

FIG. 4 is a schematic diagram of the automatic face recognition operation of the integrated sound system. When the detected target 12 enters the detection range of the sensor, the sensor collects the face and position data, and analyzes and processes the data through the core signal processor. The rotation of the drive motor causes the sounding module 2 to point to the face portion of the detected target entering the detection area.

Claims

A combined sound system for automatically recognizing and capturing face orientation, the system comprising:

a sensor module for collecting external data;

a processing control module, configured to perform integrated processing on the collected external data and the received control signal input, and implement motion control of the directional speaker unit pointing by the motion executing component;

a speaker module for playing a user-specified audio file;

Motion control module for controlling the orientation of the directional speaker unit in horizontal and vertical directions.
The combined sound system for automatically recognizing and capturing face orientation according to claim 1, wherein the sensor module recognizes data such as face data, human body features, and orientation of the human body relative to the sound system.
The combined sound system for automatically recognizing a face position according to claim 1, wherein the sensor module is composed of one or more of an image recognition sensor, an ultrasonic sensor, an infrared sensor, a thermal sensor, and the like.
The combined sound system for automatically recognizing a face positioning according to claim 1, wherein the sounding module comprises: a unit having directivity sounding, and a speaker unit having an omnidirectional sounding.
The combined sound system for automatically recognizing and capturing face positioning according to claim 4, wherein the material of the sounding unit having directivity is an array of ultrasonic transducers, or may be a PVDF or EMFi piezoelectric film. .
The combined sound system for automatically recognizing a face positioning according to claim 4, wherein the combination of the directional sounding unit and the omnidirectional sound emitting unit may be the same plane or may be Different planes.
The combined sound system for automatically recognizing a face positioning according to claim 1, wherein the direction control module comprises: a horizontal 360 degree rotational connection device, and a vertical direction rotation connection device.