CN112067054A - Intelligent dressing mirror based on BMI detects - Google Patents

Abstract

The invention discloses a smart dressing mirror based on BMI detection, comprising a hardware front-end part and a software control part, the hardware front-end part includes an OLED organic display screen, a camera, a temperature sensor, a humidity sensor, a laser head, a frame, a bracket and a power supply, The OLED organic display screen, camera, temperature sensor, humidity sensor, and laser head are installed on the front of the frame, the bracket and power supply are installed on the back of the frame, and the software control part includes a main control module, a BMI detection module, a temperature A humidity detection module, a voice module, a display module and a network module, the BMI detection module, temperature detection module, voice module, display module and network module are respectively connected with the main control module. By using the present invention, the combination of a full-length mirror and artificial intelligence is realized. As an intelligent dressing mirror based on BMI detection, the present invention can be widely used in the field of intelligent interaction.

Description

A smart full-length mirror based on BMI detection

technical field

The invention relates to the field of intelligent interaction, in particular to an intelligent full-length mirror based on BMI detection.

Background technique

With the continuous progress of human society, the full-length mirror has become an indispensable household item in people's lives. The existing full-length mirrors on the market are generally composed of a mirror surface and a mirror frame, which are mainly used for people to observe the appropriateness of clothing and the overall image when wearing clothing, and can leave a good impression on others when going to work, dating or business negotiation. However, with the progress of society, people's life rhythm is getting faster and faster. In addition to arranging their appearance in front of the full-length mirror, people also want to know their BMI value at any time to conduct a preliminary test of their health status, obtain time, weather conditions and outdoor conditions. Temperature and other information so that users can better go out.

SUMMARY OF THE INVENTION

In order to solve the above technical problems, the purpose of the present invention is to provide a smart full-length mirror based on BMI detection, which is convenient for tracking the user's body changes and obtaining external environment information, and realizes the combination of the full-length mirror and artificial intelligence.

The first technical solution adopted by the present invention is: an intelligent full-length mirror based on BMI detection, comprising a hardware front-end part and a software control part, and the hardware front-end part includes an OLED organic display screen, a camera, a temperature sensor, a humidity sensor, a laser head, Frame, bracket and power supply, the OLED organic display screen, camera, temperature sensor, humidity sensor and laser head are installed on the front of the frame, the bracket and power supply are installed on the back of the frame, and the software control part includes a main control module , BMI detection module, temperature and humidity detection module, voice module, display module and network module, the BMI detection module, temperature detection module, voice module, display module and network module are respectively connected with the main control module.

The BMI detection module further includes:

Image acquisition sub-module for acquiring 2D body images;

Feature extraction sub-module for extracting depth features of 2D body images;

The BMI calculation submodule is used to calculate the BMI value according to the depth feature of the 2D body image.

Further, the BMI detection module is an end-to-end convolutional neural network, specifically a detection network based on DenseNet.

Further, the working steps of the BMI detection module specifically include:

Obtain a 2D body image and perform human skeleton joint detection and body contour detection on the 2D body image;

Remove the background of the picture and extract the depth feature value according to the information of human skeleton joint detection and body contour detection;

The BMI value of the human body in the 2D body image is calculated according to the depth feature value.

Further, the temperature and humidity detection module is configured to acquire temperature information and humidity information of the current environment according to the temperature sensor and the humidity sensor, and transmit the temperature information and humidity information to the display module and the voice module.

Further, the network module is used to access the network to search for weather, traffic conditions and stock market information and transmit these information to the display module and the voice module.

Further, the display module is configured to receive the information transmitted from the temperature and humidity detection module, the BMI detection module and the network module according to the instructions of the main control module and display the information.

Further, the voice module is configured to receive the information transmitted from the temperature and humidity detection module, the BMI detection module and the network module according to the instruction of the main control module, and display the information in an audio format.

Further, the main control module is used to issue an instruction to control the BMI detection module, the temperature detection module, the voice module, the display module and the network module.

The beneficial effects of the invention are as follows: the intelligent dressing mirror can measure the height, weight and BMI of the user in real time through the joint action of the laser head and the camera, so as to facilitate the tracking of body changes and health monitoring, and access the network through the network module for weather, traffic It can search for information such as status and stock market, and display the obtained information through the voice module and the display module, realizing the combination of a full-length mirror and artificial intelligence, so that users can get a better experience.

Description of drawings

Fig. 1 is a kind of front view of the concrete device of the intelligent full-length mirror based on BMI detection of the present invention;

Fig. 2 is a kind of concrete device side view of the intelligent full-length mirror based on BMI detection of the present invention;

Fig. 3 is a kind of back view of the concrete device of the intelligent full-length mirror based on BMI detection of the present invention;

Fig. 4 is the structural block diagram of a kind of intelligent full-length mirror based on BMI detection of the present invention;

FIG. 5 is a network structure diagram of a BMI detection module according to a specific embodiment of the present invention.

Reference signs: 1. Camera; 2. Laser head; 3. OLED organic display screen; 4. Switch button; 5. Bracket; 6. Frame; 7. Software control part; 8. Power supply; 9. Temperature sensor; 10 , Humidity sensor; 11. Power cord.

Detailed ways

The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. The numbers of the steps in the following embodiments are only set for the convenience of description, and the sequence between the steps is not limited in any way, and the execution sequence of each step in the embodiments can be adapted according to the understanding of those skilled in the art Sexual adjustment.

As shown in Figure 1, Figure 2, Figure 3 and Figure 4, the present invention provides a smart dressing mirror based on BMI detection, including a hardware front-end part and a software control part, and the hardware front-end part includes an OLED organic display screen 3, a camera 1 , temperature sensor 9, humidity sensor 10, laser head 2, mirror frame 6, bracket 5 and power supply 8, the OLED organic display screen 3, camera 1, temperature sensor 9, humidity sensor 10 and laser head 2 are installed on the mirror frame 6 On the front, the bracket 5 and the power supply 8 are installed on the back of the frame 6, and the software control part includes a main control module, a BMI detection module, a temperature and humidity detection module, a voice module, a display module and a network module. The BMI detection module, The temperature detection module, the voice module, the display module and the network module are respectively connected with the main control module.

Further as a preferred embodiment of the present invention, the BMI detection module further includes:

Image acquisition sub-module for acquiring 2D body images;

Feature extraction sub-module for extracting depth features of 2D body images;

The BMI calculation submodule is used to calculate the BMI value according to the depth feature of the 2D body image.

As a further preferred embodiment of the method, the BMI detection module is an end-to-end convolutional neural network, specifically a DenseNet-based detection network.

Specifically, we propose an "end-to-end" convolutional neural network to calculate the BMI value, that is, the input is a single frontal 2D image of the human body, and the output is the BMI value. This method requires neither depth information nor clear face information, and only uses a 2D body image as the network input.

As a dense connection network based on skip connection, it allows the current layer to be connected to all previous layers. In addition to simply using DenseNet, we also add SE-layer (attention module) after each transition layer. Loss to learn feature weights, so that the effective feature map has a large weight, and the invalid or small effect feature map has a small weight to train the model to achieve better results. We replaced the last fully connected layer of the original DenseNet network with three fully connected layers, and finally output a BMI value.

In addition, our proposed BMI computing network has a total of four Dense Blocks, and each Dense Block is composed of a different number of Dense Units, and each unit is composed of 1*1 and 3*3. There is a Transition Layer (transition layer) and SE-layer (attention module) between each Dense Block. Each transition layer is a 1×1 convolution operation in order to reduce the number of eigenvalue parameters.

Refer to Figure 5 for the structure and detection steps of the BMI detection module, where the Dense Block represents a dense block.

Further as a preferred embodiment of the method, the working steps of the BMI detection module specifically include:

Obtain a 2D body image and perform human skeleton joint detection and body contour detection on the 2D body image;

Remove the background of the picture and extract the depth feature value according to the information of human skeleton joint detection and body contour detection;

The BMI value of the human body in the 2D body image is calculated according to the depth feature value.

Specifically, in order to avoid image background interference while extracting depth features, we remove the image background by detecting body contours in the input image. Background removal is divided into two parts, including human skeleton joint detection and body contour detection. For human skeleton joint detection, we use the Mask R-CNN method; for body contour detection, we use the Pose2Seg method, which is processed based on Mask R-CNN. On the photo, the outline of the human body is detected.

As a further preferred embodiment of the present invention, the temperature and humidity detection module is used to obtain the temperature information and humidity information of the current environment according to the temperature sensor 9 and the humidity sensor 10 and transmit the temperature information and humidity information to the display module and the voice module.

As a further preferred embodiment of the present invention, the network module is used to access the network to search for weather, traffic conditions and stock market information and transmit these information to the display module and the voice module.

Specifically, the network module can also obtain more corresponding information by presetting a specific website and transmit it to the display module and the voice module for information display, such as news hotspots.

Further as a preferred embodiment of the present invention, the display module is configured to receive the information transmitted from the temperature and humidity detection module, the BMI detection module and the network module according to the instructions of the main control module, and display the information.

Further as a preferred embodiment of the present invention, the voice module is configured to receive the information transmitted from the temperature and humidity detection module, the BMI detection module and the network module according to the instructions of the main control module, and display the information in audio format.

Specifically, the voice module can also perform functions such as alarm clock reminder and memo reminder for the user through settings.

The working principle of the present invention is as follows:

In response to the user's instruction input, the main control module controls the camera 1 to open to obtain the current user image, input it to the BMI detection module, calculate the BMI value through the built BMI calculation network, and then feed it back to the user through the OLED organic display screen and the voice module; the laser head 2 is used to project to the ground, the user stands in the designated projection, the height and weight of the user can be calculated through the photos taken by the camera 1 and the BMI detection module, and can also be fed back to the user through the OLED organic display 3 and the voice module; Switch button 4, the full-length mirror can be used as an ordinary fitting mirror, turn on the switch button 4, the full-length mirror can measure BMI, height and weight, temperature and humidity, and broadcast the weather and other functions; the bracket 5 is used to support and better fix the full-length mirror.

The intelligent full-length mirror based on BMI detection of the present invention can realize non-contact measurement of height, weight and BMI value on the basis of realizing basic functions, which is convenient for tracking body changes and health monitoring, and is also equipped with an intelligent voice assistant, which can be connected to the network. Search for information such as weather, traffic conditions, stock market, etc., and measure the temperature and humidity of the environment. Combining the full-length mirror and artificial intelligence makes the user have a better experience.

The contents in the above method embodiments are all applicable to the present system embodiments, the specific functions implemented by the present system embodiments are the same as the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

The above is a specific description of the preferred implementation of the present invention, but the present invention is not limited to the described embodiments, and those skilled in the art can make various equivalent deformations or replacements without departing from the spirit of the present invention. , these equivalent modifications or substitutions are all included within the scope defined by the claims of the present application.

Claims (9)

1. The utility model provides an intelligent dressing mirror based on BMI detects, its characterized in that, includes hardware front end portion and software control portion, and the hardware front end portion includes OLED organic display screen, camera, temperature sensor, humidity transducer, radium head, mirror holder, support and power, OLED organic display screen, camera, temperature sensor, humidity transducer and radium head are installed at the mirror holder openly, support and power are installed at the mirror holder back, software control portion includes host system, BMI detection module, temperature and humidity detection module, voice module, show module and network module, BMI detection module, temperature detection module, voice module, show module and network module are connected with host system respectively.

2. The intelligent dressing mirror according to claim 1, wherein the BMI detection module further comprises:

an image acquisition sub-module for acquiring a 2D body image;

the feature extraction submodule is used for extracting the depth features of the 2D body image;

and the BMI calculation submodule is used for calculating to obtain a BMI value according to the depth characteristics of the 2D body image.

3. The intelligent full-length mirror based on BMI detection as claimed in claim 2, wherein the BMI detection module is an end-to-end convolutional neural network, specifically a DenseNet-based detection network.

4. The intelligent dressing mirror according to claim 3, wherein the BMI detection module comprises:

acquiring a 2D body image and carrying out human skeleton joint detection and body contour detection on the 2D body image;

removing a picture background according to information of human body skeleton joint detection and body contour detection and extracting to obtain a depth characteristic value;

and calculating to obtain the human body BMI value in the 2D body image according to the depth characteristic value.

5. The intelligent full-length mirror based on BMI detection as claimed in claim 4, wherein the temperature and humidity detection module is used to obtain the temperature information and humidity information of the current environment according to the temperature sensor and humidity sensor and transmit the temperature information and humidity information to the display module and the voice module.

6. The intelligent full-length mirror based on BMI detection as claimed in claim 5, wherein the network module is used to access the network for searching weather, traffic condition and stock market information and transmitting the information to the display module and the voice module.

7. The intelligent full-length mirror based on BMI detection as claimed in claim 6, wherein the display module is used for receiving the information transmitted from the temperature and humidity detection module, the BMI detection module and the network module according to the instruction of the main control module and displaying the information.

8. The intelligent full-length mirror based on BMI detection as claimed in claim 7, wherein the voice module is used for receiving the information transmitted from the temperature and humidity detection module, the BMI detection module and the network module according to the instruction of the main control module and displaying the information in an audio format.

9. The intelligent full-length mirror based on BMI detection as claimed in claim 8, wherein the main control module is used for sending commands to control the BMI detection module, the temperature detection module, the voice module, the display module and the network module.

Images