CN208689542U - Gesture recognition control system towards Intelligent household scene - Google Patents
Gesture recognition control system towards Intelligent household scene Download PDFInfo
- Publication number
- CN208689542U CN208689542U CN201820322876.1U CN201820322876U CN208689542U CN 208689542 U CN208689542 U CN 208689542U CN 201820322876 U CN201820322876 U CN 201820322876U CN 208689542 U CN208689542 U CN 208689542U
- Authority
- CN
- China
- Prior art keywords
- module
- gesture
- gesture recognition
- device control
- fitment
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Abstract
Gesture recognition control system towards Intelligent household scene, including video acquisition module, gesture recognition module, device control module and function setting module;The input terminal connection monitoring capture apparatus of video acquisition module, obtains image by monitoring capture apparatus, output end connects the input terminal of gesture recognition module;Gesture recognition module identifies the gesture in image, the output end of gesture recognition module connects device control module, the control terminal of device control module connects fitment, and device control module passes through LAN connection function setting module according to the gesture control fitment of identification, device control module;Video acquisition module is in monitoring capture apparatus, and gesture recognition module is in server apparatus, and device control module and function setting module are in fitment;This system utilizes the gesture of gesture recognition module intelligent recognition user, controls the electric equipment operation in household, effectively increases the operability of system.
Description
Technical field
The utility model belongs to intelligent identification technology field, and in particular to a kind of gesture identification towards Intelligent household scene
Control system.
Background technique
Gesture identification refers to the images of gestures for obtaining personage in the picture, feature is extracted from images of gestures, by knowledge
The feature indescribably taken obtains the meaning of gesture.Control house electric can be reached by identifying user gesture under household scene
The effect of equipment.User just can control the operation of the equipment such as TV, air-conditioning, light using simple gesture.
According to the applicant understood, now in the market, the smart home device with gesture identification function is more rare, most hand
Gesture identification product can only identify limb action, can not identify to gesture, and recognition effect is unsatisfactory, meanwhile, these productions
Product excessively rely on networking cloud server, can not work in the absence of a network, can not to a variety of electrical equipments into
Row effectively control.
Utility model content
The shortcomings that technical problem to be solved in the utility model is, overcomes the prior art provides one kind towards intelligent family
The gesture recognition control system for occupying scene controls electrically setting in household using the gesture of gesture recognition module intelligence identification user
Received shipment row effectively increases the operability of system.
In order to solve the above technical problems, the utility model provides a kind of gesture identification control towards Intelligent household scene
System, including video acquisition module, gesture recognition module, device control module and function setting module;Video acquisition module
Input terminal connection is located at the monitoring capture apparatus in fitment, and video acquisition module obtains image by monitoring capture apparatus,
The input terminal of its output end connection gesture recognition module;Gesture recognition module identifies the gesture in image, gesture recognition module
Output end connects device control module, and the control terminal of device control module connects fitment, and device control module is according to identification
Gesture control fitment, device control module pass through LAN connection function setting module;It is equipped in function setting module
Gesture and fitment control corresponding relation database;Video acquisition module is in monitoring capture apparatus, gesture recognition module
In server apparatus, device control module and function setting module are in fitment.
The technical solution that the utility model further limits is: gesture recognition module includes images of gestures extraction unit, hand
Gesture feature extraction unit and gesture model training unit, images of gestures extraction unit are mentioned from the image that video acquisition module obtains
Images of gestures is taken out, and images of gestures is transferred to gesture feature extraction unit;Gesture feature extraction unit dividing gesture image,
Gesture feature therein is extracted, gesture model training unit is trained according to gesture feature and obtains gesture identification model.
Further, server apparatus is raspberry pie server apparatus.
Further, local area network can be set up by wireless routing or bluetooth.
Further, function setting module establishes the instruction of gesture and fitment control corresponding relationship, and instruction is logical
It crosses local area network and is sent to device control module.
The beneficial effects of the utility model are: this system can carry out intelligent control to fitment by identification gesture, together
When, the training pattern in gesture recognition module can guarantee that this system can still provide for work in the absence of a network;System
It is middle to use raspberry pie server apparatus, both meet professional platform independence, also reduces the cost for setting up smart home.
Detailed description of the invention
Fig. 1 is the system structure diagram of the utility model.
Specific embodiment
Embodiment 1
Referring to Fig. 1, a kind of gesture recognition control system towards Intelligent household scene provided in this embodiment, including view
Frequency acquisition module, gesture recognition module, device control module and function setting module;The input terminal connection position of video acquisition module
In the monitoring capture apparatus in fitment, video acquisition module obtains image, output end connection by monitoring capture apparatus
The input terminal of gesture recognition module;Gesture recognition module identifies the gesture in image, and the output end connection of gesture recognition module is set
Standby control module, the control terminal of device control module connect fitment, and device control module is according to the gesture control man of identification
Has equipment, device control module passes through LAN connection function setting module;Gesture is equipped in function setting module to set with furniture
Standby control corresponding relation database;In monitoring capture apparatus, gesture recognition module is set video acquisition module embedded in server
Standby interior, device control module and function setting module are in fitment.
Video acquisition module: the module be for obtaining video captured by Household monitor capture apparatus, theoretically,
This module may be implemented to obtain video captured by any camera.Video acquisition module is by giving collected transmission of video
Positioned at the gesture recognition module of server end, for being identified to present image.So video acquisition module is realized to family
It occupies the unified video acquisition of monitoring capture apparatus and obtains.
Gesture recognition module: including carrying the gesture identification model of installation on the server, by video acquisition mould
The image of block transmission carries out pretreatment operation and identification operation, identifies the gesture meaning in image, and gesture information is passed to
Device control module, for controlling the electrical equipment in household.
Function setting module: for setting the corresponding relationship between gesture and equipment, mould is controlled by local area network and equipment
Block connection.Local area network can be constructed by wireless routing gas or bluetooth.User can be by local area network between gesture and equipment
Corresponding relationship instruction is sent to device control module.
Device control module: for controlling the operation of electrical equipment in household.It receives what gesture recognition module was transmitted
Gesture instruction, and receive capabilities setting module is instructed by gesture transmitted by local area network and equipment corresponding relationship
Gesture recognition module includes images of gestures extraction unit, gesture feature extraction unit and gesture model training unit,
Images of gestures extraction unit extracts images of gestures from the image that video acquisition module obtains, and images of gestures is transmitted in one's hands
Gesture feature extraction unit;Gesture feature extraction unit dividing gesture image extracts gesture feature therein, gesture model training
Unit is trained according to gesture feature and obtains gesture identification model.
Images of gestures extraction unit: for extracting to the gesture in image, reduce background to the shadow of identification accuracy
It rings.Image preprocessing and Hand Gesture Segmentation two operations are wherein contained, is pre-processed to be returned using Gamma correction method to image
One changes, and carries out gaussian filtering process, and wherein Gamma correction is exactly to edit to the gamma curve of image, with to image into
The method of row non-linear tone editor detects dark parts and light-colored part in picture signal, and increases the two ratio, from
And improve picture contrast effect, reduce image local shade and illumination variation caused by influence, while can inhibit to make an uproar
The interference of sound;Gaussian filtering mainly eliminates influence of the noise to identification.Pretreated image is transformed into HSV from rgb space
Color space, wherein HSV (Hue, Saturation, Value) is the intuitive nature of color, the parameter of color in this model
It is respectively: tone (H), saturation degree (S), lightness (V).Each pixel is compared with threshold value, and is fitted with rectangle;Hand
The HSV value of the colour of skin is [95:135,30:255,45:255] in section, if the HSV value of pixel belongs to this area in image
Between, then the pixel value is set to 255, is otherwise set to 0, then mask figure can be obtained.Mask figure corresponding part is intercepted in original image, as
Images of gestures.
Gesture feature extraction unit: for carrying out feature extraction to images of gestures, for training gesture model.Gesture figure
As being divided into 9 units;Its LBP value is calculated to pixel each in image, and normalizing is carried out to the LBP value of each unit
Change, wherein LBP(Local Binary Pattern, local binary patterns) it is a kind of for describing image local textural characteristics
Operator;It has the advantages that rotational invariance and gray scale invariance etc. are significant.;Using the LBP value of each unit of statistics with histogram,
And be connected into a histogram, i.e. the LBP feature histogram of images of gestures.
Gesture model training unit: for being trained to the gesture feature of extraction, gesture identification model is obtained.Prepare instruction
Practice sample set;Including positive sample collection (various gesture samples) and negative sample collection;Training sample should be infinite number of, Er Qiexun
The various situations that may occur in actual application should be covered by practicing sample.It needs to cut sample manually, protects sample-size
It holds consistent.Extract the LBP feature of all positive and negative samples;By the LBP feature of positive negative sample, the label of positive negative sample is all input to
It is trained in SVM;Wherein instructed using radial base (the Radial Basis Function abbreviation RBF) kernel function in SVM
It tries one's hand potential model.
Server apparatus is raspberry pie server apparatus.
Local area network can be set up by wireless routing or bluetooth.
Function setting module establishes the instruction of gesture and fitment control corresponding relationship, and instruction is sent out by local area network
Give device control module.
Video acquisition module is embedded in Haikang prestige view DS-2AF1-714 camera, and gesture recognition module is embedded in raspberry
It sends in raspberry 3b, device control module mainly includes the home equipments such as electric light, air-conditioning, curtain, and function setting module is embedding
Enter in the smart phone with Android system.
In addition to the implementation, the utility model can also have other embodiments.It is all to use equivalent replacement or equivalent change
The technical solution to be formed is changed, the protection scope of the requires of the utility model is all fallen within.
Claims (3)
1. the gesture recognition control system towards Intelligent household scene, which is characterized in that including video acquisition module, gesture identification
Module, device control module and function setting module;The input terminal connection of the video acquisition module is located in fitment
Capture apparatus is monitored, the video acquisition module obtains image by monitoring capture apparatus, and output end, which connects the gesture, to be known
The input terminal of other module;Gesture in the gesture recognition module identification image, the output end connection of the gesture recognition module
The control terminal of the device control module, the device control module connects fitment, and the device control module is according to knowledge
Other gesture control fitment, the device control module pass through function setting module described in LAN connection;The function
It is equipped with gesture in setting module and fitment controls corresponding relation database;The video acquisition module is set embedded in monitoring shooting
Standby interior, in server apparatus, the device control module is embedded in the gesture recognition module with the function setting module
In fitment.
2. the gesture recognition control system according to claim 1 towards Intelligent household scene, which is characterized in that the clothes
Business device equipment is raspberry pie server apparatus.
3. the gesture recognition control system according to claim 1 towards Intelligent household scene, which is characterized in that the office
Domain net can be set up by wireless routing or bluetooth.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201820322876.1U CN208689542U (en) | 2018-03-09 | 2018-03-09 | Gesture recognition control system towards Intelligent household scene |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201820322876.1U CN208689542U (en) | 2018-03-09 | 2018-03-09 | Gesture recognition control system towards Intelligent household scene |
Publications (1)
Publication Number | Publication Date |
---|---|
CN208689542U true CN208689542U (en) | 2019-04-02 |
Family
ID=65875538
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201820322876.1U Active CN208689542U (en) | 2018-03-09 | 2018-03-09 | Gesture recognition control system towards Intelligent household scene |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN208689542U (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111079716A (en) * | 2020-01-07 | 2020-04-28 | 哈尔滨理工大学 | FPGA-based video gesture recognition intelligent home control system |
CN112232192A (en) * | 2020-10-15 | 2021-01-15 | 南京邮电大学 | Gesture convenient control system for disabled people |
CN114115524A (en) * | 2021-10-22 | 2022-03-01 | 青岛海尔科技有限公司 | Interaction method of intelligent water cup, storage medium and electronic device |
-
2018
- 2018-03-09 CN CN201820322876.1U patent/CN208689542U/en active Active
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111079716A (en) * | 2020-01-07 | 2020-04-28 | 哈尔滨理工大学 | FPGA-based video gesture recognition intelligent home control system |
CN112232192A (en) * | 2020-10-15 | 2021-01-15 | 南京邮电大学 | Gesture convenient control system for disabled people |
CN114115524A (en) * | 2021-10-22 | 2022-03-01 | 青岛海尔科技有限公司 | Interaction method of intelligent water cup, storage medium and electronic device |
CN114115524B (en) * | 2021-10-22 | 2023-08-18 | 青岛海尔科技有限公司 | Interaction method of intelligent water cup, storage medium and electronic device |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN208689542U (en) | Gesture recognition control system towards Intelligent household scene | |
CN106874871B (en) | Living body face double-camera identification method and identification device | |
CN108596140A (en) | A kind of mobile terminal face identification method and system | |
CN106372576A (en) | Deep learning-based intelligent indoor intrusion detection method and system | |
CN106462766A (en) | Image capturing parameter adjustment in preview mode | |
CN105404846A (en) | Image processing method and apparatus | |
CN109542233B (en) | Lamp control system based on dynamic gestures and face recognition | |
CN105279487A (en) | Beauty tool screening method and system | |
CN111339831B (en) | Lighting lamp control method and system | |
CN107659799B (en) | Image pickup apparatus, image processing method, and storage medium | |
CN103634680A (en) | Smart television play control method and device | |
CN108345867A (en) | Gesture identification method towards Intelligent household scene | |
CN112489143A (en) | Color identification method, device, equipment and storage medium | |
CN108711160B (en) | Target segmentation method based on HSI (high speed input/output) enhanced model | |
CN110287787A (en) | Image-recognizing method, device and computer readable storage medium | |
CN110336719A (en) | A kind of long-distance intelligent appliance control method of real-time video domestic environment equipment identification interaction | |
CN110222647B (en) | Face in-vivo detection method based on convolutional neural network | |
CN104392211A (en) | Skin recognition method based on saliency detection | |
CN108154496A (en) | A kind of power equipment appearance suitable for electric operating robot changes recognition methods | |
Lee et al. | A tank fish recognition and tracking system using computer vision techniques | |
CN111310703B (en) | Identity recognition method, device, equipment and medium based on convolutional neural network | |
Bojie et al. | Research on tea bud identification technology based on HSI/HSV color transformation | |
CN108650455B (en) | Intelligence house illumination data acquisition terminal | |
CN108270984A (en) | Image recognition type noise reduction broadcasting control system | |
CN108171195A (en) | A kind of face identification method, device and the access control system of identity-based certificate |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
GR01 | Patent grant | ||
GR01 | Patent grant |