KR101601187B1 - Device Control Unit and Method Using User Recognition Information Based on Palm Print Image - Google Patents

Abstract

The present invention relates to a device for controlling another device by using user recognition information based on the palm-print. The device includes: a palm-print sensor which senses the contact of a palm and records a palm-print image to obtain palm-print image data; an image processing unit which filters the obtained palm-print image data to execute a pre-process; a palm-print feature area detecting unit which analyzes the preprocessed palm-print image data to extract the feature point data of the palm-print image data; a user recognition unit which matches the extracted feature point data and feature point data of registered palm-print image data to execute a user authentication operation through similarity comparison; and a learning/setting control pattern database unit which receives a control signal from a device input unit and user information related to the user authentication from the user recognition unit, generates an output signal for controlling control target devices according to a control pattern set by learning the user information, and transmits the output signal to at least one of the control target devices.

Description

Technical Field [0001] The present invention relates to a device control apparatus using palm-based user recognition information,

More particularly, the present invention relates to a device control device and method thereof, and more particularly, to a device control device and a method thereof, and more particularly to a device control device and method thereof that analyze palmation information through a sensor capable of acquiring an image of a palm to discriminate user information corresponding to palm- And more particularly, to a device control apparatus using palm-based user identification information and a method thereof.

Today, the development of computers and electronic banking services using the Internet, electronic commerce, smart cards, etc., have attracted attention as a means of security and authentication of information stored in various databases.

The biomatrix is based on individual physical characteristics (face, fingerprint, iris, etc.), among which fingerprint-based identification is the most developed and proven technology.

Fingerprint recognition technology has been applied to various fields such as access control, crime investigation and resident registration certificate as the most effective personal identification method for a long time.

The fingerprint recognition method is largely classified into an optical sensing method using light and a non-optical sensing method using no light.

The fingerprint detection sensor using the optical sensing method has advantages that the surface on which the finger is closely attached is not coated, the surface is scarcely scratched, there is no corrosion, and the manufacturing cost and maintenance are easy.

Fingerprint interpretation method of the fingertip part may fail authentication if the already registered fingerprint is scratched or damaged due to any reason, and interpretation of fingerprint of the fingerprint is interpreted as another person's fingerprint.

In addition, the fingerprint interpretation method has a problem that the authentication rate of the fingerprint analysis of people who have extremely thin fingerprints such as a child and the elderly falls, and when the temperature falls in the winter, the authentication points are lowered due to the contouring of the feature points due to the contraction phenomenon of the finger skin. I have to.

The user recognition function is becoming an indispensable function in various smart devices representing modern smart phones.

However, existing user recognition technology mainly uses fingerprint, and smartphone uses face recognition.

Beyond smartphones and wearable devices, IT companies are competing for the smart home market.

Hyundai's smart home means an environment where various services can be provided through the configuration of various home appliances and sensors based on wired and wireless networks, and it is possible to provide safe and automated services anytime and anywhere. Through the home server, It means possible environment.

In such a smart home environment, the user recognition function is not yet included unless the control device represented by the remote control requires a high degree of security.

The smart home has been used to provide services only to authorized users, with fingerprinting mainly for security purposes, but no effective user recognition technology has yet been established.

In order to solve these problems, a problem to be solved by the present invention is to provide a method and apparatus for analyzing palmation information by using a sensor capable of acquiring an image of a palm to discriminate user information corresponding to palmization information, And to provide a device control apparatus and method using palm-based user recognition information to be performed.

According to an aspect of the present invention, there is provided a device control apparatus using palm-based user identification information,

A palm detection sensor for sensing the touch of the palm and capturing the palm image to acquire palm image data;

An image processor for filtering and processing the obtained palm image data;

A palm feature region detection unit for extracting the feature point data of the palm image data by analyzing the preprocessed palm image data;

A user recognition unit for matching the extracted feature point data with the feature point data of the registered palm image data and performing user authentication through comparison of similarity; And

Receiving an operation control signal from the device input unit, and generating an output signal for controlling the control target device in accordance with the control pattern learned in the user information and transmitting the control signal to at least one control target device And a learning / setting control pattern database unit.

According to another aspect of the present invention, there is provided a device control method using palm-based user identification information,

Detecting palm contact and capturing palm image to obtain palm image data;

Filtering and processing the obtained palm image data;

Extracting feature point data of the palm image data by analyzing the preprocessed palm image data;

Matching the extracted feature point data with the feature point data of the registered palm image data and performing user authentication through comparison of similarity;

Receiving user information on user authentication and an operation control signal from a device input unit; And

And generating an output signal for controlling the control target device according to the control pattern learned in the user information and transmitting the output signal to at least one control target device.

According to the above-described configuration, the present invention provides a device control service of different types for each user without any effort for user recognition by installing the hand-held sensor at the palm-contacting position of the device control device.

In the smart home environment, one control device controls various devices to provide a variety of services according to the user, thereby contributing to the development of a user-based device control device market.

1 is a block diagram of a device control apparatus using palm-based user recognition information according to an embodiment of the present invention.
2 is a block diagram showing a configuration of a learning / setting control pattern database unit according to an embodiment of the present invention.
3 is a conceptual diagram illustrating a palm-based user recognition process according to an embodiment of the present invention.
4 is a conceptual diagram illustrating a user recognition-based device control module according to an embodiment of the present invention.
5 is a flowchart illustrating a method of controlling a device using palm-based user recognition information according to an embodiment of the present invention.
6 is a flowchart illustrating an outline extraction process according to an embodiment of the present invention.
FIG. 7 is a diagram showing a resultant image when a 3 * 3 mask according to an embodiment of the present invention is applied.
8 is a view illustrating a barge removing process according to an embodiment of the present invention.
9 is a view for explaining a barge removing process according to an embodiment of the present invention.

BRIEF DESCRIPTION OF THE DRAWINGS The advantages and features of the present invention and the manner of achieving them will become apparent with reference to the embodiments described in detail below with reference to the accompanying drawings. The present invention may, however, be embodied in many different forms and should not be construed as being limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art. Is provided to fully convey the scope of the invention to those skilled in the art, and the invention is only defined by the scope of the claims. Like reference numerals refer to like elements throughout the specification.

The terminology used herein is for the purpose of illustrating embodiments and is not intended to be limiting of the present invention. In the present specification, the singular form includes plural forms unless otherwise specified in the specification. It is noted that the terms " comprises "and / or" comprising ", as used herein, do not exclude the presence or addition of one or more other elements, steps and operations.

FIG. 1 is a block diagram of a device control apparatus using palm-based user recognition information according to an embodiment of the present invention. FIG. 2 is a block diagram showing a configuration of a learning / setting control pattern database unit according to an embodiment of the present invention FIG. 3 is a conceptual diagram illustrating a palm-based user recognition process according to an embodiment of the present invention, FIG. 4 is a conceptual diagram illustrating a user recognition-based device control module according to an embodiment of the present invention, FIG. 2 is a flowchart illustrating a device control method using palm-based user recognition information according to FIG.

The device control apparatus 100 using the palm-based user recognition information according to the embodiment of the present invention includes a palm detection sensor 110, an image processing unit 120, a storage unit 130, a palm feature region detection unit 140, Unit 150 and a learning / setting control pattern database unit 170.

The palm detection sensor 110 detects the touch of the palm and projects the light source, and collects and stores the reflected light of the light source projected from the light source through the image sensing unit (S100, S110).

The imaging section includes light source means and optical sensor means which is a CMOS image sensor.

A CMOS image sensor reads an image of a palm image with a signal having a high signal-to-noise ratio (SNR) by attaching an amplifier to each pixel.

The palm detection sensor 110 captures and acquires a palm image at a specific position by capturing and acquiring an entire palm image or specifying a region in a predetermined order according to division designation of palmprint.

That is, the palm detection sensor 110 divides the palm image into a plurality of regions having different sizes and positions at the time of capturing the palm image, and photographs the palm image.

The image processor 120 converts the palm image into the monochrome image data and improves the sharpness of the monochrome image data by using a histogram stretching algorithm to improve the sharpness of the monochrome image data at step S130.

6, the image processing unit 120 extracts the outline of the palm image data by applying a pre-beat mask in order to extract the outline of the black and white image data with improved sharpness (S140).

6 and 7, the image processing unit 120 reads the palm image data stored in the storage unit 130 and converts the 3 * 3 mask or the 5 * 5 mask into the original image for the contour extraction, The mask value is multiplied by each pixel value of the palmprint image, and the sum is stored in the storage unit 130 (S142, S144, and S146).

The image processing unit 120 allocates the center pixel a5 of the palm image data to which the mask is applied, and stores the palm image in which the outline is detected in the storage unit 130 (S148 and S149).

The image processing unit 120 performs a filtering process to remove a hull other than the palm, and performs median filtering to remove the hull (S150).

8 and 9, the image processing unit 120 reads the palmprint image data stored in the storage unit 130, separates the palmprint image data into 3 * 3 block sizes, Are arranged in the size of 9 blocks, and then the pixel values of the 9 blocks arranged are sorted in ascending order (S151, S152, S153, and S154).

The image processor 120 allocates the pixel value of the 5th block, which is an intermediate value among the pixel values sorted in the ascending order, as the center value of the block area divided by 3 * 3 (S155).

As shown in FIG. 9A, when 3 * 3 blocks are arranged in the size of 9 blocks and the values of the pixels are sorted in ascending order, '2,2,2,2,4,4,4,5,10' do. If the intermediate value 4 of the aligned pixels is assigned to the center value of the 3 * 3 block, the pixel value '10', which is a noise component, is removed as shown in (b) of FIG. 9 (S156).

The image processing unit 120 stores the palm image data in which the main palm becomes clear by the median filtering and the bar code is removed in the storage unit 130. [

The image processing unit 120 performs a binarization process to remove noise of the palm image data and improve connectivity of the pixels (S160).

The image processing unit 120 performs a dual critical processing method for further weakening the stronger edge of the palm stronger, thereby removing the noise around the palm through two threshold values.

The image processing unit 120 reads the palm image data, analyzes the histogram, and extracts the brightness value of the palm image.

The image processing unit 120 sets two threshold values of T1 and T2 and classifies the pixels of '2' connected to '1' by classifying them as 0,1 (T1) and 2 (T2) All of them are converted into '1', and 256 gray images are processed as binary images of 0 and 1.

After the binocular image data is binarized, the image processor 120 eliminates the fine lines of the binoculars and emphasizes the relatively thick binoculars, and stores the binocular image data in the binarizing unit 130 so as not to break the line.

The palm feature region detecting unit 140 divides the palm image data binarized in the image processing unit 120 into a plurality of blocks, performs a secondary binarization process, and analyzes the thinned palm image data, , Direction, number, branch, and the like are extracted and feature values of position coordinates, thickness, and brightness of the feature points are measured (S170). The method of extracting these feature points and feature values is based on a feature extraction algorithm based on SIFT (Scale Invariant Feature Transform).

The palm feature region detection unit 140 generates a feature vector for the feature points and the feature values of the palm image data.

The user recognition unit 150 calculates a first partial feature vector for feature points of the palm registered in the storage unit 130 and calculates a second partial feature vector using feature points and feature values of the palm image data.

The user recognition unit 150 calculates the vector similarity degree so as to determine the degree of similarity when the second partial feature vectors are subtracted from each other and the difference value is within the tolerance range, thereby performing user's palm matching (S180 ).

As shown in FIG. 3, the user authentication process is completed using the palm matching of the user.

The learning / setting control pattern database unit 170 includes a signal processing unit 172, a user profile unit 174, and an output signal processing unit 176.

The signal processor 172 receives user authentication information from the user recognition unit 150 through user palm matching and receives an operation control signal of the device control device 100 from the device input unit 160 in operation S190, .

The user profile unit 174 searches the control pattern set by learning based on the user information received from the signal processor 172 and the operation control signal.

The output signal processing unit 176 generates an output signal for controlling one or more control target devices 180 according to the control pattern detected by the control of the signal processing unit 172 and transmits the output signal to the control target device 180, 180 are automatically operated (S192). Here, the control target device 180 includes all electronic devices such as a smart TV, a mobile terminal, a tablet PC, an audio device, an air conditioner, and a washing machine in a smart home environment.

The control pattern learning and prediction between the operation of the user's device control apparatus 100 and the learning / setting control pattern database unit 170 are performed as shown in FIG.

The user profile unit 174 sets the user information of the user recognition unit 150 and the operation control signal of the device input unit 160 as input and sets the device output signal of the output signal processing unit 176 as output, The control pattern of the control target device 180 is learned using a general error back propagation algorithm.

The user profile unit 174 calculates an output based on pattern recognition of input data having a structure in which a plurality of components are connected with a neural network algorithm.

In other words, the learning / setting control pattern database unit 170 recognizes a user who has a specific palm and learns what hardware operation the user has performed in the device control apparatus 100, repeats the learning process, The user's operation pattern for the device 100 is learned.

The user profile unit 174 generates different control patterns according to the user information.

The user inputs to generate a new operation signal through the device input unit 160 when the automatic operation of the control target device 180 controlled by the control pattern is not desired.

The learning / setting control pattern database unit 170 learns a new control pattern according to a new operation control signal and stores the new control pattern in the user profile unit 174.

The user profile unit 174 sets the operation control signal received from the device input unit 160 as a control pattern corresponding to the user information and stores the control pattern.

When a plurality of output signals are present for the user information and the operation control signal received from the device input unit 160, the user profile unit 174 outputs the generated output values The control target device 180 is selected.

When receiving the user information and the operation control signal from the device input unit 160, the user profile unit 174 generates an output signal for controlling the control target device 180 according to a predetermined control pattern, If a new operation control signal is not received from the input unit 160, that is, the user determines that a desired operation has been performed, the generated value of the output signal is recalculated and updated by applying the compensation value.

If the generation of the control pattern is normally performed and it is determined that the operation is desired by the user, the generated value can be expressed by increasing the generated value by inserting the compensation value as shown in Equation (1) below.

Here, O 'denotes an occurrence value, O denotes an existing occurrence value, a denotes a weight value, and r denotes a compensation value.

For example, when the generation value of the operation control for turning on the MBC channel of the TV is 4, and the generation value of the operation control for turning on the KBS channel is 3, the generated value is high.

The user profile unit 174 registers the control target device 180 through the signal processing unit 172 and controls the use time of the control target device 180, The control pattern of the user including the correlation with the target device 180 is learned and stored. If the control pattern occurs more than a predetermined number of times, the control pattern can be weighted.

While the present invention has been described in connection with what is presently considered to be practical exemplary embodiments, it is to be understood that the invention is not limited to the disclosed embodiments, but, on the contrary, You will understand. It is therefore to be understood that the above-described embodiments are illustrative in all aspects and not restrictive.

100: Device control device
110: Hand detection sensor
120:
130:
140: palm feature region detecting section
150:
160:
170: learning / setting control pattern database part
172: Signal processor
174:
176: Output signal processor
180: Control target device

Claims (15)

A palm detection sensor for sensing the touch of the palm and capturing the palm image to acquire palm image data;
An image processor for filtering and pre-processing the palm image data;
A palm feature region detecting unit for analyzing the preprocessed palm image data and extracting the feature point data of the palm image data;
A user recognition unit for matching the extracted feature point data with the feature point data of the registered palm image data and performing user authentication through comparison of similarity; And
And an output signal for controlling the control target device in accordance with the control pattern learned in the user information and generating the output signal for controlling the control target device, And a learning / setting control pattern database unit for transmitting the learning /
The learning / setting control pattern database unit,
And a user profile section for calculating an occurrence value indicating the frequency at which the output signal is generated,
Wherein the user profile unit receives the user information and the operation control signal to generate an output signal for controlling the control target device according to a predetermined control pattern and then receives a new operation control signal from the device input unit for a predetermined time A control unit configured to apply a compensation value to the generated value and to apply a weight to the generated value when an output signal for controlling the control target device according to the control pattern occurs more than a predetermined number of times, Based on user identification information, which is calculated by applying both the compensation value and the compensation value. The method according to claim 1,
Wherein the learning / setting control pattern database unit receives a new operation control signal from the device input unit when different control patterns are generated according to the user information and the operation of the control target device is performed by an operation that the user does not want, A device control device using palm-based user recognition information in which a new control pattern is learned and set according to a new operation control signal. The method according to claim 1,
Wherein the learning / setting control pattern database unit sets and stores an operation control signal received from the device input unit in a control pattern corresponding to the user information. The method according to claim 1,
Wherein the learning / setting control pattern database unit is configured to control the learning / setting control pattern database unit based on the user information and the operation control signal received from the device input unit, when a plurality of output signals exist, A device control device using palm-based user identification information. delete The method according to claim 1,
Wherein the palm detection sensor captures and acquires a palm image at a position of a specific region by designating the region in a predetermined order in accordance with the division designation of the palm when photographing the palm image, A device control device using palm-based user recognition information for extracting minutia data of palm image data. The method according to claim 1,
The learning / setting control pattern database unit registers the control target device to match the user information, and learns the control pattern of the user including the use time of the control target device, the type of input / output command, Device control device using palm-based user identification information. The method according to claim 1,
The image processor converts the palm image data into monochrome image data, extracts the contour of the palm image data by applying a prewitt mask to extract the contour of the monochrome image data, Based user identification information for performing median filtering and binarizing the filtered palm image data. Detecting a palm contact by the palm detection sensor and capturing a palm image to obtain palm image data;
Filtering the obtained palm image data and preprocessing the palm image data;
Extracting feature point data of the palm image data by analyzing the preprocessed palm image data by a palm feature region detecting unit;
The user recognition unit matching the extracted feature point data with the feature point data of the registered palm image data and performing user authentication through the similarity comparison;
The learning / setting control pattern database receiving the user information that has performed the user authentication and the operation control signal from the device input unit; And
Generating and outputting an output signal for controlling the control target device in accordance with the control pattern set by the learning / setting control pattern database learned in the user information, and transmitting the output signal to at least one control target device; And
The learning / setting control pattern database unit calculating an occurrence value indicating the frequency at which the output signal is generated,
Wherein the generated value is generated by receiving the user information and the operation control signal and generating an output signal for controlling the control target device according to a predetermined control pattern and then not receiving a new operation control signal from the device input unit for a predetermined time The compensation value is applied, and when an output signal for controlling the control target device according to the control pattern occurs more than a preset number of times, a weight is applied, and both the weight and the compensation value are applied to the existing generated value A device control method using palm-based user identification information. 10. The method of claim 9,
The method of claim 1,
And a control unit configured to receive a new operation control signal from the device input unit when the operation of the control target apparatus is an operation that is not desired by the user and to generate a new control pattern according to the new operation control signal, The step of learning and setting
Based user identification information, which further comprises a palm-based user identification information. 10. The method of claim 9,
The method of claim 1,
When the plurality of output signals exist for the user information and the operation control signal received from the device input unit, selecting the control target device in the ascending order of the generated values of the plurality of output signals
A device control method using palm-based user identification information. delete 10. The method of claim 9,
Dividing the palm image into a plurality of regions having different sizes and positions and designating the region in a predetermined order in accordance with the division designation of the palm when photographing the palm image; And
Extracting minutiae points of each palm image data for each of the divided regions
A device control method using palm-based user identification information. 10. The method of claim 9,
The method of claim 1,
Setting an operation control signal received from the apparatus input unit as a control pattern corresponding to the user information and storing
A device control method using palm-based user identification information. 10. The method of claim 9,
The method of claim 1,
Registering the control target device so as to match the user information and learning and storing a control pattern of the user including the use time of the control target device, the type of input / output command, and the relationship with other control target devices
A device control method using palm-based user identification information.

Images