KR101249680B1 - Apparatus for fingerprint recognition and adaptive method for capturing and processing image using thereof - Google Patents

Abstract

PURPOSE: A fingerprint recognition device and an adaptive image acquisition and a processing method therefor are provided to acquire high quality images by applying image processing algorithms to each captured mode. CONSTITUTION: An area detecting unit(100) divides an input area according to a detecting result by detecting an image correcting pad(120) formed on a fingerprint input device. A mode setting unit(102) sets an image capture mode in each input area. An image processing unit(104) captures an image acquired by the fingerprint input device according to the image capture mode. The image correcting pad is membrane including a silicon pad arranged on a platen in an input window of the fingerprint device. The area detecting unit divides the input area into an effective area and a normal area. [Reference numerals] (10) Finger print recognition device; (100) Area detecting unit; (102) Mode setting unit; (104) Image processing unit; (106) Image display unit; (108) Finger print recognition unit; (12) Finger print input device; (120) Image correcting pad;

Description

Fingerprint Recognition Apparatus and Adaptive Image Acquisition and Processing Method {Apparatus for fingerprint recognition and adaptive method for capturing and processing image using equivalent}

One aspect of the present invention relates to a user authentication technique using fingerprint recognition, and more particularly, to an image acquisition and processing technique for fingerprint recognition.

In the security and security system technology, a method of using biometric information is used as a means for authenticating the user's legitimacy of a building or a user of a specific device. Fingerprint information may be used as biometric information for authentication. Such a biometric information-based authentication system is used instead of the password method or ID card method as a method for protecting information or personal identification as well as authenticating the user of the building.

In order to improve performance of acquiring a fingerprint input through the fingerprint input device, a physical pad for image correction may be provided in an input window of the fingerprint input device. However, it is difficult to obtain a consistent high quality image by the image correction pad or the image processing time may be delayed by the corresponding region.

Prior art Korean Patent Publication No. 2009-0131757, the effective touch discrimination technology of the image processing apparatus having a touch input unit, US Patent Publication No. 20100113952, a technique for identifying a fake fingerprint, Japanese Patent Laid-Open No. 2005-063107 Techniques for facilitating fingerprint input using state and non-transmissive state changes, respectively, are disclosed.

According to an aspect, a fingerprint recognition device and an adaptive image acquisition and processing method using the same are proposed to secure a consistent high quality image and reduce image processing time with or without an image correction pad.

According to an aspect of the present invention, there is provided a fingerprint recognition device comprising: an area detector configured to detect an image correction pad formed on a fingerprint input device, and classify an input area according to a detection result; And an image processing unit for capturing an image acquired through the fingerprint input device according to the image capture mode set for each input area.

According to another aspect, there is provided a method of acquiring and processing an image of a fingerprint recognition device, the method comprising: detecting an image correction pad formed on a fingerprint input device and initializing the input area according to a detection result when the device is initialized; And acquiring an image through a fingerprint input device according to an image capture mode set for each input region, and capturing the acquired image to obtain a final image that is an image capture result.

According to an embodiment, the image correction pad formed on the fingerprint recognition device may be automatically detected, and the effective area of the image correction pad may be adaptively classified according to the detection result. Optimized image processing algorithms can be applied. In particular, it is possible to set different capture modes for the effective area and the non-image correction pad area and to apply an image processing algorithm for each set capture mode, thereby obtaining the best image quality. In addition, it saves time and internal usage resources for the entire image processing.

Furthermore, it is easy to detect whether or not the image correction pad is attached or detached through the image correction pad automatic detection process, so that different image processing algorithms can be applied to each input area. It is possible to provide a final image that maintains optimal quality whether or not it is within.

Furthermore, as the image correction pad is detected in the fingerprint input device connection process step or before the image capture process, image correction may be separately performed on the captured image in the fingerprint recognition process step through image registration and feature extraction after capture image acquisition. no need.

1 is a reference diagram illustrating an image correction pad formed on a fingerprint input device according to an embodiment of the present invention;
2 is a block diagram of a fingerprint recognition system according to an embodiment of the present invention,
3A and 3B illustrate an image correction pad effective area through a physical shape in a vertical (height) direction and a gray level of an input image when a silicon pad is formed on a platen according to an embodiment of the present invention. A reference diagram illustrating the process for
4A and 4B illustrate an image correction pad effective region through a physical shape in a horizontal (width) direction and a gray level of an input image when a silicon pad is formed on a platen according to an embodiment of the present invention. A reference diagram illustrating the process for
5 is a reference diagram illustrating an example of distinguishing an input area according to an embodiment of the present invention;
6 is a reference diagram illustrating an example of distinguishing an input region according to another embodiment of the present invention;
7 is a flowchart illustrating an image acquisition and processing method of a fingerprint recognition device at device initialization according to an embodiment of the present invention;
8 is a flowchart illustrating an image acquisition and processing method of a fingerprint recognition device when an image is acquired after device initialization according to an embodiment of the present invention;
9 is a flowchart illustrating a process of processing an image acquired through a fingerprint input device according to an image capture mode set for each input area according to an embodiment of the present invention;
10 is a block diagram of a fingerprint recognition system according to another embodiment of the present invention;
11 is a reference diagram illustrating a fingerprint input device input window for explaining an image correction pad detection process.

Hereinafter, with reference to the accompanying drawings will be described embodiments of the present invention; In the following description of the present invention, a detailed description of known functions and configurations incorporated herein will be omitted when it may make the subject matter of the present invention rather unclear. In addition, the terms described below are defined in consideration of the functions of the present invention, and this may vary depending on the intention of the user, the operator, or the like. Therefore, the definition should be based on the contents throughout this specification.

1 is a reference diagram illustrating an image correction pad 120 formed in a fingerprint input device according to an embodiment of the present invention.

Referring to FIG. 1, an image correction pad 120 is formed on a platen 122 of a fingerprint input device. The image correction pad 120 is for improving performance of acquiring a fingerprint input through the fingerprint input device. For example, by using the adsorption property of the image correction pad 120 to improve the acquisition performance for the dry fingerprint, to prevent the sliding phenomenon when the roll (roll) fingerprint acquisition, and to use a coating of a hard material such as a seal using elasticity Get the image.

The image correction pad 120 may be a silicon pad. The silicon pad is a thin membrane made of silicon that is placed on the platen 122 of the fingerprint input device in order to obtain a fingerprint image from a dry finger. Hereinafter, although the image correction pad 120 is limited to the silicon pad for ease of explanation of the present invention, the image correction pad 120 can be any type that can improve the acquisition performance of the input image in addition to the silicon pad. Do.

When the silicon pad is placed on the platen 122, the gray level characteristic of the portion is changed. For example, as shown in FIG. 1, a region in which the silicon pad is located tends to be dark, and a shadow region may be formed at the bottom of the silicon pad due to the thickness of the silicon pad. Such changes may vary in the extent, range, and shape of the silicon pad due to physical properties of the silicon pad, that is, thickness, transparency, and structural characteristics. The present invention utilizes such physical properties to detect silicon pads.

2 is a block diagram of a fingerprint recognition system according to an embodiment of the present invention.

Referring to FIG. 2, the fingerprint recognition system includes a fingerprint input device 12 and a fingerprint recognition device 10.

In the case of the optical type, the fingerprint input device 12 shoots strong light onto the platen and reflects the shape of the object such as a fingerprint or a painting on the platen, and the image of the reflected object passes through the high refractive lens to electrically transmit light. It is input to the fingerprint recognition device 10 which is a device made to be converted and read.

The fingerprint recognition device 10 may include an area detection unit 100, a mode setting unit 102, and an image processing unit 104, and may further include an image display unit 106 and a fingerprint recognition unit 108.

The fingerprint recognition device 10 adaptively detects the environment of the image correction pad 120, for example, a silicon pad, through the area detection unit 100, and sets the image capture mode according to the adaptive environment detection mode. The image processing algorithm is applied to each capture mode through the image processor 104. Accordingly, the quality of the final image, which is a capture result of the input image, may be maintained in an optimal state regardless of the presence or absence of the image correction pad 120.

The area detector 100 detects the image correction pad 120 formed on the fingerprint input device 12 and classifies the input area according to the detection result. The input area refers to a physical pixel area in the input window provided for capturing an image input through the input window of the fingerprint input device 12 and processing the captured image. The input area may be physically affected by the image correction pad 120 according to the system configuration after the fingerprint input device 12 is shipped. In this case, the quality of the captured image for fingerprint recognition is different depending on the presence or absence of the image correction pad 120. According to the present invention, the final image is always optimal regardless of the region in which the image correction pad 120 is included or not. To maintain the quality of the product.

According to an exemplary embodiment, the area detector 100 detects the image correction pad 120 by using gray level characteristics that change as much as the raised area when the image correction pad 120 is placed on the input window of the fingerprint input device 12. do. An example of detecting the image correction pad using the gray level of the area detector 100 will be described later with reference to FIGS. 3 and 4. On the other hand, using the gray level when detecting the image correction pad is only one embodiment of the present invention and other various embodiments are possible.

According to an exemplary embodiment, the area detector 100 detects the image correction pad 120 when the fingerprint input device 12 is connected and when the image capture unit 104 starts capturing through the image processing unit 104 after the fingerprint input device 12 is connected. The input area is classified according to the detection result. The input area includes an image correction pad effective area, a general area, and an unprocessable area. For example, the image correction pad effective area may be a painting input area, and the general area may be a general fingerprint input area. Alternatively, the image correction pad effective area may be a dry fingerprint input area, and the general area may be a wetland door input area. Alternatively, the image correction pad effective area may be a roll image input area, and the general area may be a flat image input area.

The mode setting unit 102 sets the image capture mode for each input area divided by the area detection unit 100. Image capture mode is a different image capture process for each input area. For example, in the painting input area mode, the image capture process is a series of processes for correcting the painting contrast, performing image binarization, and then performing image post-processing to obtain a final image. In the dry fingerprint input area mode, the image capture process is a series of processes of processing the image for improving the dry fingerprint, correcting the dry fingerprint contrast, and performing image post-processing to obtain a final image. An example of a process of processing an image for each image capture mode will be described later with reference to FIG. 9.

On the other hand, throughout the present specification, the image capture mode is understood as a comprehensive concept that changes not only the hardware settings necessary for image acquisition depending on the mode setting, but also the software settings required for image processing for the acquired image. Should be.

The image processor 104 captures an image acquired through the fingerprint input device 12 according to an image capture mode set for each input area through the mode setting unit 102. That is, the image processor 104 acquires images through the input window, and obtains a captured image in the input area according to the image capture mode set through the mode setting unit 102 among the acquired images. One of the acquired images becomes a capture image. The captured image is the final image output through the image capture of the image processing unit 104, but is also an image received by the fingerprint recognition unit 108 to be described later for feature extraction and fingerprint recognition. Images acquired through the image processor 104 and captured images may be displayed on the screen in real time through the image display unit 106.

The fingerprint recognition unit 108 extracts a feature from the captured image acquired through the image processing unit 104, analyzes the extracted feature, and recognizes a user's fingerprint through comparison with previously registered fingerprint information. The fingerprint recognition unit 108 receives the captured image from the image processing unit 104, and since the feature extraction and fingerprint recognition processes performed by the fingerprint recognition unit 108 correspond to general fingerprint recognition techniques, a detailed description thereof will be omitted.

3A and 3B illustrate an image correction pad effective area through a physical shape in a vertical (height) direction and a gray level of an input image when a silicon pad is formed on a platen according to an embodiment of the present invention. It is a reference figure for demonstrating the process for this.

When the silicon pad 120 is placed on the platen 122, an image of an aspect as illustrated in FIG. 3A may be obtained. The area in which the silicon pad 120 is placed tends to be slightly darker than the area in the normal platen 122, which becomes darker as the thickness of the silicon pad 120 becomes thicker. In addition, the lower band 120a of the silicon pad 120 generates a black stripe region proportional to the thickness of the pad.

Referring to FIG. 3B, the input area is divided into an image correction pad effective area A and an unprocessable area B corresponding to an area where the silicon pad 120 is located based on the gray level of the input image. The image correction pad effective area A shows a slightly lower gray level than the area in the normal platen 122, and the unprocessable area B is the lowest gray level area.

4A and 4B illustrate an image correction pad effective region through a physical shape in a horizontal (width) direction and a gray level of an input image when a silicon pad is formed on a platen according to an embodiment of the present invention. It is a reference figure for demonstrating the process for this.

4A and 4B, when the silicon pad 120 is formed on the platen 122, an image in the horizontal direction may be obtained as shown in FIG. 4A. Referring to FIG. 4B, in the horizontal direction, the image correction pad effective area A having a gray level slightly lower than that in the general platen 122 may be detected as in the vertical direction.

5 is a reference diagram illustrating an example of distinguishing an input region according to an exemplary embodiment of the present invention.

Referring to FIG. 5, the gray level of the input image is analyzed in the vertical (height) and horizontal (width) directions through the image correction pad automatic detection process, and when the analysis results are collected, the image correction pad as shown in FIG. The effective area A can be detected. In the image correction pad effective area A, an image capture mode suitable for the image correction pad is set by the mode setting process, and image processing is performed by the image processing process.

6 is a reference diagram illustrating an example of distinguishing an input area according to another exemplary embodiment of the present invention.

Referring to FIG. 6, in the image correction pad effective area A, the acquisition performance of the dry fingerprint is improved. This is because of the adsorption property of the image correction pad, and due to such characteristics, it is possible to obtain an effect of preventing slip when the roll fingerprint is acquired. In addition, by using the elasticity of the image correction pad, it is possible to acquire a surface image of a hard material such as a seal. In this way, the combination of the fingerprint input device and the image correction pad allows various images to be acquired according to the purpose of use. When the effective area of the image correction pad is automatically detected, the image acquisition mode for each region may be separately set and used. .

For example, in the case of a two-fingerprint input device, when the silicon pad is placed on the platen, an image correction pad effective area A as shown in FIG. 5 can be obtained. As described above, the image correction pad effective area A and the general area B can be obtained. In this case, an image capture mode may be set for each input area. That is, the image capture mode as shown in Table 1 below can be set.

Example of utilization
Silicon Pad Effective Area General area Painting input solution
Dojo Acquisition Mode 1 acquisition mode Fingerprint input device
Dry Fingerprint Mode
Normal Fingerprint, Wetland Gate Acquisition Mode Single Roll Image Acquisition Mode
1 flat image acquisition mode

7 is a flowchart illustrating an image acquisition and processing method of a fingerprint recognition device upon device initialization according to an embodiment of the present invention.

2 and 7, the fingerprint recognition device 10 detects an image correction pad 120 formed on the fingerprint input device 12 when the fingerprint input device 12 is connected, and inputs the result according to the detection result. The image capturing mode is divided and an image capture mode is set for each input input region (710). Subsequently, an image is acquired 720 according to an image capture mode set for each input area, and the obtained image is processed 730 to obtain a final image that is an image capture result (740).

8 is a flowchart illustrating an image acquisition and processing method of a fingerprint recognition device when an image is acquired after device initialization according to an embodiment of the present invention.

Referring to FIG. 8, when the image acquisition process is initialized after the fingerprint input device 12 is initialized, the fingerprint input device 12 detects an image correction pad 800 and acquires an image through the fingerprint input device 12 ( 810). Subsequently, it is determined whether the divided input area is an image correction pad effective area (820). If the divided input area is an image correction pad effective area, the image is processed according to the image correction pad effective area mode (830). If not, the image is processed 840 according to the general area mode to obtain a final image (850).

Accordingly, the effective area of the image correction pad existing in the platen and the general area can be adaptively distinguished, and thus optimized algorithms may be applied differently to objects such as a finger and a seal input to the corresponding area. In this case, the final image having the optimal quality may be obtained using the image correction pad autodetection result. For example, if a silicon pad is placed on the platen during use or the silicon pad being used is removed, the processing algorithm for the acquired image is applied internally without any additional image capture mode or setting change. It is always possible to obtain a final image with optimal quality.

9 is a flowchart illustrating a process of processing an image acquired through a fingerprint input device according to an image capture mode set for each input area according to an embodiment of the present invention.

2 and 9, when the fingerprint recognition device 10 acquires an image through the fingerprint input device 12 (900), the fingerprint recognition device 10 determines whether the acquired image is an image acquired in the effective area of the image correction unit (FIG. 902). As a result of the determination, if the image is acquired in the image correction unit effective area, it is determined whether the image capture mode is the painting input area mode (904). As a result of the determination, in the painting input area mode, the painting contrast correction 906, image binarization 908, and image post-processing 910 are performed to obtain a final image (912). Image binarization 908 may use gray level binarization. Image post-processing 910 includes noise removal.

In contrast, when the image capture mode is the general fingerprint input area mode, the final image is obtained by performing the contrast correction 914 and the image post-processing for the general fingerprint (912). In addition, when the acquired image is not an image acquired in the image correcting unit effective area, it is determined whether or not the dry fingerprint input area mode (916), and if the dry fingerprint input area mode, the image processing for dry fingerprint (918), the dry fingerprint cone Last correction is performed 920 and image post-processing to obtain a final image (912). On the other hand, the above-described embodiment is only one embodiment to help understanding of the present invention, it is apparent that various other embodiments are possible.

10 is a block diagram of a fingerprint recognition system according to another embodiment of the present invention.

The configuration of the fingerprint recognition system with reference to FIG. 10 is that the pad detection unit 110 is added as compared to the fingerprint recognition system with reference to FIG. 2.

The pad detector 110 detects a predetermined image correction pad among a plurality of image correction pads formed in the fingerprint input device 10. In this case, the pad detection unit 110 detects a predetermined image correction pad by detecting a preset pattern for each image correction pad when generating an input signal through the fingerprint input device 10. The preset pattern for each image correction pad is unique in its thickness and characteristics.

The mode performing unit 112 performs an image processing mode allocated to a predetermined image correcting pad detected by the pad detecting unit 110. In detail, the mode performing unit 112 includes an area detecting unit 100, a mode setting unit 102, an image capturing unit 103, and an image processing unit 104, and the functions thereof are the same as described above with reference to FIG. 2. Detailed description will be omitted. However, in addition to the configuration of FIG. 2, the image capture unit 103 obtains an image through the fingerprint input device according to the image capture mode set for each input area, and the image processor 104 acquires the image through the fingerprint input device. Performs a preset image processing mode for each input area for an image. Furthermore, the fingerprint recognition device 10 may further include an image display unit 106 and a fingerprint recognition unit 108 as described above with reference to FIG. 2.

11 is a reference diagram illustrating a fingerprint input device input window for explaining an image correction pad detection process.

10 and 11, an image processing embodiment in which four different image correction pads 1100, 1110, 1120, and 1130, for example, silicon pads are mounted on a single fingerprint input device 12 is described. Shows. Each of the mounted silicon pads is printed or drawn with distinct patterns in the pads, and the thickness and characteristics thereof may be different. The specific pattern of each image correction pad is set in advance, and the fingerprint recognition system has information about the patterns of the image correction pads in advance.

In the state where each area is separated by using the pattern formed on the image correction pad, when a user input signal is detected on the image correction pad, the fingerprint recognition device 10 may perform a specific image processing function or mode assigned to the area having the corresponding pattern. The operation method is applied differently depending on the function or mode. That is, the fingerprint recognition device 10 assigns a specific function or mode to the corresponding area by using the distinguishable patterns of the image correction pad, and detects a user input signal on a predetermined pad among the image correction pads. Process the assigned function or mode.

The embodiments of the present invention have been described above. Those skilled in the art will appreciate that the present invention can be implemented in a modified form without departing from the essential features of the present invention. Therefore, the disclosed embodiments should be considered in an illustrative rather than a restrictive sense. The scope of the present invention is defined by the appended claims rather than by the foregoing description, and all differences within the scope of equivalents thereof should be construed as being included in the present invention.

10: fingerprint recognition device 12: fingerprint input device
100: area detection unit 102: mode setting unit
104: image processing unit 106: image display unit
108: fingerprint recognition unit 120: image correction pad
110: pad detection unit 112: mode execution unit

Claims (15)

An area detector which detects an image correction pad formed on the fingerprint input device and classifies the input area according to a detection result;
A mode setting unit for setting an image capture mode for each input area divided by the area detection unit; And
An image processor configured to capture an image acquired through the fingerprint input apparatus according to an image capture mode set for each input area;
Fingerprint recognition device comprising a. The method of claim 1, wherein the image correction pad,
And a membrane comprising a silicon pad placed on a platen in an input window of the fingerprint input device. The method of claim 1, wherein the area detection unit,
And an image correcting pad effective area and a general area among the input areas according to the image correcting pad detection result. The method of claim 3, wherein
The image correction pad valid region is at least one of a paint input region, a dry fingerprint input region, and a roll image input region.
And the general area is at least one of a general fingerprint input area, a wetland door input area, and a flat image input area. The method of claim 3, wherein the area detection unit,
The gray level is measured with respect to at least one image input through the fingerprint input device, and the input area is at least one of an image correction pad active area, a general area, and an unprocessable area according to the gray level changing for each position of the input area. Fingerprint recognition device characterized in that divided into areas. The method of claim 5, wherein the area detection unit,
And a gray level is measured in the vertical direction and the horizontal direction of the fingerprint input device, and the measurement results are collected to distinguish the input area. The method of claim 1, wherein the area detection unit,
And the image correcting pad is detected when the fingerprint input device is connected and when the image is acquired after the fingerprint input device is connected. The method of claim 1,
An image display unit displaying the captured image in real time;
Fingerprint recognition device further comprising. The method of claim 1, wherein the fingerprint recognition device
A fingerprint recognition unit which extracts a feature from the captured image, analyzes the extracted feature, and recognizes a user's fingerprint through comparison with pre-registered fingerprint information;
Fingerprint recognition device further comprising. A pad detector configured to detect a predetermined image correction pad among a plurality of image correction pads formed in the fingerprint input device; And
A mode execution unit that performs an image processing mode assigned to the detected image correction pad;
Fingerprint recognition device comprising a. The method of claim 10, wherein the pad detector,
And a predetermined image correction pad is detected when the user input signal is generated through the fingerprint input device by detecting a preset pattern for each image correction pad. The method of claim 11,
Fingerprint recognition device, characterized in that the predetermined pattern for each image correction pad is unique in thickness and characteristics. The method of claim 11, wherein the mode performing unit,
An area detector configured to classify an input area based on the detected image correction pads when a predetermined image correction pad is detected by the pad detector;
A mode setting unit for setting an image capture mode for each input area divided by the area detection unit; And
An image capture unit for acquiring an image through the fingerprint input device according to an image capture mode set for each input area; And
An image processing unit configured to perform a preset image processing mode for each input area with respect to an image acquired through the fingerprint input device;
Fingerprint recognition device comprising a. In the image acquisition and processing method of the fingerprint recognition device,
Detecting an image correction pad formed on the fingerprint input device when the device is initialized and classifying the input area according to the detection result;
Setting an image capture mode for each of the divided input regions;
Acquiring an image through the fingerprint input device according to an image capture mode set for each input area; And
Capturing the obtained image to obtain a final image that is an image capture result;
Image acquisition and processing method comprising a. 15. The method of claim 14,
Detecting an image correction pad formed in an input window of a fingerprint input device when the image acquisition process is initialized after the device initialization;
Acquiring an image through the fingerprint input device; And
Capturing the acquired image according to an image capture mode set for each input area to obtain a final image that is an image capture result;
Image acquisition and processing method further comprising.

Images