JPH10255051A - Palm print detection processing method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To detect a palm print of desired precision in a short time. SOLUTION: A camera photographs a depressed palm, the contour of a palm print image that is photographed by the camera is extracted (A4) when the palm print image is displayed on a display part, the inside of the contour is divided into blocks of the prescribed number of pixels (A5), each block is compared with a paint out decision threshold and a thinning decision threshold, whether it is a paint out block or not (A6) and whether it is a thinning block or not are decided (A7), and blocks of decision results are displayed together with a palm print image on a display part (A9). After that, the palm print images of paint out blocks and thinning blocks are separately undergone such processing as to lower a trough line density or to raise a ridge line density.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a palm print processing method for processing and collecting a palm print image picked up by a camera. Like fingerprints, palm prints are also used for personal identification. This palm print has a larger area than the fingerprint,
It is desired that a clear image can be easily obtained in the palm print collection.

[0002]

FIG. 12 is a schematic perspective view of a conventional example.
Reference numeral 1 denotes an impression portion; 52, a light source such as a light emitting diode or an incandescent lamp; 53, a camera such as a CCD camera;
Is a mounting plate 54 on the surface of which a transparent elastic thin film or the like is formed.
And a case where the reflection film 56 is formed by a prism 55 formed at one end. The palm is placed on the placement plate 54, pressed, and illuminated by the light source 52, so that the path of the dashed line, that is, The palm print can be imaged by the camera 53 via the reflection film 56 of the prism 55, the reflection mirror 57, and the lens 58.

FIG. 13 is a flowchart of a conventional example of palm print collection, in which the palm of the subject is placed on the impression unit 51 (E).
1) The person in charge of collection adjusts the pressing force of the subject's palm on the imprinting part 51 (E2), illuminates with the light source 52, and images the palm print with the camera 53 (E3). This camera 53
(E4), and the person in charge of collection observes the displayed palm print image to determine whether or not there is blur (E5). The process proceeds to step (E2), where the person in charge of collection adjusts the pressing of the palm of the subject. By repeating such processing, when a palm print image without blurring is displayed, the palm print image is stored in the storage unit (E6).

[0004]

In the conventional example of palm print collection, the palm print area and the unevenness are large compared to the finger print area at the time of fingerprint collection.
There is a problem that three or more palmprint collection operations are required, and the time required for palmprint collection becomes long. In addition, when a re-collection operation is performed due to the presence of a blurred portion of the palm print image, the re-collection operation is performed only by the judgment of the person in charge of collection. SUMMARY OF THE INVENTION It is an object of the present invention to collect a palm print image with a desired accuracy in a short time without causing individual differences.

[0005]

According to the present invention, there is provided a palm print collecting method. (1) A palm pressed on an impression portion is illuminated by a light source, and the palm is imaged by a camera such as a CCD camera. In a palm print processing method including processing a signal by an image processing unit and displaying a palm print image on a display unit such as a CRT (cathode ray tube) display unit or a liquid crystal panel display unit, the outline of the palm print image captured by the camera is obtained. Is extracted, the inside of the contour is divided into blocks each having a predetermined number of pixels, and each block is compared with a crush determination threshold and a blur determination threshold,
It is determined whether the inside of the block is an unclear image due to crushing and whether the image is an unclear image due to dinging, and including the step of displaying the crushed block or the dingy block as a result of this determination on the display unit together with the palm print image, By increasing the pressing force on the blurred block side and decreasing the pressing force on the crushed block side, an optimal palm print image can be obtained.

(2) In the palm print processing method, a contour of a palm print image picked up by a camera is extracted, and the inside of the contour is divided into blocks each having a predetermined number of pixels. The number (or the number of valley pixels) is counted, and when the number of ridge pixels in each block is equal to or greater than the crush determination threshold (or the number of valley pixels is equal to or less than the crush determination threshold), the block is determined to be a crushed block, and When the number of pixels of the ridge line is equal to or less than the blur determination threshold (or the number of pixels of the valley line is equal to or greater than the blur determination threshold), the block is determined to be a blurred block, and the resulting block or blurred block is displayed together with the palm print image. It includes a process of displaying the information in the section.

(3) In the palm print collection processing method, a contour of a palm print image picked up by a camera is extracted, the inside of the contour is divided into blocks each having a predetermined number of pixels, and each block is crushed or blurred. Is determined by comparing with the collapse determination threshold value and the blur determination threshold value, and the density of the ridge is reduced (or the density of the valley line is increased) for the collapsed block, and the density of the ridge is increased for the blurred block. This includes a process of performing a process of lowering (or lowering the density of a valley line), and it is possible to correct a block-corresponding crush or blur by image processing and obtain an optimal palm print image.

(4) The palm pressed on the impression portion is illuminated by a light source, the palm is imaged by a camera, the image signal of the camera is processed by an image processing section, and a palm print image is displayed on a display section. In a palm print processing method including a process, a camera is provided with a shutter for controlling an incident light amount corresponding to a block,
The contour of the palm print image picked up by the camera is extracted, the inside of the contour is divided into blocks each having a predetermined number of pixels, and whether or not each block is crushed or blurred is compared with a crushing threshold and a blurring threshold. The method includes a step of controlling the opening and closing of the shutter corresponding to the block in accordance with the result of the determination. In the camera, the amount of incident light is reduced for the block of the crushed block, and the amount of incident light is increased for the block of the blurred block. By controlling the shutter corresponding to the block as described above, it is possible to obtain a palm print image without crushing or blurring.

[0009]

FIG. 1 is a flow chart of a first embodiment of the present invention, in which the palm of a subject is placed on an impression section (A1), and the palm of the subject is placed on the impression section. The person in charge of collection adjusts the pressure (A2), and captures a palm print with a camera (A
3) The contour of the palm print is extracted (A4), and the inside of the contour is divided into n × n blocks (A5).

Then, for each block, it is determined whether or not the pixels in the block are equal to or larger than a threshold x, that is, whether or not the inside of the block is equal to or larger than a threshold value for judging (A6). That is, it is determined whether the inside of the block is equal to or less than the blur determination threshold (A7). In accordance with each determination result, the crushed block or the dim block is displayed together with the captured palm print image (A9). Since the crushed block or the blurred block is displayed, the person in charge of sampling removes the palm of the subject by reducing the pressing force on the crushed block and increasing the pressing force on the blurred block (step A3).
By performing the subsequent processing, it is possible to obtain a palm print image without crushing and blurring, and the palm print image is stored in the storage unit (A8).

FIG. 2 is a functional block diagram of an embodiment of the present invention. Reference numeral 1 denotes a sampling unit similar to a conventional example, such as a prism, 2 denotes a light source such as a light emitting diode or an incandescent lamp, and 3 denotes a C light source.
A camera such as a CD camera, 4 is an image processing unit, 5 is a display unit such as a CRT or a liquid crystal panel, 6 is an image capturing unit, 7 is a storage unit such as a floppy device, a magnetic disk device, or a magneto-optical disk device. The contour extraction unit 9 is a division unit, and 10 is a determination unit.

The palm placed and pressed on the sampling unit 1 is
The image capturing unit 6 captures an image signal for one screen, converts the captured image signal into a digital image signal, and transfers the digital image signal to the image processing unit 4. The contour extracting unit 8 of the image processing unit 4 extracts the contour of the palm print image based on the captured image signal from the image capturing unit 6. The dividing unit 9 divides the inside of the contour into blocks of n × n pixels. Judgment unit 10
Determines whether each block is crushed or blurred, using the crushing determination threshold and the blurring determination threshold. The functions of each unit and other image processing functions in the image processing unit 4 can be realized by a configuration including a microprocessor and a memory.

The image processing section 4 binarizes the picked-up image signal from the image capturing section 6, for example, displays the palm print image on the display section 5, and blocks the judgment result of the judgment section 10 in the collapse. Alternatively, a blurred block is displayed, and when a palmprint image without a blurred or blurred block is displayed, the palmprint image is stored in the storage unit 7. In addition, together with the palm print image displayed on the display unit 5, for example, a crushed block is displayed in a red frame, and a dim block is displayed in a green frame. I do.

Therefore, the person in charge of collection indicates that if the crushed block is generated over the entire surface, the display content of the display unit 5 indicates that the overall pressing force is too large. On the other hand, if a faint block is formed over the entire surface, it indicates that the overall pressing force is insufficient, and the overall pressing force of the palm is increased. If the crushed block is unevenly distributed, reduce the pressing force of the palm on the crushed block side,
If the blurred blocks are unevenly distributed, the adjustment is made so as to increase the pressing force of the palm on the blurred blocks. This makes it possible to obtain a palm print image without crushed or blurred blocks.

FIG. 3 is an explanatory diagram of the palm print image. The outline of the palm print image 12 displayed on the display screen 11 is extracted. In this contour extraction, for example, a position where the luminance value of the pixel suddenly changes from the background portion can be determined as a contour by comparing the target pixel with surrounding pixels. Since the space between the fingers corresponds to the background portion above the palm print, this point can be used as the outline of the palm print. Regarding the lower part of the palm print, a position corresponding to a portion where the contour is greatly curved can be used as the contour.

Then, as shown by the dashed line, the image is divided into blocks 13 of n × n pixels. In this case, for example, n =
16 or n = 32 or the like. Then, for each block 13, it is determined whether the block is a crushed block or a blurred block using the crushing determination threshold and the blurring determination threshold, and the determination result is displayed.

FIGS. 4 and 5 are explanatory diagrams for explaining the contour extraction of the palm print image. For example, as shown in FIG.
, The boundary is found as indicated by a black dot based on the luminance difference between the background pixel and the palm print pixel. In such processing, a luminance difference between adjacent pixels along the scanning line 15 is obtained, and a point having a luminance difference equal to or more than a predetermined value can be set as a boundary. Alternatively, it is possible to find the boundary between the background and the palm print image by applying the means of the outline emphasis processing in the image processing.

Next, as shown in FIG. 4B, a boundary is found as indicated by a black dot on the basis of the luminance difference between the background pixel and the palm print pixel according to the scanning line 16 from the left side to the right side. Next, as shown in FIG. 5C, a boundary is found as indicated by a black point based on the luminance difference between the background pixel and the palm print pixel in accordance with the scanning line 17 that goes upward from below. Next, as shown in FIG. 5D, the scanning lines 18 from the right side to the left side.
, The boundary is found as indicated by a black dot based on the luminance difference between the background pixel and the palm print pixel.

By connecting the points of the boundary by the scanning lines 15 to 18, the outline 19 of the palm print image shown in FIG.
Can be extracted. Although the border points are indicated by white circles and the intervals are shown relatively large, the border points can be found at intervals corresponding to the resolution of the palm print image. Also, it is not necessary to connect the boundary points exactly,
The outline 19 can be formed by being connected in a straight line. And
The inside of the contour 19 is divided into blocks 20 each including a plurality of pixels as shown in FIG.

In this case, the block 20 needs to have a size including a plurality of ridges and valleys, respectively.
In order to detect a crushed or blurred area as detailed as possible, the size of the block 20 is reduced. For example, it is set to include two ridge lines and two valley lines.
Normally, the line width between the ridge line and the valley line is about 200 μm. For example, if the resolution is 50 μm, (200/5
0) × 2 × 2 = 16, and the block 20 has a size of 16 pixels × 16 pixels. The size of the outline 19 of the palm print image is relatively large among individuals,
Since it is about 100 mm x about 100 mm, the total number of pixels is 2
000 pixels × 2000 pixels. Then, as described above, when divided into blocks 20 of 16 × 16 pixels,
There are 15625 blocks.

In order to speed up the detection of a crushed or blurred area, it is better that the number of blocks is as small as possible. For example, the area can be divided into 100 blocks. In that case, the size of one block is 200 pixels × 2
00 pixels.

In the palm print image, when the luminance is represented by 256 gradations, the average luminance value of the ridge line is 200, and the average luminance value of the valley line is 20. Becomes As described above, assuming that the number of ridges and the number of valleys existing in one block are the same, the average luminance value of one block is (200 + 20) / 2 = 110. However, in the case of crushing, the average luminance value increases, and in the case of crushing, the average luminance value decreases.

Such a change in the average luminance value differs depending on the configuration of the optical system, the AD conversion parameters, and the like.
10% → blurred image, AB <110−10% → blurred image. That is, ± of the judgment average luminance value
When it exceeds the range of 10%, it can be determined that the image is crushed or blurred. Note that it is possible to determine only the pixel data of the ridge in the block, or to determine only the pixel data of the valley.

FIG. 7 is a flowchart of the second embodiment of the present invention, in which steps (B1) to (B5) and (B
8) and (B9), steps (A1) to (A5) and (A8) and (A9) in the flowchart shown in FIG.
Is the same as In this embodiment, it is determined whether or not the ridge data in the block is equal to or larger than the threshold value u (B6), and it is determined whether or not the ridge data in the block is equal to or smaller than the threshold value v (B7).

In this case, the luminance value (multi-valued image data) of the pixel of the captured palm print image is compared with a threshold value. For example, a pixel having a luminance value equal to or higher than the threshold value is defined as a valley line (white), and a pixel having a luminance value smaller than the threshold value is determined. Is determined as a ridge (black), the number of pixels constituting the ridge in the block is counted and used as ridge data, and the ridge data is compared with a collapse determination threshold u. Further, the ridge data is compared with a blur determination threshold value y. Then, the crushed block or the dim block is displayed on the display unit (B9).

In step (B6), the number of pixels forming the valley line in the block is counted to obtain valley line data, and this valley line data is compared with the collapse determination threshold value. Sometimes it is determined to be collapsed, and step (B7)
In step (b9), the valley line data is compared with the blurring determination threshold value, and when the blurring threshold value is exceeded, the blurring is determined, and the blurred block or the blurred block is displayed on the display unit according to each determination result (B9). .

FIGS. 8A and 8B are diagrams for explaining the determination of crushing and blurring in the second embodiment of the present invention. FIG. 8A shows that the ridges 22 and the valleys 23 in the block 21 are almost one-to-one. In the palm print image without crushing or blurring, the ridges 22 and the valleys 23 in the block are each 50% ± the area of the block 21 (the number of pixels).
j%.

FIG. 8B shows that the ridges 32 in the block 31 are 50% of the area (the number of pixels) of the block 31.
+ J% or more (the number of pixels) (ridge data in FIG. 4), and the valley line 33 is 50% of the area (the number of pixels) of the block 31 (valley line data in FIG. 4). % -J%
The case of the following area (number of pixels) is shown. Therefore, the threshold value u in step (B6) in FIG. 7 is set to 50% + j%, and whether or not the ridge 32 is collapsed is determined.
In the case of FIG. 8B, it is determined that the block 31 is a collapsed block. Similarly, when determining whether or not the valley line 33 is crushed, the crush determination threshold is set to 50% -j%. Therefore, in the case of FIG.
Since the area (the number of pixels) is 0% -j%, it is determined that the block 31 is a crushed block.

In the case of the block 41 shown in FIG. 8C, the ridge 42 has an area (the number of pixels) of 50% -j% or less of the area (the number of pixels) of the block 41. Line 4
3 is 50% + with respect to the area (the number of pixels) of the block 41
The case where the area (the number of pixels) is j% or more is shown. Therefore, FIG.
Assuming that the threshold value v of the step (B7) is 50% −j%,
It is determined whether or not the ridge 42 is blurred.
In FIG. 8C, the ridge 42 is 50% -j
%, The area (the number of pixels) is determined to be a faint block 41. When determining whether or not the valley line 43 is blurred, the blurring determination threshold is set to 50% +
j%, the block 4 shown in FIG.
1 is determined to be a faint block.

FIG. 9 is a flowchart of the third embodiment of the present invention. Steps (C1) to (C5) are described in steps (A1) to (A5) in the flowcharts of FIGS. This is the same as (B1) to (B5), and a duplicate description will be omitted. In step (C6), the presence or absence of blurring or crushing of the palm print image is determined. This determination step is performed in step (A) of FIG.
6), (A7) or steps (B6), (B7) in FIG.
Can be applied.

Then, the ridge density (the luminance value of the pixels forming the ridge) and the valley density (the luminance value of the pixels forming the valley) are extracted for the blurred or crushed block (C).
8) In the case of collapse, the density of valley lines is lowered, and in the case of blurring, the density of ridges is increased. Thereby, in the crushed block and the blurred block, the image processing is performed so that the density difference between the ridge line and the valley line is widened.

That is, in the collapsed block, the density of the ridge line and the valley line is large and the density difference is small.
Therefore, by lowering the density of the valleys and increasing the difference from the density of the ridges, the area of the collapse is narrowed. Similarly, in the blurred block, the density of the ridge and the valley are small and the density difference is small. Therefore, by increasing the density of the ridge, the difference between the density of the valley and the density is increased. , Make the blurred area relatively narrow. Therefore, a palm print image having an optimum density can be obtained by the image processing, and the palm print image can be loaded into the storage unit by one palm print collection operation (C7).

FIG. 10 is a flow chart of the fourth embodiment of the present invention. Steps (D1) to (D5) are described in steps (A1) to (A) of FIGS.
5), (B1) to (B5), and (C1) to (C5), and step (D6) is the same as step (C) in FIG.
Since it is the same as 6), duplicate description is omitted.

In this embodiment, a camera 3 (see FIG. 2)
Is provided with a liquid crystal shutter capable of controlling the amount of incident light corresponding to the block. FIG. 11 is a view for explaining the control of the liquid crystal shutter. 40 is a liquid crystal shutter, 46 and 47 are driving units for applying a voltage selectively controlled to the electrodes of the liquid crystal shutter, 45 is a liquid crystal shutter control unit, and 4 is an image processing unit. 2, a contour extraction unit, 9 a division unit, and 10 a determination unit, and the illustration of the camera 3, the image capture unit 6, and the like in FIG. 2 is omitted.

The liquid crystal shutter 40 has electrodes corresponding to blocks for dividing the palm print image into a plurality of blocks.
6 and 47, a voltage is selectively applied to control the amount of transmitted light corresponding to the block. In the initial state, the driving units 46 and 47 are controlled so that the amount of transmitted light of the block corresponding portion of the liquid crystal shutter 40 becomes the same. , A voltage to be applied to the electrodes is set, and for the crushing determination block in the determination unit 10, the driving units 46 and 47 are controlled from the liquid crystal shutter control unit 45 to apply the voltage to the electrodes corresponding to the crushing determination block. Is reduced to reduce the amount of transmitted light. Conversely, for the blurring determination block, the driving units 46 and 47 are controlled from the liquid crystal shutter control unit 45 to increase the voltage applied to the electrodes corresponding to the blurring determination block. Increase the amount of transmitted light.

That is, the image processing unit 4 determines whether the crush determination block or the blur determination block is
The voltage applied to the electrode of the liquid crystal shutter 45 corresponding to the block is controlled by the liquid crystal shutter control unit 45, and the block corresponding to the block is determined by closing the block corresponding portion of the liquid crystal shutter 40 by g% from the normal state. Reduce the brightness value of the pixels in the block. By opening the block corresponding portion of the liquid crystal shutter 40 corresponding to the blur determination block by h% from the normal state, the luminance value of the pixel of the blur block is increased (D8).

In this case, g% and h% can be selected in consideration of the luminous intensity of the light source and the characteristics of an optical system such as a lens. Then, the procedure proceeds to step (D3), and the palm print image is again captured, whereby an optimal palm print image can be obtained, and the palm print image is stored in the storage unit (D7). Therefore, usually, an optimal palm print image can be obtained by two palm prints. The size of the block obtained by dividing the palm print image is
The selection is made according to the electrode configuration of the liquid crystal shutter 45. For example, if it is divided into 100 blocks,
Each of the driving units 46 and 47 is configured to apply a voltage to ten electrodes, and the output voltages of the driving units 46 and 47 are controlled so that a desired voltage is applied to the electrode at the intersection.

The present invention is not limited to the above-described embodiments, but can be added and changed in various ways. For example, a palm print image stored in a storage unit can be transmitted to a center using a transmission device. It is also possible to transmit. Further, in the determination of crushing or blurring, the luminance values of the pixels in the block can be integrated for the ridges and valleys and compared with the determination threshold. Further, it is also possible to control the light source so as to correspond to a block so that the luminous intensity of the light source block corresponding to the crushed block is reduced and the luminous intensity of the light source block corresponding to the blurred block is increased.

[0039]

As described above, according to the present invention, the captured palm print image is divided into blocks each having a predetermined number of pixels, and each block is compared with a threshold value for determining whether each block is a collapsed block to determine whether the block is a collapsed block. Also, it is possible to determine whether or not the block is a fading block by comparing it with the fading determination threshold value, display the determination result together with the palm print image on the display unit, and give an instruction to determine the pressing adjustment by the person in charge of collection. , At most 2-3
An optimum palm print image can be obtained by the second palm print imaging process. In addition, there is an advantage that the displayed content does not include individual differences among collection persons, and that even a collection person who is not a proficient person can easily perform an optimal palm print collection process.

Also, by performing the correction processing of the density in the block by image processing in accordance with the determination result of the crushed block or the dim block, one palmprint imaging operation can be performed, so that the palmprint image can be collected in a short time. There are advantages that can be made possible. Further, by controlling the amount of transmitted light by the liquid crystal shutter corresponding to the blocks obtained by dividing the palm print image, there is an advantage that an optimum palm print image can be obtained in a short time without repeating the pressing adjustment.

[Brief description of the drawings]

FIG. 1 is a flowchart of a first embodiment of the present invention.

FIG. 2 is a functional block diagram of the embodiment of the present invention.

FIG. 3 is an explanatory diagram of a palm print image.

FIG. 4 is an explanatory diagram of contour extraction of a palm print image.

FIG. 5 is an explanatory diagram of contour extraction of a palm print image.

FIG. 6 is an explanatory diagram of contour extraction and division of a palm print image.

FIG. 7 is a flowchart according to a second embodiment of the present invention.

FIG. 8 is an explanatory diagram of judgment of crushing and blurring in the second embodiment of the present invention.

FIG. 9 is a flowchart according to a third embodiment of the present invention.

FIG. 10 is a flowchart according to a fourth embodiment of the present invention.

FIG. 11 is a control explanatory diagram of a liquid crystal shutter according to a fourth embodiment of the present invention.

FIG. 12 is a schematic perspective view of a conventional example.

FIG. 13 is a flowchart of palmprint collection in a conventional example.

[Explanation of symbols]

 (A1)-(A9) Step

Claims (4)

[Claims] 1. A process of illuminating a palm pressed on an embossing unit with a light source, capturing an image of the palm with a camera, processing an image signal of the camera by an image processing unit, and displaying a palm print image on a display unit. In the palm print collection processing method, a contour of a palm print image picked up by the camera is extracted, and the inside of the contour is divided into blocks each having a predetermined number of pixels, and each block is compared with a crush determination threshold and a blur determination threshold. And determining whether the inside of the block is an unclear image due to crushing and an unclear image due to dinging, and displaying the crushed block or the unshaded block as a result of the determination on the display unit together with the palm print image. A palm print collection processing method comprising: 2. An outline of a palm print image picked up by the camera is extracted, and the inside of the outline is divided into blocks each having a predetermined number of pixels, and the number of pixels of ridges (or the number of pixels of valleys) in the blocks is determined. Is counted, and when the number of ridge pixels in the block is equal to or greater than the crushing threshold (or the number of valley pixels is equal to or less than the crushing threshold), the block is determined to be a crushed block, and the number of pixels of the ridge is equal to the crushing threshold. The following (or the number of pixels of the valley line is equal to or greater than the blur determination threshold), the process includes a step of determining a blurred block and displaying the crushed block or the blurred block as a result of the determination on the display unit together with the palm print image. 2. The palm print collecting method according to claim 1, wherein: 3. A contour of a palm print image picked up by the camera is extracted, the inside of the contour is divided into blocks each having a predetermined number of pixels, and whether or not each block is crushed or blurred is determined by a crushing threshold and a blurring threshold. The density of ridges is reduced for a collapsed block (or the density of a valley is increased), and the density of a ridge is raised for a blurred block (or the density of a valley is reduced). 3. The palm print collection processing method according to claim 1, further comprising the step of: 4. A process of illuminating a palm pressed on an impression portion with a light source, capturing an image of the palm with a camera, processing a captured image signal of the camera by an image processing unit, and displaying a palm print image on a display unit. In the palm print processing method, the camera is provided with a shutter for controlling the amount of incident light corresponding to each block, an outline of a palm print image captured by the camera is extracted, and the inside of the outline is divided into blocks each having a predetermined number of pixels. Determining whether each block is crushed or blurred by comparing it with a crushing determination threshold and a blurring determination threshold, and controlling the opening / closing of the shutter corresponding to the block in accordance with the determination result. Palm print collection processing method.