JP2007249950A - Image acquisition device and image acquisition method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To easily acquire fine images by efficiently detecting the movement of a photographed object while attaining miniaturization without a movable part. <P>SOLUTION: This image acquisition device having an imaging element section comprising a plurality of pixels, acquires the whole image by imaging a plurality of times while relatively moving the photographed object. The imaging element section has a first region 51 of high resolution and a second region 52 of low resolution, and composes the whole image from partial images acquired in the first region 51 based on the moving information of the photographed object obtained from partial images acquired in the second region 52. Image acquisition in the second region 52 is intermittently performed between image acquisition in the first region 51. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to an image acquisition apparatus and an image acquisition method for acquiring an image by scanning an imaging region of a subject, and in particular, having an imaging element unit composed of a plurality of pixels and performing a plurality of times while relatively moving the subject. The present invention relates to an image acquisition device that captures an image and acquires an entire image. The image acquisition device is applied as a scanner device or a biometric authentication device.

  Digital processing of information is widely performed using a computer. Also in the field of image processing, when an object is imaged, an image is picked up by an image pickup device which is a photoelectric conversion device, and thereafter, recorded or displayed by digitized data. Digital processing of image information is excellent in rapidity, processability, and downsizing of the apparatus, and is widely used as an imaging apparatus such as a CCD camera.

  In recent years, in order to ensure security for personal information and confidential information, biometric authentication systems such as fingerprint authentication devices have attracted attention, and demand for office devices and portable devices is also increasing. Such a biometric authentication system using fingerprints, faces, irises, palm prints, etc. acquires a biometric image in an image acquisition device, extracts features from the acquired image, and collates with registered data based on that information To authenticate you.

  Here, as a detection method of the image acquisition device used in the biometric authentication system, there are an optical method using an imaging element such as a CCD or a CMOS sensor, a capacitance method, a pressure detection method, a thermal method, an electric field detection method, and the like. is there. As another classification, there are an area type that collects subject images in a batch using a two-dimensional area sensor and an imaging method called a sweep type. In the sweep type, an entire image is obtained from an image obtained by sequentially capturing the subject in the sub-scanning direction using a one-dimensional line sensor or a belt-like two-dimensional sensor having about 2 to 20 pixels in the sub-scanning direction (movement direction) of the subject. Synthesize and get.

  The sweep type method is required to be more troublesome for the subject to move the subject than the area type using the two-dimensional area sensor, but the device can be downsized and the manufacturing cost can be reduced. . Note that the method of fixing the subject and moving the image sensor unit is also a sweep type, but there is a disadvantage that the mechanism is complicated compared to the method of moving the subject.

  In a sweep type apparatus in which a subject is moved with respect to the image sensor unit, the following situation occurs with respect to image reading.

  20a and 20b are diagrams for explaining image acquisition by an image sensor unit using one line sensor (solid-state image sensor), and FIGS. 21a and 21b are image sensors using a plurality of line sensors. It is a figure for demonstrating the image acquisition by a part.

  20a and 20b are examples of a scanner using a one-dimensional sensor. FIG. 20a of one line sensor is a diagram illustrating a case where the longitudinal direction of the line sensor 111 is the main scanning direction and the moving direction of the subject is the sub-scanning direction, and the relative moving speed of both is sufficiently slow, and FIG. It is the figure which showed the case where a relative movement speed is quick.

  In FIG. 20a, the plurality of partial images acquired by the line sensor 111 have overlapping portions, and the entire image can be reconstructed. However, as shown in FIG. 20b, when the relative movement speed is increased, the movement speed is not known, and when the entire image is synthesized from the acquired partial images, the image expands and contracts.

  21a and 21b are examples of a fingerprint sensor using a two-dimensional sensor. FIG. 21A is a diagram illustrating a case where the relative movement speed of both the line sensors 112 is sufficiently slow when the longitudinal direction of the line sensors 112 is the main scanning direction and the moving direction of the subject is the sub-scanning direction. FIG. 21 b is a diagram showing a case where the relative movement speed is high. As in the case of FIG. 20b, imaging cannot catch up with the moving speed, and when the entire image is synthesized from the acquired partial images, the image expands and contracts. In particular, when the resolution is increased, the frame rate is lowered, making it difficult to catch up. If the frame rate is increased to avoid this, the clock speed must be increased, which leads to an increase in current consumption and an increase in cost. Further, higher resolution leads to increased IF circuit cost and IC area.

  In view of this, movement of a subject as a finger is detected and the information is used for image synthesis. An example of a sweep type method for detecting movement of a subject will be described below.

  Patent Document 1 discloses an example of a conventional sweep type image acquisition apparatus that moves a subject.

  FIG. 22 is a diagram illustrating a conventional sweep-type image acquisition device (fingerprint data reading device) disclosed in Patent Document 1. In FIG.

  The surface 142a of the finger 142 on the table 141 is irradiated with light from the light source 145 through the prism 141a. The light source 145 is composed of a line light source that irradiates one line. The reflected light from the surface 142a is received by an optical sensor 146 configured with a line optical sensor that converts the reflected light for one line into an electrical signal. The electrical signal converted by the optical sensor 146 is input to the image memory 148 via the A / D converter 147. On the other hand, the movement distance detection sensor 144 attached to the finger presser base 143 detects the movement distance of the finger that is moved substantially at right angles to the optical sensor 146.

  In the processing unit 149, two-dimensional fingerprint image data is stored in the image memory 148 based on the output of the movement distance detection sensor 144 and the output of the optical sensor 146.

  Patent Document 2 discloses another example of a conventional sweep type image acquisition apparatus that moves a subject.

  FIG. 23 is a diagram illustrating a conventional sweep-type image acquisition device (fingerprint detection device) disclosed in Patent Document 2. In FIG.

  The fingerprint detection device 151 optically detects the fingerprint 153 on the surface of the finger 152. The fingerprint 153 is detected by bringing the finger 152 into contact with the outer peripheral surface of the reading roller 154 and angularly displacing the reading roller 154 while moving it in the direction of the arrow 152a.

  Inside the fingerprint detection device 151, an optical system 155 and a movement detection unit 156 that detects the movement of the finger 152 by the angular displacement of the reading roller 154 are provided. The optical system 155 includes a light source 157 for irradiating light on the surface of the finger 152 in contact with the reading roller 154, and a line sensor 158 for imaging reflected light from the fingerprint 153. The line sensor 158 is a CCD configured by a plurality of light receiving elements arranged one-dimensionally. A two-dimensional image can be captured by moving the finger 152 in the sub-scanning direction intersecting the main scanning direction, for example, in a direction orthogonal to the main scanning direction, with the arrangement direction of the light receiving elements as the main scanning direction. The movement detector 156 detects the direction and distance of movement of the finger surface portion based on the angular displacement of the reading roller 154. The image data captured by the line sensor 158 is accumulated every time it is detected that it has moved by a predetermined distance in a certain direction.

  The one-dimensional image captured by the line sensor 158 is accumulated by the image synthesizing unit 159, synthesized with an image such as a two-dimensional fingerprint, and output.

  In addition, in response to the above problem, using the partial image of the subject itself, the movement information of the subject is detected from the relationship between the partial images before and after the movement, and the entire image is synthesized from the partial image based on the movement information. The method of doing is known. The movement information is obtained from the partial image itself used for image acquisition. Patent Document 3 discloses such a sweep type fingerprint sensor.

  FIG. 24 is a diagram showing a conventional sweep-type image acquisition device (fingerprint reading system) disclosed in Patent Document 3. As shown in FIG.

  The finger 161 slides, and partial images I0 to In of the fingerprint 162 at the instants t0 to tn are generated. The partial images I0 to In partially overlap and are correlated to obtain movement information and reconstruct it to obtain an entire fingerprint image.

In this method, since image acquisition and movement detection are performed by one sensor, there is an advantage that downsizing and cost reduction are possible.
JP-A-63-273976 JP 2001-184490 A JP 10-091769 A

  In the conventional sweep-type image acquisition device that detects the movement of the subject and uses the information for image synthesis, it can achieve a certain size reduction and easily acquire a good image. However, since it has a movable part mechanically, there is a limit to downsizing, and the number of parts and the manufacturing cost are increased.

  Further, in a sweep type apparatus, obtaining movement information of a subject from the relationship between the partial images of the subject and synthesizing the entire image based on the movement information are concerned with the following restrictions. .

  An example in which a plurality of line sensors 112 are used as a sweep type two-dimensional sensor will be described again with reference to FIGS. The longitudinal direction of the line sensor 112 is the main scanning direction, and the moving direction of the subject is the sub-scanning direction. FIG. 21A shows a case where the relative movement speed of the subject and the sensor (image acquisition device) is slow relative to the image acquisition interval, and FIG. 21B shows a case where the relative movement speed is high relative to the image acquisition interval. It is a figure.

  When the relative movement speed of the subject and the sensor is slower than the image acquisition interval, there is an overlapping region where the same subject portion has moved between the acquired partial images. For this reason, the relative movement information can be calculated from the partial images by obtaining the size of the overlapping region using the front and rear partial images.

  On the other hand, when the relative movement speed of the subject and the sensor is faster than the image acquisition interval, there is no overlapping region where the same subject portion has moved between the acquired partial images. For this reason, it is impossible to calculate the relative movement information from the partial images by obtaining the size of the overlapping region using the front and rear partial images. For this reason, since accurate movement information cannot be obtained, the image must be synthesized from an image obtained by guessing to some extent, and the image expands and contracts.

  In particular, when the resolution is increased, the amount of data increases, so the time for acquiring one image becomes longer and the image acquisition interval becomes longer. In addition, since the moving speed can be obtained even if the relative moving speed is high, the amount of data increases even when the number of rows in the sub-scanning direction is increased, resulting in a long imaging interval. turn into. As described above, there is a trade-off between the imaging capability with respect to higher resolution and faster subject movement speed.

  As described above, in a device that detects a moving speed using a two-dimensional sensor having no moving parts, high resolution does not provide a necessary image acquisition interval when the data communication time is long and the finger moving speed is high. There was a restriction that the image could not be synthesized.

  Therefore, the present invention provides an image acquisition apparatus that can acquire a high-resolution image by efficiently detecting the moving speed even when the moving speed of the subject is high, while achieving downsizing without moving parts. It aims to be realized.

  In order to solve the above-described problems, an image acquisition apparatus according to the present invention has an imaging element unit including a plurality of pixels, and acquires an entire image by performing multiple imaging while relatively moving a subject. The image sensor unit has a first region having a high resolution and a second region having a low resolution, and is based on subject movement information obtained from a partial image acquired in the second region. The entire image is synthesized from the partial images acquired in the first area.

  According to another aspect of the present invention, there is provided an image acquisition apparatus that includes an imaging element unit including a plurality of pixels and that captures a plurality of times while relatively moving a subject to acquire an entire image. The imaging element unit has a first region having a high resolution and a second region having a low resolution, and image acquisition in the second region is intermittent between image acquisitions in the first region. The entire image is synthesized from the partial image acquired in the first area based on the movement information of the subject obtained from the partial image acquired in the second area.

  According to the present invention, there is no movable part for obtaining the movement information of the subject, thereby achieving miniaturization. Even when the moving speed of the subject is high, the image and movement information of the subject can be efficiently detected from the high-resolution region and the low-resolution region of the image sensor unit, and a good image can be easily acquired.

[Outline of Embodiment]
In the present invention, attention has been paid to the fact that both the increase in resolution of an acquired image and the increase in the number of rows in the sub-scanning direction for obtaining movement information are in a trade-off relationship with shortening of the imaging interval. Further, it has been found that this trade-off relationship is caused by obtaining movement information using partial images themselves for synthesizing the entire image. Therefore, a partial image for synthesizing the entire image and a partial image for obtaining movement information are separated, and each is acquired at the optimal resolution and timing, so that the relative movement speed that can cope with higher resolution of the acquired image It has been found that both high speeds can be achieved.

  Since the moving speed is not obtained from the high-resolution partial image for obtaining the acquired image, the number of rows may be small. However, in order to synthesize the image, the imaging interval needs to be short. On the other hand, low-resolution partial images that calculate the moving speed need the number of rows in the sub-scanning direction to ensure image overlap even with fast movement, but the images are not synthesized, so the speed is always observed for each image capture. do not have to. Utilizing this fact, the following configuration was adopted.

  The image acquisition apparatus according to the first embodiment includes an imaging element unit including a plurality of pixels, performs imaging a plurality of times while relatively moving the subject, and obtains an entire image from a one-dimensional or two-dimensional partial image of the subject. It presupposes an image acquisition device to be synthesized. The imaging element unit includes a first area having a high resolution and a second area having a low resolution, and the first element is obtained based on the movement information of the subject obtained from the partial image acquired in the second area. The entire image is synthesized from the partial images acquired in the area.

  As a result, even when the moving speed of the subject is high, the moving speed can be detected efficiently, and a high-resolution image can be acquired.

  The image acquisition apparatus according to the second embodiment includes an imaging element unit including a plurality of pixels, performs imaging a plurality of times while relatively moving the subject, and obtains an entire image from a one-dimensional or two-dimensional partial image of the subject. It presupposes an image acquisition device to be synthesized. The imaging element unit includes a first area having a high resolution and a second area having a low resolution, and the first element is obtained based on the movement information of the subject obtained from the partial image acquired in the second area. The point which synthesize | combines the whole image from the partial image acquired in the area | region is the same as that of Embodiment 1. FIG. At that time, the image acquisition in the second region is performed intermittently between the image acquisitions in the first region.

  The imaging interval for acquiring a partial image at a high resolution is different from the imaging interval for acquiring a partial image at a low resolution.

  This makes it possible to reduce the amount of low-resolution image data required for measuring movement information while acquiring sufficient high-resolution image data necessary for image recognition, increasing the image acquisition speed and increasing the resolution. To achieve both.

  The frequency at which image acquisition in the second area is intermittently performed between image acquisitions in the first area is increased or decreased based on movement information of the subject obtained from image acquisition in the second area. The That is, by increasing or decreasing the imaging frequency for acquiring a partial image at a low resolution according to the relative movement speed of the subject, the required amount of data of the low resolution image can be set to an optimal value according to the relative movement speed of the subject. This makes it possible to achieve both high image acquisition speed and high resolution.

  Next, embodiments of the present invention will be described in detail with reference to the drawings. 1 to 4 are diagrams for explaining matters common to the first embodiment.

[Embodiment 1]
FIG. 1 is a schematic diagram of a sensor unit of an optical sweep sensor according to Embodiment 1 of the present invention.

  In FIG. 1, reference numeral 1 denotes a finger that is a subject of fingerprint authentication as a subject, and moves in the direction of an arrow 7 that is a sub-scanning direction. 2 is LED as a light source which irradiates a finger | toe part. Reference numeral 3 denotes an optical member that guides the optical difference of the uneven pattern of the fingerprint to the image sensor 4. The image sensor 4 that is an image sensor section is a strip-shaped sensor having pixels in a plurality of rows (lines) in the sub-scanning direction, and is specifically a CMOS type image sensor. A plurality of pixels in one row are electronically scanned in the main scanning direction. Reference numeral 5 denotes an incident direction of light from the LED 2 to the finger 1. Reference numeral 6 denotes an emission direction of light from the finger 1 toward the optical member 3 and the image sensor 4 after the light emitted from the LED 2 is scattered in the finger 1. Note that the imaging target may be a vein outside the fingerprint.

  FIG. 2 is a block diagram showing the overall configuration of the optical sweep sensor.

  When the sweep of the finger is started, photographing of the fingerprint partial image 10 is started by the pixel unit 12 which is an image sensor unit. The fingerprint partial image 10 is temporarily held in the frame memory unit 13 in the sensor unit 11 and then stored in the memory unit 16 in the microcomputer unit 15 through the A / D converter 14. This operation is performed n times until the finger sweep is completed, and fingerprint partial images 10 for n frames are photographed. These fingerprint partial images 10 are sequentially stored in an arbitrary memory in the memory unit 16. For example, the first frame is stored in the memory (1), the second frame is stored in the memory (2), and the nth frame is stored in the memory (n).

  The frame memory unit 13 in the sensor unit 11 is for shortening the time from the start to the end of photographing one frame. Since the sweep sensor photographs a moving finger, it is important to shorten this time.

  The n fingerprint partial images 10 are sent to the image processing unit 17 where image reconstruction, image enhancement, and the like are performed. Here, the reconstructed fingerprint composite image 10 is stored in the fingerprint database 18 or collated with the fingerprint in the fingerprint database 18 in the collation unit 19.

  FIG. 3 is a diagram simply showing the circuit configuration of the unit pixel of the pixel unit, the frame memory unit, and the unit memory of the optical sweep sensor.

Unit pixels, the unit pixels 41 of the high-resolution, low-resolution unit pixel 42 both of the photodiodes PD, the switch _{M TX,} reset switch _{M RES,} amplifiers _{M SF,} and a switch _{M SEL.} The switch M _TX , the reset switch M _RES , the amplifier M _SF , and the switch M _SEL are FETs.

The switch _MTX is an element that transfers the signal charge of the photodiode PD to the subsequent amplifier, and the reset switch _MRES is an element that removes residual charge at the gate portion of the amplifier. The amplifier _MSF is an amplifier that buffers the signal voltage of the gate unit, and the switch _MSEL is an element that controls reading of the output from the amplifier _MSF to the signal line. The photodiode signal from the pixel and the reset noise of the pixel amplifier are read after being temporarily stored in the capacitors C _S and C _N of the unit memory 43 in the frame memory. _{M 1, M 2, M 3} , M 4 _is a switch for controlling writing and reading with respect to C S, _{C N.} The switches M ₁ , M ₂ , M ₃ , and M ₄ are FETs.

  FIG. 4 is a diagram schematically showing images of the mth frame 46 and the m + 1st frame 47 of the fingerprint partial image 10.

  The reconstruction performed in the image processing unit 17 will be described with reference to FIG.

  Reference numeral 48 in FIG. 4 denotes an overlapping portion of images obtained by photographing the same place of the finger in the mth frame 46 and the m + 1st frame 47. The image processing unit 17 detects this overlapping portion between two consecutive partial images for each frame. When connecting the preceding and following partial images, this overlapping portion is omitted from the partial image of either the m-th frame 46 or the (m + 1) -th frame 47 so that the overlapping portion does not appear twice in the reconstructed image. Then connect. For the front and rear partial images, a characteristic shape is extracted and connected based on the relative positional relationship between them.

  FIG. 5 is a diagram showing a pixel portion including a first region and a second region in the first example described below. The pixel unit will be described as a representative of the first embodiment.

  What is shown in FIG. 5 is a fingerprint sensor. The first area 51 of the image sensor unit is a high-resolution part, and is an area for acquiring a partial image, which is composed of one row of strip-shaped solid-state image sensors. One line of strip-shaped solid-state imaging devices are arranged in the main scanning direction which is the direction of electronic scanning. As described above, the entire image of the finger is synthesized from the series of partial images.

The second area 52 is a low-resolution part, and is an area for acquiring relative movement information of a finger, which includes a plurality of solid-state imaging elements. The movement information is information such as a finger speed, a movement distance, and a direction. The finger moves in another direction different from the main scanning direction, for example, a right-angled direction to make the sub-scanning direction. The second area 52 is provided adjacent to the first area 51 in the finger movement direction, that is, in the sub-scanning direction. In the second region 52, the pixel interval in the main scanning direction in one row is larger than the pixel interval in one row of the first region 51.
Next, specific examples of the image sensor unit according to the first embodiment of the present invention will be described with reference to first to fourth examples.

(First embodiment)
In this embodiment, the image sensor unit is a fingerprint sensor composed of two regions having different resolutions.

  FIG. 6 is a timing chart in the first embodiment.

  In FIG. 5, the first region 51 has one row of 256 pixels with an interval of 25 μm, and the resolution is 256 pixels × 1 row and 1016 DPI. In the second region, 128 pixels with an interval of 50 μm are 4 rows, and the resolution is 128 pixels × 4 rows and 508 DPI. The distance between one row of the first region 51 and the uppermost row of the second region 52 in contact with the first region 51 and the space between the rows of the second region are 50 μm.

  In FIG. 6, the even-numbered pixels 128 and odd-numbered pixels 128 in the first row of the first region 51 are output in the section of the signals V1 and V2. The pixels 128 in the first row, the pixels 128 in the second row 128, the pixels 128 in the third row, and the pixels 128 in the fourth row in the second region 52 are output in the period of the signals V3, V4, V5, and V6. The output of the first area 51 is divided and output in accordance with the output timing of the second area 52 in the main scanning direction.

  Only one line has a high resolution for image synthesis, and the other lines have a low resolution for movement speed detection. Since the high-resolution area is only one line, the frame rate is not so high, and the overall current consumption and cost can be suppressed. In addition, an increase in cost and area can be suppressed by adjusting the IF circuit to a low resolution.

  In order to match the outputs of the first region 51 and the second region 52, the aperture ratios of the respective pixels are α1, α2, the amplifier gains are G1, G2, and α1 × G1 = α2 × G2, and the sensitivity is the same. It becomes.

(Second embodiment)
The present embodiment is a scanner sensor in which an image sensor unit is composed of two regions having different resolutions.

  FIG. 7 is a diagram illustrating a pixel portion including a first area and a second area in the second embodiment.

FIG. 8 is a timing chart in the second embodiment.
FIG. 7 shows a scanner sensor. In FIG. 7, the first region 61 has three rows with 256 pixels having an interval of 25 μm as one row, and the resolution is 256 pixels × 3 rows and 1016 DPI. The interval between the rows is 25 μm. The second area 62 has 6 rows, each including 64 pixels with an interval of 50 μm, and the resolution is 64 pixels × 6 rows and 508 DPI. The interval between the lowermost row of the first region 61 and the uppermost row of the second region 62 in contact with the first region 61 and the interval between the rows of the second region are 50 μm.

  In FIG. 8, 1 to 64 pixels and 65 to 128 pixels in the first row of the first region 61 are output in the section of the signals V1 and V2. Subsequently, 193 to 256 pixels in the third row of the first region 61 are sequentially output in the section of the signal V12. In the section of the signals V13, V14, and V15, 1 to 64 pixels in the first row of the second area 62, 1 to 64 pixels in the second row, and 1 to 64 pixels in the third row are output. The output of the first area 61 is divided and output in accordance with the output timing of the second area 62 in the main scanning direction.

  In three lines, the resolution is increased for image synthesis, and in the other lines, the resolution is decreased for detecting the moving speed. Since the high-resolution area has only three rows, the frame rate is not so high, and the overall current consumption and cost can be suppressed. In addition, an increase in cost and area can be suppressed by adjusting the IF circuit to a low resolution.

  In order to match the outputs of the first region 61 and the second region 62, the aperture ratios of the respective pixels are α1, α2, the amplifier gains are G1, G2, and α1 × G1 = α2 × G2, and the sensitivity is the same. It becomes.

(Third embodiment)
The present embodiment is a scanner sensor in which an image sensor unit is composed of three regions having different resolutions.

  FIG. 9 is a diagram showing the pixel portion including the first area, the second area, and the third area in the third embodiment. The second and third areas are sequentially provided in the first area in the direction in which the subject is relatively moved.

  FIG. 9 shows a scanner sensor. The first area 71 has the same configuration as that of the second embodiment. The first area 71 has three rows each including 256 pixels with an interval of 25 μm, and the resolution is 256 pixels × 3 rows and 1016 DPI. The interval between the rows is 25 μm.

  The second area 72 has the same configuration as that of the second embodiment. The second area 72 has six lines, each including 64 pixels with an interval of 50 μm, and the resolution is 64 pixels × 6 lines and 508 DPI. The interval between the bottom row of the first region 71 and the top row of the second region 72 in contact with the first region 71 and the interval between the rows of the second region are 50 μm.

  There is a third region 73 in parallel with the second region 72 in the sub-scanning direction. There are four rows with 32 pixels having an interval of 50 μm as one row, and the resolution is 2 pixels × 4 rows, 254 DPI. The interval between the lowermost row of the second region 72 and the uppermost row of the third region 73 in contact with the second region 72 and the interval between the rows of the third region are 100 μm.

  The first area 71 has a high resolution for image composition, the second area 72 has a medium resolution, and is used for both moving speed detection and image acquisition, and the third area 73 has a low resolution, It is for moving speed detection.

  Since the high-resolution area has only three rows, the frame rate is not so high, and the overall current consumption and cost can be suppressed. In addition, an increase in cost and area can be suppressed by adjusting the IF circuit to a low resolution.

  In order to match the outputs of the first region 71 and the second region 72, the aperture ratio of each pixel is α1, α2, the amplifier gains are G1, G2, and α1 × G1 = α2 × G2, and the sensitivity is the same. It becomes.

  By matching the first area 71 with the resolution and pitch of the second area 72, the second area 72 can also acquire an image together with the first area 71. In addition, the circuit scale of the IC can be reduced by matching the IF to the second area of medium resolution.

  Furthermore, by setting a low resolution area in the third area 73 so as to have a narrow width, it is possible to follow even when the relative speed is higher.

  The second and third regions 72 and 73 are narrower than the first region 71. In this case, the vacant region is a circuit unit such as an AMP unit, a TG unit, or an A / D conversion unit. It is possible to effectively use the area of the IC by using it.

  In each of the first to third embodiments, the pixel interval of the low resolution second region is larger (coarse) than the high resolution first region in the sub-scanning direction. The pixel interval of the first region in the scanning direction may be matched with the pixel interval of the second region.

(Fourth embodiment)
FIG. 10 is a diagram illustrating a pixel portion including a first area and a second area in the fourth embodiment.

  Although there is a second region 82 provided side by side with the first region 81 in the sub-scanning direction, the pixel spacing in both the main scanning direction and the sub-scanning direction is the same.

  Here, in order to make the second region 82 a low resolution portion, a part of the pixels is thinned out. For example, the thinning interval is 8 pixels. That is, the spatial arrangement of pixels from which signals are read differs between the two regions. As a result, the first region 81 for image composition has a high resolution, and the second region 82 has a low resolution for detecting the moving speed. The pixel arrangement is the same throughout and can be easily manufactured.

[Embodiment 2]
(Fifth embodiment)
Next, a fifth embodiment to which the present invention is applied will be described.

  In this embodiment, the image sensor unit is a fingerprint sensor composed of two regions having different resolutions.

  The sensor unit of the optical sweep sensor and the circuit configuration of the unit pixel in the pixel unit, the frame memory unit, and the unit memory in this example are the same as those in the first embodiment, and are shown in FIGS.

  FIG. 11 is a block diagram showing the overall configuration of the optical sweep sensor in the present embodiment.

  When the sweep of the finger is started, photographing of the fingerprint partial image 10 (10a, 10b) is started by the pixel unit 22 which is an image sensor unit. The fingerprint partial image 10 is temporarily stored in the frame memory unit 23 in the sensor unit 21 and then stored in the memory unit 26 in the microcomputer unit 25 through the A / D converter 24. This operation is performed n times until the finger sweep is completed, and fingerprint partial images 10 for n frames are photographed. These fingerprint partial images 10 are sequentially stored in an arbitrary memory in the memory unit 26. For example, the first frame is stored in the memory (1), the second frame is stored in the memory (2), and the nth frame is stored in the memory (n).

  The frame memory unit 23 in the sensor unit 21 is for shortening the time from the start to the end of photographing one frame. Since the sweep sensor photographs a moving finger, it is important to shorten this time.

  Here, the fingerprint partial image in the present embodiment can be output by switching between a high-resolution partial image and a low-resolution partial image. Of the n fingerprint partial images 10, a low resolution partial image is represented as 10a, and a high resolution partial image is represented as 10b. The low-resolution partial image 10a is sent to the speed detection unit 27, and a correlation between the two partial images is obtained to obtain a movement speed that is movement information of the subject. As the movement information, a movement distance or acceleration may be used.

  On the other hand, the high-resolution partial image 10b is sent to the image processing unit 28, and is reconstructed (combined) from the high-resolution partial image using the movement information 34, which is speed information of the speed detection unit, to acquire the entire image. . The fingerprint composite image of the entire subject obtained by the reconstruction is further subjected to image processing such as image enhancement in the image processing unit 28, and then feature extraction is performed and converted into fingerprint feature information 32. . The fingerprint feature information 35 is feature point distribution information such as end points and branch points of fingerprint ridges. The fingerprint feature information 35 is sent to the collation / registration unit 29, registered in the fingerprint database 30, or collated with the registered fingerprint feature information in the fingerprint database 30. On the other hand, the movement information 31 obtained by the speed detection unit 27 is also sent to the image acquisition routine control unit 34 in the timing generation unit (TG unit) 31 in the sensor unit 21. Here, based on the movement information 34, a resolution control switching signal 36 for switching the resolution is sent to the resolution mode switching unit 33, so that the resolution mode switching unit 33 outputs the sensor drive pulse 37 according to the switched resolution mode to the pixel unit. Select output to.

  By using FIG. 4 again, the images of the m-th frame 32 and the m + 1-th frame 33 of the low-resolution fingerprint partial image 10a can be schematically represented.

  The acquisition of reconstruction movement information performed by the speed detection unit 17 will be described with reference to FIG.

  Reference numeral 48 in FIG. 4 denotes an overlapping portion of images obtained by photographing the same place of the finger in the mth frame 46 and the m + 1st frame 47. The speed detector 27 detects this overlapping portion between two consecutive partial images for each frame. By obtaining the correlation, movement information (distance and speed) indicating how far the overlapping portion is located is detected. Using the movement information obtained in this manner, the image processing unit 18 forms the entire fingerprint image by arranging the partial images 10b of the high-resolution area side by side.

  When connecting the preceding and following partial images, this overlapping portion is omitted from the partial image of either the m-th frame 46 or the (m + 1) -th frame 47 so that the overlapping portion does not appear twice in the reconstructed image. Then connect. For the front and rear partial images, a characteristic shape is extracted and connected based on the relative positional relationship between them.

  The pixel portion composed of the first region and the second region in this example is shown in FIG. 5 in the first embodiment.

  What is shown in FIG. 5 is a fingerprint sensor. The first area 51 of the image sensor unit is a high-resolution part, and is an area for acquiring a partial image, which is composed of one row of strip-shaped solid-state image sensors. One line of strip-shaped solid-state imaging devices are arranged in the main scanning direction which is the direction of electronic scanning. As described above, the entire image of the finger is synthesized from the series of partial images.

  The second area 52 is a low-resolution part, and is an area for acquiring relative movement information of a finger, which includes a plurality of solid-state imaging elements. The movement information is information such as a finger speed, a movement distance, and a direction. The finger moves in another direction different from the main scanning direction, for example, a right-angled direction to make the sub-scanning direction. The second region 52 is provided side by side with the first region 51 in the finger movement direction, that is, in the sub-scanning direction. In the second region 52, the pixel interval in the main scanning direction in one row is larger than the pixel interval in one row of the first region 51.

  12a, 12b, 13a, and 13b are timing charts for explaining the operation of this embodiment.

  12a shows the first readout mode, FIG. 12b shows the second readout mode, FIG. 13a shows the case where the relative movement speed with the subject is slow, and FIG. 13b shows the timing when the relative movement speed with the subject is fast. It is a chart.

  FIG. 14 is a flowchart for explaining the operation of this embodiment.

  In FIG. 5, the first region 51 has one row of 256 pixels with an interval of 25 μm, and the resolution is 256 pixels × 1 row and 1016 DPI. In the second region, 128 pixels with an interval of 50 μm are 4 rows, and the resolution is 128 pixels × 4 rows and 508 DPI. The distance between one row of the first region 51 and the uppermost row of the second region 52 in contact with the first region 51 and the space between the rows of the second region are 50 μm. Only one line has a high resolution for image synthesis, and the other lines have a low resolution for movement speed detection.

  In FIGS. 12a and 12b, signals V1, V2, V3, V4, V5, and V6 indicate the output signals of the shift register in the sub-scanning direction for selecting the pixel row to be read. DATA indicates a signal of the read pixel row.

  In the interval between the signals V1 and V2, 128 pixels are output for the even pixels in the first row of the first region 51, and 128 pixels are output for the odd pixels. In the section of the signals V3, V4, V5, and V6, the second row 52 has the same number of pixels as the first row of 128 pixels, the second row of 128 pixels, the third row of 128 pixels, and the fourth row. As many as 128 pixels are output. The output of the first area 51 is divided into two in accordance with the output timing of the second area 52 in the main scanning direction.

  FIG. 12a shows the case of the first readout mode in which the first region that is the high resolution portion and the second region that is the low resolution portion are successively read out as partial images. One frame of partial image is output for 768 pixels. FIG. 12b shows a case of the second readout mode in which only the first region which is the high resolution portion is continuously read out as a partial image. One frame of partial image is an output for 256 pixels. The data transfer time in the second read mode is calculated to be 1/3 of the data transfer time in the first read mode in which the data in the first area and the second area are transferred together.

  In order to match the outputs of the first region 51 and the second region 52, the aperture ratio of each pixel is α1, α2, the amplifier gains are G1, G2, and α1 × G1 = α2 × G2, and the sensitivity Are the same.

  13A, 13B, and 14 will be used to describe the operation when acquiring partial images in succession.

  In FIGS. 13a and 13b, DATAOUT indicates a fingerprint partial image signal read from the pixel portion. In the fingerprint partial image signal, a portion described as 500 DPI is a low-resolution partial image, and a portion described as 1000 DPI is a high-resolution partial image.

  FIG. 13A is a timing chart when the moving speed of a finger as a subject is slow, and reading is continuously performed in a first reading mode in which high resolution and low resolution are combined. The reading speed is 1000 fps (frame / second).

  FIG. 13B is a timing chart when the moving speed of the finger as a subject is fast. After outputting two frames in the first readout mode combining the high resolution and the low resolution, 4 in the second readout mode of only the high resolution. A total of 6 frames are output repeatedly. The imaging interval at the high resolution is 2 frames at 1000 fps and 4 frames at 3000 fps, which is equivalent to 1800 fps on average.

  Thus, when the moving speed of the finger is fast, the imaging interval at high resolution is increased from 1000 fps to 1800 fps, thereby enabling imaging that follows the movement of the subject. On the other hand, the low-resolution imaging interval for obtaining the moving speed extends from the 2-frame interval to the 6-frame interval, but the moving speed is interpolated from the preceding and following moving speeds obtained at the 6-frame interval. Can be sought. Therefore, it is possible to distribute the data amount reduced by thinning out the low-resolution imaging to high resolution and high-speed imaging. Further, when the relative moving speed of the subject is slow, the imaging interval may be slow, and accordingly, the moving speed measurement interval is shortened to distribute the moving speed with high accuracy. In this way, optimization can be performed by changing the combination of the imaging intervals according to the moving speed.

  The flowchart shown in FIG. 14 shows an image acquisition routine performed by the image acquisition routine control unit 23 of FIG.

  In FIG. 14, when the image acquisition routine is started in S (step) 110 with a control signal (not shown) from the microcomputer unit 25 in FIG. 11, first in S111, a first readout mode in which high resolution and low resolution are combined. Thus, two partial images are acquired. In S112, the movement information 34 obtained from the image of the low resolution portion of the two partial images is acquired, and in S113, the threshold Vt is compared with the movement speed of the subject. If the speed is lower than the threshold value, the process proceeds to S118. If the image acquisition is not terminated by a control signal (not shown) from the microcomputer unit 25, the process proceeds to S111, and the first readout mode combining the high resolution and the low resolution. Continue to get images at.

  On the other hand, the threshold value Vt is compared with the moving speed of the subject in S113, and if the speed is higher than the threshold value, the process proceeds to S114, the frame count number n is initialized, and only the high resolution is read in the second readout mode in S115. Get an image. In step S116, the frame count number n is incremented, and in step S117, it is compared whether the frame count number has reached 4 or more. If the frame count is less than 4 in S117, the process proceeds to S114, and if it is 4 or more, the process proceeds to S118. If the image acquisition is not terminated by a control signal (not shown) from the microcomputer unit 25, the process proceeds to S111, and an image is acquired again in the first readout mode that combines high resolution and low resolution. In the case of termination, the process proceeds to S119 and ends.

  In S111, a process for acquiring an image at a low resolution is performed, and in S111 and S114 to 117, a process for acquiring an image at a high resolution is performed. When the moving speed is high, the period of S114 to 117 is entered so that the low resolution is obtained. The process of acquiring images at is intermittently performed. Further, the imaging interval is optimized by switching the operation according to the moving speed in S113.

  In this way, the step of acquiring an image at a low resolution and the step of acquiring an image at a low resolution are separated, and the step of acquiring an image at a low resolution is intermittently performed. Here, an image at a high resolution is an image necessary for image recognition, and an image at a low resolution is for acquiring auxiliary information (here, movement information) for acquiring an image at a high resolution. It is an image.

  This makes it possible to reduce the amount of low-resolution image data required to measure movement information to the required level while acquiring sufficient high-resolution image data necessary for image recognition, and increase the image acquisition speed. And high resolution.

  In addition, the method further includes a control step of increasing or decreasing the frequency of intermittently performing the step of acquiring an image at a low resolution between the steps of acquiring an image at a high resolution based on movement information of the subject.

  This makes it possible to optimize the amount of low-resolution image data required to measure movement information according to the relative movement speed of the subject, and to achieve both higher image acquisition speed and higher resolution. To do.

(Sixth embodiment)
The present embodiment is a scanner sensor in which an image sensor unit is composed of two regions having different resolutions.

  FIG. 15 is a diagram illustrating a pixel portion including a first area and a second area in the sixth embodiment.

  FIG. 16A is a full-screen output mode, FIG. 16B is a timing chart in the first readout mode, and FIG. 16C is a timing chart in the second readout mode.

  FIG. 17A is a timing chart when the relative movement speed with the subject is low, and FIG. 17B is a timing chart when the relative movement speed with the subject is high.

  FIG. 18 is a flowchart for explaining the operation of the sixth embodiment.

  FIG. 15 shows a scanner sensor.

  In FIG. 15, the first region 91 has three rows with 256 pixels having an interval of 25 μm as one row, and the resolution is 256 pixels × 3 rows and 1016 DPI. The interval between the rows is 25 μm. The second region 92 has 24 rows with 64 pixels having an interval of 50 μm as one row, and the resolution is 64 pixels × 24 rows and 508 DPI. The interval between the lowermost row of the first region 91 and the uppermost row of the second region 92 in contact with the first region 91 and the interval between the rows of the second region are 50 μm. Three lines have a high resolution for image synthesis, and another 24 lines have a low resolution for movement speed detection.

  In FIG. 16a, signals V1, V2, V3,... V36 indicate output signals of the shift register in the sub-scanning direction for selecting the pixel row to be read. DATA indicates a signal of the read pixel row.

  1 to 64 pixels and 65 to 128 pixels in the first row of the first region 91 are output in the section of the signals V1 and V2. Thereafter, 193 to 256 pixels in the third row of the first region 61 are output in order of 64 pixels in the interval of the signal V12. In the interval of the signals V13 and V14 to V36, 1 to 64 pixels in the first row of the second region 92, 1 to 64 pixels in the second row, etc., and then 1 to 64 pixels in the 24th row are output. The The output of the first area 91 is divided into eight and output in accordance with the output timing of the second area 92 in the main scanning direction.

  Three lines have a high resolution for image synthesis, and the other lines have a low resolution for movement speed detection. Since the high-resolution area has only three rows, the frame rate is not so high, and the overall current consumption and cost can be suppressed. In addition, an increase in cost and area can be suppressed by adjusting the IF circuit to a low resolution.

  In order to match the outputs of the first region 91 and the second region 92, the aperture ratios of the respective pixels are α1, α2, the amplifier gains are G1, G2, and α1 × G1 = α2 × G2, and the sensitivity is the same. It becomes.

  FIG. 16b shows the case of the first readout mode in which only the second region, which is the low resolution portion, is continuously read out as a partial image. One frame of partial image is output for 1536 pixels. FIG. 16c shows a case of the second readout mode in which only the first region which is the high resolution portion is continuously read out as a partial image. One frame of partial image is output for 1536 pixels. The time for data transfer in the second read mode is equal to that in the first read mode.

  The operation when acquiring partial images successively will be described with reference to FIGS.

  17A and 17B, DATAOUT indicates a partial image signal read from the pixel portion. Of the partial image signal, a portion described as 500 DPI is a low-resolution partial image, and a portion described as 1000 DPI is a high-resolution partial image.

  FIG. 17A is a timing chart when the relative movement speed of the scanner with respect to the paper as the subject is slow.

  After reading two frames at a low resolution, which is the first readout mode, the high resolution, which is the second readout mode, is read periodically and repeatedly as a set of two frames. The effective reading speed is about half of 500 fps when each mode is 1000 fps (frame / second).

  FIG. 17B is a timing chart when the relative movement speed of the scanner with respect to the subject is high.

  After reading two frames at a low resolution, which is the first readout mode, the high resolution, which is the second readout mode, is read out periodically with one set of 6 frames. The interval between effective imaging at high resolution is 750 fps.

  As described above, when the relative movement speed of the subject is fast, the imaging interval at high resolution is increased from 500 fps to 750 fps, thereby enabling imaging that follows the relative movement of the subject. On the other hand, the low-resolution imaging interval for obtaining the moving speed extends from the 4-frame interval to the 8-frame interval, but the moving speed is obtained by interpolating the moving speed between the previous and next moving speeds obtained at the 8-frame interval. Can do. Therefore, the amount of data reduced by thinning out low-resolution imaging is distributed to high resolution and high-speed imaging. Further, when the relative moving speed of the subject is slow, the imaging interval may be slow, and accordingly, the moving speed measurement interval is shortened to distribute the moving speed with high accuracy. In this way, optimization can be performed by changing the combination of the imaging intervals according to the moving speed.

  FIG. 18 is a flowchart in the present embodiment.

  This flowchart shows an image acquisition routine performed by the image acquisition routine control unit 23 of FIG.

  In FIG. 18, when the image acquisition routine is started by a control signal (not shown) from the microcomputer unit in S120, two partial images are first acquired in a reading mode having a low resolution in S121. In S122, the movement information obtained from the two partial images is acquired, and in S123, the threshold value Vt is compared with the relative movement speed of the subject. If the speed is higher than the threshold, the process proceeds to S124, and the frame limit number m is set to 6. When the speed is lower than the threshold value, the process proceeds to S125 and the frame limit number m is set to 2. In S126, the frame count number n is initialized. In S127, an image having only a high resolution is acquired. In S128, the frame count number n is incremented. In S129, it is compared whether the frame count number has reached m or more. If the frame count is less than m in S129, the process proceeds to S126, and if it is greater than or equal to m, the process proceeds to S130. If the image acquisition is not terminated by a control signal (not shown) from the microcomputer unit, the process proceeds to S120, and the image is acquired again in the low resolution mode. In the case of termination, the process proceeds to S131 and ends.

  A step of acquiring an image with a low resolution is performed at S121, and a step of acquiring an image with a high resolution is performed at S126 to 129. In S123, the interval of the process of acquiring images at low resolution that is performed intermittently is optimized by switching the number of image acquisitions between S126 and 129 according to the moving speed.

  In this way, an image of a high-resolution area necessary for performing image recognition and an image of a low-resolution area for acquiring auxiliary information (here, movement information) for acquiring the image of the high-resolution area. Separate the acquisition process. In addition, the image acquisition process of the low resolution area is intermittently performed. This makes it possible to reduce the amount of low-resolution image data required to measure movement information to the required level while acquiring sufficient high-resolution image data necessary for image recognition, and increase the image acquisition speed. And high resolution.

  In addition, the method further includes a control step of increasing or decreasing the frequency of intermittently performing the step of acquiring an image at a low resolution between the steps of acquiring an image at a high resolution based on the relative movement information of the subject.

  This makes it possible to optimize the amount of low-resolution image data required to measure movement information according to the relative movement speed of the subject, and to achieve both higher image acquisition speed and higher resolution. To do.

(Seventh embodiment)
Next, a seventh embodiment to which the present invention is applied will be described.

  The present embodiment shows an example of a biometric authentication sensor similar to the third embodiment, but here, it is exemplified as a capacitance type fingerprint sensor. As described above, as a fingerprint sensor, in addition to the optical type, a capacitance method, a pressure detection method, a thermal method, an electric field detection method, and the like are known. However, the essence of the present invention is that the resolution and acquisition timing of the acquired image for synthesizing the whole image and the image for speed detection are separately optimized, and it goes without saying that the detection method is not used.

  19a to 19c are diagrams illustrating a pixel portion of a capacitive fingerprint sensor to which the present invention is applied. FIG. 19a shows a pixel configuration, FIG. 19b shows a high-resolution reading, and FIG. 19c shows a low-resolution reading.

  Here, the high-resolution first area and the low-resolution second area shown in FIG. 15 partially overlap each other as indicated by 101 and 102. In other words, at least a part of the second area shares the area with the first area. When reading at high resolution, all pixels of only the high resolution portion are read as shown in FIG. 19b. On the other hand, when reading at a low resolution, a part of the high resolution part is thinned out together with the low resolution part as shown in FIG. As an operation, the image acquisition process at a high resolution and the image acquisition process at a low resolution for acquiring movement information are separated as shown in FIGS. The image acquisition process at the resolution is performed intermittently. This makes it possible to reduce the amount of low-resolution image data required to measure movement information to the required level while acquiring sufficient high-resolution image data necessary for image recognition, and increase the image acquisition speed. And high resolution.

  Although the fingerprint sensor and the scanner sensor have been described in the first and second embodiments, the present invention may be applied to other biometric authentication image acquisition apparatuses such as a sensor that acquires a vein image instead of a fingerprint image. In addition, in terms of capturing an image to recognize an object, the present invention may be applied to an industrial camera other than a scanner or an object recognition camera such as a robot eye. In addition to the optical method, the imaging method may be other methods such as a capacitance method or a method using an electric field.

Schematic of sensor part of optical sweep sensor in the present invention The block diagram showing the whole structure of the optical sweep sensor in Embodiment 2 of this invention A simple representation of the circuit configuration of the unit pixel in the pixel section of the optical sweep sensor, the frame memory section, and the unit memory The figure which represented typically the image of the mth frame 12 of the fingerprint partial image 10, and the m + 1st frame 13 The figure which showed the pixel part which consists of 1st area | region and 2nd area | region in the 1st Example of Embodiment 1 Timing chart in the first embodiment The figure which showed the pixel part which consists of 1st area | region and 2nd area | region in 2nd Example. Timing chart in the second embodiment The figure which showed the pixel part which consists of a 1st area | region, a 2nd area | region, and a 3rd area | region in 3rd Example. The figure which showed the pixel part which consists of a 1st area | region and a 2nd area | region in a 4th Example. The block diagram showing the whole structure of the optical sweep sensor in Embodiment 2 of this invention Timing chart of first read mode in fifth embodiment Timing chart of second read mode in fifth embodiment Timing chart when the relative movement speed with respect to the subject in the fifth embodiment is slow Timing chart when the relative movement speed with respect to the subject is high in the fifth embodiment Flow chart in the fifth embodiment The figure which showed the pixel part which consists of a 1st area | region and a 2nd area | region in a 6th Example. Timing chart of full screen output mode in sixth embodiment Timing chart of first read mode in sixth embodiment Timing chart of second read mode in sixth embodiment Timing chart when the relative movement speed with the subject in the sixth embodiment is slow Timing chart when the relative movement speed with the subject in the sixth embodiment is high Flow chart in the sixth embodiment Diagram for explaining image acquisition of fingerprint sensor in seventh embodiment, pixel configuration Similarly when reading high resolution Similarly when reading low resolution The figure for demonstrating the image acquisition by the image pick-up element part using one line sensor (solid-state image pick-up element), when the relative movement speed with a to-be-photographed object is slow enough Similarly, when the relative movement speed with the subject is fast Diagram for explaining image acquisition by the image sensor unit using multiple line sensors, when the relative movement speed with the subject is sufficiently slow Similarly, when the relative movement speed with the subject is fast The figure which shows the image acquisition apparatus of the conventional sweep type system disclosed by patent document 1 The figure which shows the image acquisition apparatus of the conventional sweep type system disclosed by patent document 2 The figure which shows the image acquisition apparatus of the conventional sweep type system disclosed by patent document 3

Explanation of symbols

1 ... finger 2 ... LED (light source)
DESCRIPTION OF SYMBOLS 3 ... Optical member 4 ... Imaging device 5 ... Light incident direction 6 ... Light emission direction 7 ... Finger movement direction 11, 21 ... Sensor part 15, 25 ... Microcomputer part 51, 61, 71, 81, 91, 101 ... 1st area | region 52,62,72,82,92,102 ... 2nd area | region 73 ... 3rd area | region

Claims (18)

  In an image acquisition apparatus that has an image sensor unit composed of a plurality of pixels and captures a plurality of times while relatively moving a subject, the image sensor unit has a high resolution first An entire image from the partial image acquired in the first region based on the movement information of the subject obtained from the partial image acquired in the second region. An image acquisition apparatus characterized by combining the above.   The image acquisition apparatus according to claim 1, wherein the second area is provided adjacent to the first area in a direction in which the subject is relatively moved.   There is a third region having a resolution lower than that of the second region, and the second and third regions are sequentially arranged alongside the first region in the direction of relative movement of the subject, and the second region The movement information of the subject is obtained from the partial images acquired in the third area, and the entire image is synthesized from the partial images acquired in the first area and the second area based on the movement information. The image acquisition apparatus according to claim 1.   2. The image acquisition according to claim 1, wherein the imaging element unit includes a plurality of rows of pixels in a sub-scanning direction in which a direction in which pixels in one row are scanned is a main scanning direction and a subject is relatively moved. apparatus.   The image acquisition apparatus according to claim 4, wherein the second region has a pixel sensitivity matched to that of the first region.   The image acquisition apparatus according to claim 4, wherein the second region has a pixel interval in the main scanning direction larger than a pixel interval of the first region.   The image acquisition apparatus according to claim 6, wherein the second region has a pixel interval in the sub-scanning direction larger than a pixel interval of the first region.   The image acquisition apparatus according to claim 6, wherein the first area is divided and output in accordance with an output timing of the second area in a main scanning direction.   The image acquisition apparatus according to claim 6, wherein the first region has a pixel interval in the sub-scanning direction that matches the second region.   The second region has the same pixel spacing in the main scanning direction and the sub-scanning direction as the first region, and the resolution is lowered by reading out part of the pixels in the second region. The image acquisition apparatus according to claim 4.   A scanner device, wherein the image acquisition device according to claim 1 is used.   The biometric authentication apparatus using the image acquisition apparatus according to claim 1, wherein the imaging target is a fingerprint or a vein.   In an image acquisition apparatus that has an image sensor unit composed of a plurality of pixels and captures a plurality of times while relatively moving a subject, the image sensor unit has a high resolution first A second region having a low resolution, and the image acquisition in the second region is intermittently performed between image acquisitions in the first region, and acquired in the second region. An image acquisition apparatus comprising: combining an entire image from a partial image acquired in the first area based on movement information of a subject obtained from a partial image.   The frequency with which image acquisition in the second region is intermittently performed during image acquisition in the first region is increased or decreased based on movement information of the subject. Image acquisition device.   15. The image acquisition apparatus according to claim 13, wherein at least a part of the second area shares the area with the first area.   The image acquisition apparatus according to claim 13, wherein the subject is a living body, and an image for biometric authentication is acquired.   In an image acquisition method for acquiring an entire image by performing multiple imaging while moving relative to a subject, a step of acquiring an image at a high resolution, a step of acquiring an image at a low resolution, and a portion acquired at the low resolution Obtaining subject movement information from an image, and synthesizing an entire image from partial images obtained in the step of obtaining an image at the high resolution based on the movement information, and obtaining an image at the low resolution. A process is performed intermittently between the processes of acquiring an image at the high resolution.   The method further comprises a control step of increasing or decreasing the frequency of intermittently performing the step of acquiring an image at the low resolution during the step of acquiring an image at the high resolution based on movement information of a subject. The image acquisition method according to 17.

Images