JP6805521B2 - Image reading device, correction method of scanned image in the device, and correction program of scanned image - Google Patents

Description

The present invention relates to an image reading device used in an image forming apparatus or the like and reading an image of a document conveyed in a sub-scanning direction, a method for correcting a scanned image in the apparatus, and a correction program for the scanned image.

In an image reading device equipped with an automatic document transporting device mounted on an image forming device or the like, two types of light receiving sensors currently used are a CCD (Charge Coupled Device) sensor and a CIS (Contact Image Sensor) sensor. Be done.

The CIS unit has the advantage of being smaller than the CCD unit, and is used as a backside sensor inside the document transfer device in office machines. However, there are problems that the depth of focus is shallow and the cost is high. On the other hand, the CCD unit has a problem in that the configuration becomes large and it lacks versatility.

In order to solve the above problems, there is a technique for reducing the size of a CCD sensor, which is currently composed of one sensor, by dividing it into a plurality of sensors in the main scanning direction and shortening the optical path. As a result, it is possible to realize a small reading unit having a deeper depth of focus than the CIS sensor and a lower cost than one CCD sensor.

However, since the optical path is short, there is a problem that it is easily affected by the variation in the distance between the reading surface and the CCD sensor during document transportation. Specifically, when the document is separated from the reference scanning position when the document is conveyed, the scanned image is reduced and the scanned images are duplicated at the boundary of the adjacent CCD sensors. On the other hand, when the document is separated from the scanning position P0 below, the scanned image is enlarged, and in some cases, the image is lost.

As described above, the reading unit in which the CCD sensor is divided and configured has a problem that the magnification fluctuation due to the variation in the distance between the reading surface and the CCD sensor during document transport becomes larger than that of the CCD or CIS having a single configuration. ..

Note that Patent Document 1 discloses a technique in which a reference line is drawn at a reading position of a joint portion of a plurality of CCD sensors, and the position shift and the magnification shift are corrected in advance using the reference line.

Japanese Unexamined Patent Publication No. 2000-175001

However, the technique disclosed in Patent Document 1 merely corrects the positional deviation and the magnification deviation in advance by using the reference line, and the variation in the position of the conveyed original, in other words, the reading surface during the original conveying. It is not possible to correct the magnification variation of the scanned image due to the variation in the distance between the and the CCD sensor. Therefore, in response to the problem that the magnification variation due to the variation in the distance between the scanning surface and the CCD sensor during document transport is large. It did not provide a sufficient countermeasure.

The present invention has been made in view of such a technical background, and is an image reading device capable of appropriately correcting magnification fluctuations due to variations in the distance between the reading surface and the light receiving sensor during document transport. It is an object of the present invention to provide a correction method and a correction program for a scanned image in the above.

The above problem is solved by the following means.
(1) A plurality of optical reduction type light receiving sensors arranged at intervals in the main scanning direction and for reading an image of a document conveyed in the sub scanning direction, and the main scanning of two light receiving sensors adjacent to each other. The boundaries of the reading range in the direction are arranged so as to coincide with each other in the vicinity of the reference transport surface of the document or the reference transport surface on the light receiving sensor side, and the distance from the reference transport surface to the light receiving sensor increases. A plurality of light receiving sensors that can read a part of the main scanning direction of the document in an overlapping manner by two adjacent light receiving sensors when the document is conveyed at a position, and a light receiving sensor that is adjacent to each other among the plurality of light receiving sensors. A duplication detecting means for detecting duplication of images obtained by a sensor, a scanning magnification calculating means for calculating the scanning magnification of a document from the duplication width when duplication is detected by the duplication detecting means, and the scanning magnification calculating means. An image reading device including a correction means for correcting the magnification of an image in a target area based on the reading magnification calculated by the above.
(2) When the duplication detection means does not detect duplication of images obtained by reading originals by light receiving sensors adjacent to each other and there is no loss of images obtained by the adjacent light receiving sensors , the correction means. the image reading apparatus according to item 1, remain equal magnification magnification of the image of the target area.
(3) When the duplication detection means does not detect duplication of images obtained by reading originals by light receiving sensors adjacent to each other, the image is equalized from the edge portion of the image obtained by the adjacent light receiving sensors. It is provided with an equal-magnification / enlargement determination means for determining whether the image is magnified or magnified, and when the equal-magnification / enlargement determination means determines that the image is magnified, the correction means measures the magnification of the image obtained by the light receiving sensor. When it is determined that the image is magnified by the same magnification / enlargement determination means, the correction means corrects the magnification of the image in the target area in the reduction direction. ..
(4) When the duplication detection means does not detect duplication of images obtained by reading originals by light receiving sensors adjacent to each other, and detects a defect width of images obtained by the adjacent light receiving sensors. The image reading device according to item 2 or 3 above, wherein the correction means reduces an image in a target area at a reduction magnification corresponding to the defect width.
(5) The image reading device according to any one of the above items 1 to 4, wherein the correction means corrects the magnification of the image in the target area in the main scanning direction for each position in the sub-scanning direction.
(6) A document size detecting means for detecting the size of the document and
After correcting the magnification of the image by the correction means, adjustment is made to adjust the magnification of the image read by each light receiving sensor so that the size of the image read by the corrected light receiving sensor matches the document size detected by the document size detecting means. The image reading device according to any one of the preceding items 1 to 5, further comprising means.
(7) A feature amount detecting means for detecting the feature amount of the image read by the light receiving sensor is provided, and the duplicate detection means is an image read by two adjacent light receiving sensors from the feature amount detected by the feature amount detecting means. It is determined whether or not the image cannot detect the overlap width even if the overlap is generated, and if it is determined that the image cannot detect the overlap width, the overlap detection operation is not performed. The image reader described.
(8) A plurality of optical reduction type light receiving sensors arranged at intervals in the main scanning direction and for reading an image of a document conveyed in the sub scanning direction, and the main scanning of two light receiving sensors adjacent to each other. The boundaries of the reading range in the direction are arranged so as to coincide with each other in the vicinity of the reference transport surface of the document or the reference transport surface on the light receiving sensor side, and the distance from the reference transport surface to the light receiving sensor increases. An image reading device including a plurality of light receiving sensors capable of overlapping and reading a part of the main scanning direction of the document by two adjacent light receiving sensors when the document is conveyed at a position is a device of the plurality of light receiving sensors. Of these, a duplication detection step that detects duplication of images obtained by scanning documents by light receiving sensors adjacent to each other, and if duplication is detected by the duplication detection step, scanning that calculates the scanning magnification of the document from the overlap width. A method for correcting a read image in an image reading device, which comprises executing a magnification calculation step and a correction step for correcting the magnification of an image in a target area based on the reading magnification calculated by the reading magnification calculation step. ..
(9) A plurality of optical reduction type light receiving sensors arranged at intervals in the main scanning direction and for reading an image of a document conveyed in the sub scanning direction, and the main scanning of two light receiving sensors adjacent to each other. The boundaries of the reading range in the direction are arranged so as to coincide with each other in the vicinity of the reference transport surface of the document or the reference transport surface on the light receiving sensor side, and the distance from the reference transport surface to the light receiving sensor increases. When the document is conveyed at a position, the computer of the image reading device provided with a plurality of light receiving sensors capable of overlapping and reading a part of the main scanning direction of the document by two adjacent light receiving sensors receives the plurality of light. Among the sensors, a duplication detection step that detects duplication of images obtained by scanning documents by light receiving sensors adjacent to each other, and if duplication is detected by the duplication detection step, the scanning magnification of the document is calculated from the overlap width. A scanning image correction program in an image reading device for executing a reading magnification calculation step and a correction step for correcting the magnification of an image in a target area based on the reading magnification calculated by the reading magnification calculation step.

According to the invention described in the previous section (1), a plurality of optical reduction type light receiving sensors arranged at intervals in the main scanning direction and for reading an image of a document conveyed in the sub scanning direction, and each other. The boundary of the reading range in the main scanning direction of the two adjacent light receiving sensors is arranged so as to coincide with each other in the vicinity of the reference transport surface of the document or the reference transport surface on the light receiving sensor side , and the reference transport surface is connected to the light receiving sensor. When a document is conveyed at a position distant from each other in the direction of increasing the distance, a part of the main scanning direction of the document can be read by two adjacent light receiving sensors in an overlapping manner. Duplicate of the image obtained by scanning the original by the light receiving sensor is detected, and the scanning magnification of the original is calculated from the overlapping width. Then, since the magnification of the target area of the image obtained by the light receiving sensor is corrected based on the calculated reading magnification, the distance between the reading surface of the document and the CCD sensor varies during the document transfer, and the magnification fluctuates. However, this magnification fluctuation can be corrected appropriately.

According to the invention described in the preceding paragraph (2), when the duplication of the image obtained by reading the original by the light receiving sensors adjacent to each other is not detected and there is no loss of the image obtained by the light receiving sensor adjacent to each other. , the magnification of the image of the target area from the left magnification, also a region overlapping the image is not detected, it is possible to perform appropriate correction.

According to the invention described in the previous section (3), when duplication of images obtained by reading a document by light receiving sensors adjacent to each other is not detected, an image is taken from the edge portion of the image obtained by adjacent light receiving sensors. Is determined to be 1x or magnified, and if it is determined to be 1x, the magnification of the image in the target area remains the same, and if it is determined to be magnified, the magnification of the image is reduced. Since the correction is made in the direction, it is possible to perform more accurate correction in the region where the duplication of the image is not detected.

According to the invention described in the preceding paragraph (4), duplication of images obtained by reading originals by light receiving sensors adjacent to each other is not detected , and the defect width of images obtained by the adjacent light receiving sensors is detected. If no, because the image of the target area at a reduction ratio corresponding to the defect width of the image is small contraction, the area overlapping the image is not detected it can be performed with high accuracy correction.

According to the invention described in the previous section ( 5 ), as a result of correcting the magnification of the scanned image in the main scanning direction for each position in the sub-scanning direction, the magnification changes during the transfer of the original to the entire scanned image of the original. Appropriate corrections can be made to suppress the effects of.

According to the invention described in the preceding paragraph ( 6 ), it is possible to suppress the influence of magnification fluctuation during document transport and obtain a scanned image suitable for the document size.

According to the invention described in the preceding paragraph ( 7 ), when it is determined from the feature amount of the read image that the read image is an image whose overlap width cannot be detected, the overlap detection operation is not performed.

According to the invention described in the preceding paragraph ( 8 ), even if the distance between the reading surface of the document and the CCD sensor fluctuates during the document transport and the magnification fluctuates, the magnification fluctuation can be appropriately corrected.

According to the invention described in the preceding paragraph ( 9 ), even if the distance between the reading surface of the document and the CCD sensor fluctuates during the transfer of the document and the magnification fluctuates, the process capable of appropriately correcting the magnification fluctuation is performed. It can be executed by the computer of the image reader.

It is a figure which shows the whole structure of the image forming apparatus which mounted the image reading apparatus which concerns on one Embodiment of this invention. It is a figure which shows typically the structure of the document reading part seen from the front side in the sub-scanning direction (the document conveying direction). It is a figure which shows an example of the case where a part or all of the reading surface moves up and down with a reference transport surface in between, and the distance between a reading surface and a light receiving sensor varies during transportation. It is a figure which shows another example of the case where a part or the whole of the reading surface moves up and down with the reference transport surface in between, and the distance between the reading surface and the light receiving sensor varies during transportation. It is a figure which shows still another example in the case where a part or the whole of the reading surface moves up and down with a reference transport surface in between, and the distance between a reading surface and a light receiving sensor varies during transport. It is a figure for demonstrating the state of an error at the time of performing magnification complementation using only the image data of an overlap area. For the region where the overlapping region is not detected, it is a figure for demonstrating the state of the error when the condition that the image is enlarged or reduced to the same size is added to the complement condition. FIGS. (A) to (C) are diagrams for explaining a case where the width (defective width) of the defective region in which the image is defective can be detected. It is a figure which shows that the overlap area of an image becomes wider as the distance between the reference transport surface and a light receiving sensor becomes larger as the reading surface of a document becomes. (A) to (C) are concrete explanatory views when duplication occurs in an image. It is a figure which shows an example of an image which cannot detect the overlap width even if it overlaps with the scanned image by the adjacent light receiving sensor. FIG. 3 is a functional block diagram for performing an operation related to magnification correction in the control unit shown in FIG. 1. It is a flowchart which shows the magnification correction processing with respect to the reading image data obtained by a light receiving sensor.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.

FIG. 1 is a diagram showing an overall configuration of an image forming apparatus equipped with an image reading apparatus according to an embodiment of the present invention. As shown in the figure, the image forming apparatus includes an automatic document conveying device 10, a document reading unit 20 as an image reading device, an image forming unit 30, an automatic double-sided unit 40, a paper feeding unit 50, and a paper feeding cabinet. It includes 60, an operation panel 70, a facsimile unit 90, a communication interface (I / F) unit 91, a control unit 100, and a storage unit 120.

The automatic document transport device 10 automatically transports a plurality of documents set on the document paper feed tray one by one to a predetermined document reading position set on the platen glass which is the scanning glass of the document reading unit 20. This is a known device that ejects a document image onto a document ejection tray when the document reading unit 20 reads the document image. Further, the automatic document transfer device 10 includes a document set sensor 11, and the document set sensor 11 is composed of a known tact switch, detects whether or not a document has been set, and sends the result as a signal to the control unit 100.

The document reading unit 20 scans the document image according to the size of the document conveyed to the document reading position, receives the reflected light of the light emitted from the light source on the document image as the entrance light, and receives the incident light as electricity. It is a device that converts it into a signal and sends it as image data to the control unit 100. Further, the document reading unit 20 includes a device lifting sensor 21, and the device lifting sensor 21 is composed of a known magnetic sensor, detects whether or not the automatic document transporting device 10 has been lifted, and signals the result to the control unit 100. Send as.

The operation panel 70 is a known user interface and includes a touch panel input unit 71 and a key input unit 72. Further, the operation panel 70 includes an auxiliary power switch 80. The sub power switch 80 is a switch for the user to directly instruct the transition to the sleep mode, which is the power saving operation mode.

The control unit 100 includes a CPU, ROM, and RAM, performs various data processing such as shading correction on the received read data, reads out each main scanning line in synchronization with the paper supply, and drives the laser diode. Is output, etc., and the entire image forming apparatus is controlled in an integrated manner. Further, in this embodiment, the magnification of the image data acquired by the CCD sensor is corrected, and this point will be described later.

The facsimile unit 90 is an interface for connecting to a public telephone line and transmitting / receiving image data.

The communication I / F unit 91 is an interface for connecting to an external network to which a personal computer or the like is connected. The external network includes LAN and USB.

The storage unit 120 stores image data and other data sent from the control unit 100, and is composed of, for example, a hard disk device (HDD).

The image forming unit 30 forms an image by a well-known electrophotographic method, and includes photoconductor drums 31a, 31b, 31c, 31d, exposure scanning units 32a, 32b, 32c, 32d, a transfer belt 33, and the like. Although not shown, a front door cover and a front door sensor 34 that protect these units are provided. Further, the image forming unit 30 corresponds to four colors of yellow, magenta, cyan, and black. Based on the drive signal output from the control unit 100, the laser beam generated by the exposure scanning unit 32 is exposed and scanned onto the photoconductor drum 31. The front door sensor 34 is composed of a known tact switch, detects whether or not the front door cover is opened, and sends the result as a signal to the control unit 100. The transfer belt 33 superimposes all the toner images on the photoconductor drum 31 corresponding to each color and transfers them to the paper conveyed from the paper feed unit 50.

The paper feed unit 50 includes paper feed cassettes 51 and 53 for storing paper and pickup rollers 52 and 54 for feeding out the paper, and supplies paper to the image forming unit 30.

Like the paper feed unit 50, the paper feed cabinet 60 includes paper feed cassettes 61 and 63 for storing paper and pickup rollers 62 and 64 for feeding out the paper, and passes through the paper feed unit 50. Then, the image forming unit 30 is replenished with paper.

The automatic double-sided unit 40 enables double-sided printing by temporarily switching back on the paper passing path and feeding the paper again in order to reverse the paper on which one side is printed.

FIG. 2 is a diagram schematically showing the configuration of the document reading unit 20 of FIG. 1 when viewed from the front side in the sub-scanning direction (document transport direction).

In the figure, reference numeral 200 is a reading glass (platen glass). Since the original is conveyed by the automatic document conveying device 10 with a slight gap from the scanning glass 200, there is a document conveying surface on which the original passes above the reading glass surface 201. In this embodiment, it is assumed that the target document transport surface on which the document is scheduled to pass is set as the reference transport surface 201 and exists at the position P1 in the vertical direction. The gap between the reading glass 200 and the reference transport surface 201 is secured by a spacer such as a sheet material arranged at a required portion.

Below the reading glass 200, CCD sensors (also simply referred to as CCDs) 1 to N as a plurality of optical reduction type light receiving sensors that receive a reduced image via the lens 202 are located in the main scanning direction (FIG. 2). The images on the reading surface of the document in the left-right direction) are arranged at intervals so that they can be read separately. Further, in this embodiment, the boundary portions of the reading ranges of the two CCDs adjacent to each other in the main scanning direction are arranged so as to coincide with each other on the reference transport surface 201. Therefore, when the document completely passes through the reference transport surface 201, overlapping of images read by two adjacent CCDs and image loss due to a non-read area do not occur. It should be noted that the boundary portions of the reading ranges of the two CCDs adjacent to each other in the main scanning direction may be arranged so as to coincide with each other slightly below the reference transport surface 201. In this state, when the document is transported on the reference transport surface 201, the same region of the reading surface of the document is read by two CCDs adjacent to each other in an overlapping manner.

Each CCD1 to N is composed of RGB channels arranged in the sub-scanning direction, which is the document transport direction (also referred to as the FD direction in the following description), or RGB channels and Gr channels. ..

Next, the correction of the image data of the document read by the CCDs 1 to N, which is executed by the control unit 100, will be described.

As described above, when the document completely passes through the reference transport surface 201, duplication of images read by two light receiving sensors adjacent to each other and image loss do not occur. However, in reality, a part or all of the reading surface of the original moves up and down with the reference transport surface 201 in between, and the distance between the reading surface and the CCDs 1 to N varies.

For example, as shown in FIG. 3, when the document 300 passes through the reference transport surface 201 at the reference position P1, the scanned image obtained by each CCD 1 to N becomes like I1. On the other hand, when the document 300 passes through the position of P2 above the reference position P1, the scanned image obtained by each CCD1 to N becomes like I2, and is in a reduced state as compared with the scanned image I1. ing. Therefore, enlargement correction of the scanned image I2 is required. In addition, a part of the document is read by two adjacent CCDs in an overlapping manner. For example, the CCD1 and CCD2 read the region of D1 in an overlapping manner.

Further, as shown in FIG. 4, when the document 300 is tilted downward to the left with respect to the reference transport surface 201 at the reference position P1, the scanned image obtained by each of the CCDs 1 to N becomes like I3. The scanned image I3 is in a reduced state in the area where the document 300 is above the reference transport surface 201 (for example, CCD3, CCD4), and is the same size in the region where the document 300 is on the reference transport surface 201 (for example, CCD2). The region below the reference transport surface 201 is in an enlarged state (for example, CCD1). Therefore, if the scanned image I3 is reduced, enlargement correction is required, and if it is enlarged, reduction correction is required.

Further, as shown in FIG. 5, the document 300 may be transported in a wavy manner with respect to the reference transport surface 201 in the main scanning direction due to bending or distortion of the document 300. In this case as well, the region where the document 300 is above the reference transport surface 201 is in a reduced state, and the region on the reference transport surface 201 is the same size and is below the reference transport surface 201. The area will be expanded. Therefore, if the scanned image I3 is reduced, enlargement correction is required, and if it is enlarged, reduction correction is required. Further, for a part of the region above the reference transport surface 201, a part of the region of the document is read by two adjacent CCDs in an overlapping manner.

In FIG. 5, the duplication of reading the original by two adjacent CCDs occurs at two locations, the region between CCD1 and CCD2 and the region between CCD3 and CCD4. Since the region between the CCD2 and the CCD3 is a defective region in which the image is missing, the magnification cannot be obtained accurately. When the magnification is complemented using only the image data of the overlapping region, it is expressed as an approximate value shown by a alternate long and short dash line with respect to the actual document shown by the solid line in FIG. 6, but the image data between CCD2 and CCD3 Does not exist, so the error is large.

Therefore, for the region where the overlapping region is not detected, the condition that the image is enlarged or reduced to the same size is added to the complementary condition. When this condition is added, the approximate value shown by the alternate long and short dash line in FIG. 7 is obtained, and the complement accuracy is improved.

In this way, in the area where the overlapping area is not detected, the image is enlarged or reduced to the same size to complement it, thereby appropriately correcting the magnification fluctuation due to the variation in the distance between the reading surface and the CCDs 1 to N during document transportation. can do.

However, depending on the type of image, it is possible to detect the width of the defective region (defective width) in which the image is missing, and it is possible to accurately determine the magnification to be corrected by detecting the defective width.

For example, when there is no image data between the CCD2 and the CCD3, an edge image may exist at the right end of the read image of the CCD2 and the left end of the read image of the CCD3. As an example thereof, as shown in FIG. 8A, a case where the original image has diagonal lines 301 is shown. FIG. 8B shows a state in which the right end of the read image 250 of the CCD2 and the left end of the read image 251 of the CCD3 that read the diagonal line 301 are connected. Due to the image defect, there is a step between the left and right images 250 and 251 at the joint.

In this case, approximate lines consisting of continuous lines are set for the left and right images 250 and 251 respectively, and the images are reduced so that the approximate lines overlap. The reduction ratio when the approximate lines overlap becomes the magnification correction value of the target area as it is. Alternatively, instead of setting an approximate line, as shown in FIG. 8C, at least one of the images 250 and 251 is shifted to obtain the position where the images 250 and 251 are connected, and the defect width is calculated. .. Obtain the required reduction ratio from the defect width and use it as the correction magnification.

By correcting the image in the target area with the reduction magnification obtained in this way, more accurate correction can be performed. When the right end of the read image 250 of the CCD2 and the left end of the read image 251 of the CCD3 are connected, if there is no step between the left and right images at the joint, it means that the image is read on the reference transport surface 201. In this case, the magnification is the same, and no magnification correction is required.

Next, the magnification correction of the overlapping region will be described.

As described above, when the reading surface of the document 300 is above the reference transport surface 201, duplication occurs in the images read by the two adjacent CCDs. As shown in FIG. 9, the longer the distance from the CCDs 1 to N, the wider the overlapping region indicated by the upward arrow. That is, the overlap width becomes large.

To further explain this situation with reference to FIG. 10, it is assumed that the character 302 of "oka" is present on the reading surface of the document 300 as shown in FIG. 10A at the reading boundary portion between the CCD1 and the CCD2. When the document 300 is transported on the reference transport surface 201, an image of the same size is obtained. Therefore, as shown in FIG. (B), there is no overlap between the image read by CCD1 and the image 254 read by CCD2. The image 255 when both images 253 and 254 are combined corresponds to the characters 302 of the original, and the magnification correction is unnecessary. On the other hand, when the reading surface of the document 300 is located at a position farther from the reference transport surface 201 with respect to CCD1 to N, as shown in FIG. 10C, the image 253 read by CCD1 and the image 254 read by CCD2 Duplicates occur. In this case, the overlap width is obtained from the combined image 255 of both images, the enlargement magnification is calculated from the obtained overlap width to obtain the correction magnification, and the read image of the overlap portion is corrected using this correction magnification.

By the way, even if the scanned image is duplicated by the adjacent CCD, the overlap width may not be detected. For example, in the case of an image having high frequency characteristics such as a halftone dot pattern in which black dots are continuous in the vertical and horizontal directions as shown in FIG. 11, since the feature amount of the image cannot be detected, the overlap width cannot be detected even if the overlap occurs. Further, the overlap width cannot be detected even for an image having a low frequency characteristic with no gradation change.

Therefore, in this embodiment, the feature amount of the image is detected to determine whether or not the image cannot detect the overlap width even if the overlap occurs, and if the image cannot detect the overlap width, the presence or absence of the overlap is detected. Is not done.

FIG. 12 is a functional block diagram for performing an operation related to magnification correction in the control unit 100 shown in FIG. 1, and is a reading unit 101, a feature amount detection unit 102, an image correction unit 103, an image memory 104, and an overlap width detection. It includes a magnifying power calculation unit 106, a defect width detection unit 107, a reduction magnification calculation unit 108, a magnification correction table creation unit 109, a variable magnification unit 110, a missing pixel interpolation unit 111, and an output unit 112. The feature amount detection unit 102, the image correction unit 103, the overlap width detection unit 105, the enlargement magnification calculation unit 106, the defect width detection unit 107, the reduction magnification calculation unit 108, the magnification correction table creation unit 109, the scaling unit 110, and the defect. The pixel interpolation unit 111 is functionally configured by the CPU 130 provided in the control unit 100.

The reading unit 101 is composed of a plurality of CCDs 1 to N, and outputs the reading data of the original as RGB data.

The feature amount detection unit 102 detects the feature amount of the original image data. As a result, the character area, the halftone dot area, and the photographic area are separated.

The image correction unit 103 corrects the image data such as edge enhancement and smoothing from the result of the feature amount detection unit 102, and the image memory 104 stores the output of the image correction unit 103.

The overlap width detection unit 105 detects duplication of image data for adjacent CCDs among a plurality of CCDs 1 to N, and detects the overlap width when there is duplication. At this time, from the result of the feature amount detection unit 102, the overlap width is detected only in the region where the overlap width can be detected.

The enlargement magnification calculation unit 106 obtains the enlargement magnification required for the target area from the overlap width detected by the overlap width detection unit 105.

The defect width detection unit 107 detects the defect and the defect width of the image data for the adjacent CCDs among the plurality of CCDs 1 to N. At this time, the defect width detection unit 107 is defective only when the overlap width is not detected by the overlap width detection unit 105 even though the overlap width is a detectable region from the result of the feature amount detection unit 102. Detect the width.

The reduction magnification calculation unit 108 obtains the reduction magnification required for the target region from the defect width detected by the defect width detection unit 107.

The magnification correction table creation unit 109 creates a magnification correction table in which a continuous correction magnification including the main scanning direction or the sub-scanning direction is defined from the enlargement magnification calculation unit 106 or the reduction magnification calculation unit 108. Specifically, an approximate expression is obtained using the enlargement magnification of the target area calculated by the enlargement magnification calculation unit 106 and the reduction magnification calculation unit 108, or further, and the magnification of the area other than the target area is calculated from this approximate expression. This is performed over the entire reading area in the main scanning direction to specify the continuous correction magnification in the main scanning direction. By repeating this process for each position in the sub-scanning direction, a magnification correction table in which a continuous correction magnification including the main scanning direction or the sub-scanning direction is defined is created.

The scaling unit 110 performs scaling processing of the image data stored in the image memory 104 according to the magnification correction table created by the magnification correction table creation unit.

The missing pixel interpolation unit 111 complements the data of the region that is considered to be missing from the results of the reduction magnification calculation unit 108 or the missing width detection unit 107 by using peripheral data.

The image data corrected in this way is output from the output unit 112 to the image forming unit 30 and the like.

FIG. 13 is a flowchart showing a magnification correction process for the read image data obtained by CCD1 to N. This process is executed by the CPU 130 of the control unit 100 operating according to an operation program stored in a recording medium such as a ROM (not shown).

In step S201, the feature amount is extracted using the image data read by CCD1 to N. In step S202, it is determined from the result of feature amount extraction whether or not the target region is an image having characteristics in which the overlap width can be detected. If the area is capable of detecting the overlap width (YES in step S202), the overlap width is detected in step S203, and then the process proceeds to step S204. If the overlap width cannot be detected in the region (NO in step S202), the process proceeds to step S211 without setting the magnification of the target region in step S210.

In step S204, it is checked whether or not the overlap width can be detected, and if the overlap width can be detected (YES in step S204), in step S205, the enlargement magnification is calculated from the detected overlap width. On the other hand, when the overlap width cannot be detected (NO in step S204), the defect width is detected in step S206, and it is examined whether or not the defect width can be detected in step S207.

If the defect width can be detected (YES in step S207), the reduction ratio is calculated from the detected defect width in step S208, and then the process proceeds to step S211. If the defect width cannot be detected (NO in step S207), the scaling value of the target area is set to the same size in step S209, and then the process proceeds to step S211.

In step S211 it is determined whether or not the detection of the image data from each CCD 1 to N is completed. If the detection is not completed (NO in step S211), the process returns to step S201.

When the detection is completed (YES in step S211), in step S212, a magnification correction table for correcting the entire image is calculated from the obtained magnification correction value. Then, in step S213, the magnification correction table is used to correct the magnification of the image read by each CCD1 to N.

Next, in step S214, it is determined whether or not there is a missing region (missing pixel) in the scanned image data, and if there is a missing region (YES in step S214), in step S215, missing pixel complementation is performed in the target region. And end the process. If the missing pixel region does not exist (NO in step S214), the process ends as it is.

As described above, in this embodiment, the reading magnification of the original is calculated from the overlapping width, and the images obtained by CCD1 to N are corrected based on the calculated reading magnification. If no duplication of images is detected, the images obtained by CCD1 to N are left at the same magnification or corrected in the reduction direction. Therefore, even if the distance between the reading surface of the document and the CCD sensor varies during the document transport and the magnification fluctuates, the magnification fluctuation can be appropriately corrected.

Although one embodiment of the present invention has been described above, the present invention is not limited to the above embodiment. For example, an original size detecting means capable of detecting the size of the original 300 is further provided, and the magnification of the image read by each CCD 1 to N is readjusted so as to match the original size detected by the original size detecting means. You may adjust the overall size of the data. As a result, it is possible to prevent the size of the image data read by the CCDs 1 to N from becoming larger than the original size or conversely becoming too small, and it is possible to obtain a scanned image suitable for the original size.

10 Automatic document feeder 20 Document reading unit 100 Control unit 101 Reading unit 102 Feature amount detection unit 105 Overlap width detection unit 106 Magnification magnification calculation unit 107 Missing width detection unit 108 Reduction magnification detection unit 109 Magnification correction table creation unit 110 Variable magnification unit 200 Reading glass 201 Reference transport surface 300 Original CCD1 to N light receiving sensor

Claims (9)

A plurality of optical reduction type light receiving sensors for reading an image of a document conveyed in the sub scanning direction, which are arranged at intervals in the main scanning direction, and are read by two light receiving sensors adjacent to each other in the main scanning direction. The boundary of the range is arranged so as to coincide with the reference transport surface of the document or the vicinity of the reference transport surface on the light receiving sensor side, and the position of the document is separated from the reference transport surface in the direction in which the distance from the light receiving sensor increases. A plurality of light receiving sensors that can read a part of the document in the main scanning direction in an overlapping manner by two adjacent light receiving sensors when the
Of the plurality of light receiving sensors, a duplication detecting means for detecting duplication of images obtained by light receiving sensors adjacent to each other, and
When duplication is detected by the duplication detecting means, a scanning magnification calculating means for calculating the scanning magnification of the original from the overlapping width and a scanning magnification calculating means.
A correction means for correcting the magnification of the image in the target area based on the reading magnification calculated by the reading magnification calculation means, and
An image reading device characterized by being equipped with. When the duplication detection means does not detect duplication of images obtained by scanning documents by light receiving sensors adjacent to each other and there is no loss of images obtained by the adjacent light receiving sensors , the correction means is a target. The image reading device according to claim 1, wherein the magnification of the image in the area remains the same. When the duplication detection means does not detect duplication of images obtained by scanning documents by light receiving sensors adjacent to each other, the image is magnified or equal in size from the edge portion of the image obtained by the adjacent light receiving sensors. It is provided with the same magnification / enlargement determination means for determining whether or not the image is magnified, and when the same magnification / enlargement determination means determines the same magnification, the correction means increases the magnification of the image obtained by the light receiving sensor to the same magnification. The image reading device according to claim 2, wherein the correction means corrects the magnification of the image in the target area in the reduction direction when it is determined that the image is enlarged by the same magnification / enlargement determination means. When the duplication detection means does not detect duplication of images obtained by reading originals by light receiving sensors adjacent to each other and detects a defect width of images obtained by the adjacent light receiving sensors, the correction means. The image reading device according to claim 2 or 3, wherein the image of the target area is reduced at a reduction magnification according to the defect width. The image reading device according to any one of claims 1 to 4, wherein the correction means corrects the magnification of the image in the target area in the main scanning direction for each position in the sub-scanning direction. A document size detecting means for detecting the size of the document and
After the correction of the image magnification by the correction means, the magnification of the image read by each light receiving sensor is adjusted so that the size of the image read by the corrected light receiving sensor matches the document size detected by the document size detecting means. The image reading device according to any one of claims 1 to 5, further comprising means. A feature amount detecting means for detecting a feature amount of an image read by the light receiving sensor is provided.
The duplication detecting means determines from the feature amount detected by the feature amount detecting means whether or not the image cannot detect the overlap width even if the images read by the two adjacent light receiving sensors are overlapped, and the duplication is detected. The image reading device according to any one of claims 1 to 6, wherein if it is determined that the image has a width that cannot be detected, the duplicate detection operation is not performed. A plurality of optical reduction type light receiving sensors for reading an image of a document conveyed in the sub scanning direction, which are arranged at intervals in the main scanning direction, and are read by two light receiving sensors adjacent to each other in the main scanning direction. The boundary of the range is arranged so as to coincide with the reference transport surface of the document or the vicinity of the reference transport surface on the light receiving sensor side, and the position of the document is separated from the reference transport surface in the direction in which the distance from the light receiving sensor increases. An image reading device provided with a plurality of light receiving sensors capable of overlapping and reading a part of the main scanning direction of the document by two adjacent light receiving sensors when the document is conveyed.
Among the plurality of light receiving sensors, a duplication detection step for detecting duplication of images obtained by scanning a document by light receiving sensors adjacent to each other, and
When duplication is detected by the duplication detection step, a scanning magnification calculation step of calculating the scanning magnification of the original from the overlapping width and a scanning magnification calculation step
A correction step for correcting the magnification of the image in the target area based on the reading magnification calculated by the reading magnification calculation step, and
A method of correcting a scanned image in an image reader, which comprises executing. A plurality of optical reduction type light receiving sensors for reading an image of a document conveyed in the sub scanning direction, which are arranged at intervals in the main scanning direction, and are read by two light receiving sensors adjacent to each other in the main scanning direction. The boundary of the range is arranged so as to coincide with the reference transport surface of the document or the vicinity of the reference transport surface on the light receiving sensor side, and the position of the document is separated from the reference transport surface in the direction in which the distance from the light receiving sensor increases. To a computer of an image reader equipped with a plurality of light receiving sensors that can overlap and read a part of the main scanning direction of the document by two adjacent light receiving sensors when the document is conveyed.
Among the plurality of light receiving sensors, a duplication detection step for detecting duplication of images obtained by scanning a document by light receiving sensors adjacent to each other, and
When duplication is detected by the duplication detection step, a scanning magnification calculation step of calculating the scanning magnification of the original from the overlapping width and a scanning magnification calculation step
A correction step for correcting the magnification of the image in the target area based on the reading magnification calculated by the reading magnification calculation step, and
A correction program for a scanned image in an image reader for executing.