Patent Application
20230216970
This application claims the benefit of Japanese Patent Application No. 2022-000682, filed Jan. 5, 2022, which is hereby incorporated by reference herein in its entirety.
The present invention relates to a technique for correcting inclination of image data read from an original.
A flat-bed scheme, a sheet-feed scheme, and the like are generally known as schemes for reading apparatuses that read an image of a sheet-type original. A flat-bed scheme is a scheme in which the original is placed on a flat base and scanned by moving a linear image sensor that includes the base or a light source, and a light-receiving unit being integrated. A sheet feed scheme is a scheme in which, the original is read with conveying the original, by rotating a conveyance roller with a conveyance motor, at a constant speed in a direction orthogonal to the arrangement direction of the linear image sensor that is fixed inside the apparatus. For reading a particularly large original exceeding the A3 size, a sheet feed scheme, which allows easy downsizing of the apparatus, is usually employed, and in such a case a wide image is read by conveying the original with a plurality of linear image sensors arranged in a staggered manner in which the plurality of linear image sensors are shifting at a regular interval in the conveyance direction of the original.
In either case, distortion may occur in the read image when the orthogonal relation between the arrangement direction of the sensor and the movement direction of the original or the sensor is loosen due to the placement state of the original or the mounting state of the linear image sensor. Although inclination of the original may be rectified by the placement or conveyance state of the original, inclination of the linear image sensor is particularly difficult to adjust from a state in which once they are assembled on the apparatus, and therefore needs to be electrically corrected.
To correct the inclination, there is known a method of adjusting and correcting the access position to a memory. The aforementioned method temporarily stores, in the memory, the image data of the original read in an inclined state, and reads out the data with shifting the address in accordance with the inclination when reading out the stored image data from the memory again.
In the method described in Japanese Patent No. 3740910, deviation in the sub-scanning direction due to inclination of the read image is corrected, while the readout line position in a sub-scanning direction is being changed, in reading image data, for each area that is obtained equally dividing by an integer in the main-scanning direction. However, correction of the readout line position in the sub-scanning direction is performed pixel by pixel, and therefore the number of pixels to be read in the main-scanning direction may be equally divided, provided that the maximum amount of line deviation in the sub-scanning direction due to inclination is an integer multiple of the pixel width. For example, inclination of two pixels in the sub-scanning direction can be accurately corrected by equally dividing the main-scanning pixel width by two. However, an accurate reading width in the main-scanning direction cannot be set when the maximum amount of line deviation includes a component smaller than one pixel, and thus the precision of correction may degrade.
The present invention, which has been made in view of the above-described problems, improves the correction precision on correcting inclination of an image read by an image reading apparatus.
According a first aspect of the present invention, there is provided an image processing apparatus comprising: a storage unit configured to store image data read by scanning, with a linear sensor including pixels arranged in a main-scanning direction, an original in a sub-scanning direction orthogonal to the main-scanning direction; an acquisition unit configured to acquire inclination information indicating inclination of the image data; and a readout unit configured to read out the image data from the storage unit, with shifting, in accordance with the inclination information, a line position in the sub-scanning direction by each predetermined number of pixels in the main-scanning direction, wherein the readout unit reads out the image data from the storage unit, based on the inclination information, differentiating the predetermined number of pixels between an edge in the main-scanning direction and other parts.
According to a second aspect of the present invention, there is provided an image reading apparatus comprising: the image processing apparatus described above; and a reading unit configured to read by scanning an original in a sub-scanning direction orthogonal to the main-scanning direction, with a linear sensor including pixels arranged in a main-scanning direction.
According to a third aspect of the present invention, there is provided a method for controlling an image processing apparatus including a storage unit configured to store image data read by scanning, with a linear sensor including pixels arranged in a main-scanning direction, an original in a sub-scanning direction orthogonal to the main-scanning direction, the method comprising: acquiring inclination information indicating inclination of the image data; and reading out the image data from the storage unit, shifting, in accordance with the inclination information, a line position in the sub-scanning direction by each predetermined number of pixels in the main-scanning direction, wherein the reading out reads the image data from the storage unit, based on the inclination information, differentiating the predetermined number of pixels between an edge in the main-scanning direction and other parts.
Further features of the present invention will become apparent from the following description of exemplary embodiments with reference to the attached drawings.
Hereinafter, embodiments will be described in detail with reference to the attached drawings. Note, the following embodiments are not intended to limit the scope of the claimed invention. Multiple features are described in the embodiments, but limitation is not made an invention that requires all such features, and multiple such features may be combined as appropriate. Furthermore, in the attached drawings, the same reference numerals are given to the same or similar configurations, and redundant description thereof is omitted.
The scanner 100 includes an operation unit 103, which includes a physical key, an LCD panel (liquid crystal display panel) or the like on the upper surface of the main body, and thus setting of reading conditions or inputting of the size of the original can be performed. In addition, the scanner 100 includes an upper cover 104 on the upper surface thereof, and the reading unit or the like can be accessed by opening the upper cover 104 to upside, allowing maintenance of the main body.
In the conveyance path, the original 110 passes between a glass plate 109 and an original pressing plate 111. The original pressing plate 111 serves to press the original 110 against the glass plate 109 with a predetermined pressure. A CIS (contact image sensor) 106 is a linear image sensor which includes the reading surface arranged in a direction orthogonal to the conveyance direction of the original 110. The arrangement direction is defined as the main-scanning direction, the conveyance direction of the original 110 is defined as the sub-scanning direction. In the following, description is provided assuming that the linear image sensor is the CIS 106.
The reading surface of each of the CISs 106, which is facing the glass plate 109, is designed such that the focal position of reading lies on the contact surface between the original 110 and the glass plate 109. A downstream side original conveyance roller 108 is configured to be driven by a belt (not illustrated) in association with an upstream side original conveyance roller 107, and serves to discharge the original 110, to downstream side, which has exited from the area where the original is pressed against the glass plate 109 by the original pressing plate 111. An image reading control unit 120 includes a system controller 200 described below, and also includes detection sensors, a motor (not illustrated) configured to rotate the upstream side original conveyance roller 107, a circuit board configured to control the CIS 106, the operation unit 103, and the like.
Returning to the description of
The conveyance motor 202 is controlled by the CPU 207 via a motor driver, and rotates the upstream side conveyance roller 107 and the downstream side conveyance roller 108. Outputs from the original detection sensor 105 and the edge detection sensor 112 are input to the CPU 207, and the CPU 207 determines the drive timing of the CIS 106 and performs reading control (image data acquisition), based on the change of the output signals from the sensors and the state of the conveyance motor 202.
The CIS 106 outputs the read image to the A/D converter 204 as an analog signal, and the A/D converter 204 converts the signal into a digital signal and inputs the converted signal to the system controller 200. The data converted into digital signals in the A/D converter 204 can be transmitted as image data to an external device connected by USB or LAN via the IF unit 201, after the data is subjected to a predetermined process. The power source unit 206 generates a voltage required by each unit and supplies electric power thereto.
In
The image processing unit 209 reads, by using the DMAC 210, the image data written to the memory 205 by the read image control unit 208, and corrects the inclination of the image data. For example, the CIS 106 is assembled with an inherent inclination in accordance with the mounting tolerance. The original is conveyed in a direction perpendicular to the main-scanning direction with respect to the CIS 106 including the reading surface arranged in the main-scanning direction, however, if the CIS 106 has a certain angle with respect to the main-scanning direction, distortion (inclination) may occur in the image read by the CIS 106.
As illustrated in
The starting address processing unit 213 generates, in cooperation with the address generation unit 214, the head address at the start of the line processing. The address generation unit 214 generates a reading address of image data at a constant interval from the head address in the main-scanning direction, depending on the presence or absence of correction. In case of executing inclination correction, readout is performed with shifting the readout line in the sub-scanning direction by the shift determination unit 215 as illustrated in
As a reading procedure, for example, the user places the leading edge of the original 110 such that the central part of the original lies at the center of the original feed port on the original feeding base 102, and inserts the original into the original feeding port 101 by sliding the original on the base. When the original detection sensor 105 and the edge detection sensor 112 detects that the original 110 is set, the user operates the operation unit 103 to start reading.
The read image is stored in the memory 205 by the read image control unit 208. Here, different areas are assigned for the memory area to store above described read image, and the memory area for the image data processed by the image processing unit 209. As for the memory buffer area (an area for storing a part of the image data) of the CIS 106 to which the read image control unit 208 writes data, for example, twice the number of lines of the line shift numbers in a case of occurrence of a maximum mounting tolerance in the CIS 106, are secured as a margin. In case the maximum inclination amount is four lines, an area of eight lines is secured, and the buffer is made to perform ring operation with adjusting the timings of writing by the read image control unit 208 and reading out by the image control unit 209, and thus memory space is saved.
When the read image control unit 208 has completed writing of as many lines as the maximum shifts, the CPU 207 having received the notification of the completion causes the image processing unit 209 to execute the processing. The image processing unit 209 performs reading out with shifting of the accessed memory address by one line in the sub-scanning direction, in accordance with the inclination of the image data stored by the read image control unit 208, each time a certain amount of image data is read out in the main-scanning direction. Accordingly, distortion of the image due to inclination is corrected on the last output result.
Readout of image data is performed by burst access in order to increase the transfer efficiency. It is assumed that the burst length is fixed, and a readout amount of addresses per burst (unit readout amount) is set to a value evenly divisible by the scanning width. A readout address corresponding to each burst access is generated, in which the basic calculation method is to calculate by adding the address read out at the previous timing with an amount of addresses corresponding to the burst length and the bus width. Letting K be the added value, K = [bus width] × [burst length] holds. At the same time, the amount of data read out in the main-scanning direction is counted for shifting the line and determining completion of processing of one line. In addition, the register unit 211 is set therein the pixel width in the main-scanning direction (shift readout width, in the following) for shifting in the sub-scanning direction in accordance with inclination, and the line is shifted when the amount of data read out in the main operation direction reaches the shift readout width.
The basic processing is performed for each line, with applying an individual setting of the edge only to the shift readout width at the head of each line, the precision of inclination correction is improved.
Before the operation of the flowchart is started, the register unit 211 is accessed by the CPU 207, receives parameters related to inclination correction of image data, and passes the parameters to respective blocks required for the inclination correction. The register unit 211, holding parameters as illustrated in
The inclination correction process is performed line by line. There are provided, as counters to be used in the process, a Cn that counts the number of processed lines, an address counter Cl that counts the amount of data of one line, and an address counter Cs that counts the amount of data of the shift readout width.
Simultaneously to the start of processing, an edge setting value Wi1 is set to the shift readout width Ws (S500). The reading address Ar generated in one line is provided with numbers (x = 0, 1, 2, ..., n), and the readout address are denoted by Ar(x) in the following.
The outline of the inclination correction process according to the present embodiment will be further described, referring to
In
In addition, readout of data from the buffer memory is performed with a readout width B using the burst length as a reference, as has been already described. Accordingly, as illustrated in
Returning to description of
The shift determination unit 215 determines whether or not to execute a line shift process. The address generated in the shift execution is calculated (S516a to S516e) from the scanning width and the inclination direction of one line, which is set in the register unit 211.
In
When a ring process of the data storage buffer is required as with the case of the address Ar(0), readout is performed after the ring process is performed (S519 to S520, S523). However, as has already been explained, such as x = 2 and x = 3 as well as x = 5 and x = 6 in
Specifically, there will be described a case where the inclination amount d of the scanning width L in
Whether or not to execute the concatenation process of shift data is determined based on the value of the shift counter Cs, and the concatenation process is executed when the condition Cs > Ws is satisfied. The first half data is temporarily buffered without being written by the DMAC 210, and written after the concatenation is executed when the second half data is read out (S517 to S518).
The reset process of the shift counter Cs sets Cs - Ws, which is the extend-off amount from the shift readout width (S511). In addition, since two reading addresses are generated at a same main-scanning position, the one-line data amount Cl is not incremented when generating the address of the first half data. When the data amount Cs of the shift readout width becomes identical value with the shift readout width (Cs = Ws) at a certain scanning timing such as x = 8 in
The shift readout width Ws set at the head of each line is the parameter Wi1 that is for edge, when the line is shifted once, the edge processing is completed, and the shift readout width Ws is set by replacing to the parameter Wi2 of the central portion (S512).
Subsequent processing is continued by using the same parameter, and when the number of processing pixels reaches the data amount Cl of one line with regard to the last line, the processing is terminated with one line regardless of the shift readout width Ws, Cl and Cs are reset, and Cn is incremented (S528 to S529).
In addition, along with generation of the head address Ar(0) of the next line, the shift readout width Ws is changed to Wi1.
As illustrated in
As illustrated in
The internal memory 212 has prepared therein image data buffer area covering three lines, in addition to the storage area for filter coefficients. Data of two lines are stored at the start of processing, and subsequently when data readout of the third line is started, the data of the third line is stored into the internal memory 212, and also gradation correction is executed on data of the former two lines using data of three pixels in the sub-scanning direction of the data of the third line, and then the corrected data is written to the memory 205. Upon completion of reading data of the third line, the data of the fourth line is overwritten on the memory area of the first line, and simultaneously, the gradation correction process is performed using data of the second and the third lines, and the corrected data are written to the memory 205. Similar processing continues thereafter.
The aforementioned process is repeated on each line as many processes as the required number of lines in accordance with the size of the original, and the overall processing by the image processing unit 209 is terminated (S530).
As has been described above, the aforementioned embodiment allows for accurately correcting inclination with respect to the inclination of the linear image sensor, even when the maximum deviation amount in the sub-scanning direction due to inclination is not an integer multiple of the pixel width, whereby inclination can be corrected with a higher precision.
In the present embodiment, in addition to the correction by the first embodiment, in a case in which image data from a plurality of CISs 106 are concatenated, the inclination correction reference position of each of the plurality of CISs 106 in the sub-scanning direction are adjusted in units of less than one pixel. Accordingly, the precision in concatenating image data of each of the plurality of CISs 106 are improved.
With the sheet feed scheme, reading is performed by rotating the conveyance rollers 107 and 108 with the conveyance motor, and conveying at a constant speed in a direction orthogonal to the arrangement direction of the linear sensor fixed within the apparatus. The aforementioned configuration requires concatenation of the data read by each of the CISs 106, and the mounting offset amount in the sub-scanning direction is absorbed at the timing of storing in the memory by the read image control unit.
As illustrated in
The CISs 106 have respective inclination components depending on the mounting state, and each parameters of the CISs 106 are also set in the register of the image processing unit 209. The image processing unit 209 performs processes corresponding to respective inclination for each of the CISs 106.
As in the right-hand side of
A specific calculation method will be presented below.
The offset amount in the sub-scanning direction of the adjacent CISs 106 is preliminarily set by calibration. For example, a chart having a predetermined pattern printed thereon is read by the CISs 106 while the chart is conveyed at a constant speed, and a deviation amount Yh in the sub-scanning direction is determined from the difference in reading timings of the overlapping area between the adjacent CISs 106. Next, a deviation amount Yv in the sub-scanning direction of the head pixel of the front CIS and the rear CIS is calculated from preliminarily measured inclination information of the front CIS.
First, the deviation amount in the sub-scanning direction in overlapping area at the same main-scanning position is acquired by reading the chart. Particularly, a deviation amount Yh is acquired between a position located at the last pixel of the front CIS after the number of pixels Xt of the overlapping area is subtracted from the last pixel of the front CIS, and the head pixel of the rear CIS. For example, the distance is calculated from a sensor that detects the conveyance amount of the original and the detection timing of the pattern in each of the CISs 106.
The distance Yv between head pixels can be calculated by Yv = Yh + (L -Xt/cosθ)sinθ, using the scanning width L of the CIS 106 and the inclination angle θ of the front CIS. Settings of the edge are adjusted from the value of the decimal part Yvd (Yv = Yvi + Yd) without using the integer part Yvi of Yv. When Yvd is 0, the starting position of the weighting filter need not be changed. When 0 < Yvd < 1 holds, the filtering starting position is selected from its relation with the shift readout width. For example, in a case where the number of the central shift readout widths and the filter coefficients is 8, and 0.5 is included after the decimal point of the aforementioned calculation result Yv, in order to adjust the 0.5-pixel deviation, filters are used starting from the fifth filter assuming that the shift readout width of the edge being four pixels. In the event of a line shift, the coefficient returns to the first coefficient.
Although the shift readout width is presented as being equally divided in the example described above, in a case where the shift readout width of the edge in the inclination correction of the rear CIS is different from that of the center, the width of the leading edge required for the concatenation process is set and the remaining pixel width is processed (distributed) at an end point as remainder.
Letting Wi1′ be the head width for concatenation, as illustrated in
When connecting the third CIS 106, the second and the third CISs 106 are processed similarly as the front CIS and the rear CIS, based on the second correction center calculated from the relation between the first and the second CISs 106.
Although a single-body configuration of the scanner has been described in the aforementioned embodiments, there may also be a configuration in which a printer is included as a separate unit, and a built-in system controller inside the printer executes similar image processing.
As has been described above, in the aforementioned embodiment, inclination can be accurately corrected even when the maximum deviation amount in the sub-scanning direction due to inclination is not an integer multiple of the pixel width with respect to the inclination of the linear image sensor, whereby inclination can be corrected with a higher precision. In addition, when image data from a plurality of linear image sensors are concatenated as in a case of a large-scale scanner that reads a wide original, the concatenation precision can be improved by adjusting the inclination correction reference position of each linear image sensor in the sub-scanning direction in units of less than one pixel.
Embodiment(s) of the present invention can also be realized by a computer of a system or apparatus that reads out and executes computer executable instructions (e.g., one or more programs) recorded on a storage medium (which may also be referred to more fully as a ‘non-transitory computer-readable storage medium’) to perform the functions of one or more of the above-described embodiment(s) and/or that includes one or more circuits (e.g., application specific integrated circuit (ASIC)) for performing the functions of one or more of the above-described embodiment(s), and by a method performed by the computer of the system or apparatus by, for example, reading out and executing the computer executable instructions from the storage medium to perform the functions of one or more of the above-described embodiment(s) and/or controlling the one or more circuits to perform the functions of one or more of the above-described embodiment(s). The computer may comprise one or more processors (e.g., central processing unit (CPU), micro processing unit (MPU)) and may include a network of separate computers or separate processors to read out and execute the computer executable instructions. The computer executable instructions may be provided to the computer, for example, from a network or the storage medium. The storage medium may include, for example, one or more of a hard disk, a random-access memory (RAM), a read only memory (ROM), a storage of distributed computing systems, an optical disk (such as a compact disc (CD), digital versatile disc (DVD), or Blu-ray Disc (BD)™), a flash memory device, a memory card, and the like.
While the present invention has been described with reference to exemplary embodiments, it is to be understood that the invention is not limited to the disclosed exemplary embodiments. The scope of the following claims is to be accorded the broadest interpretation so as to encompass all such modifications and equivalent structures and functions.
| Number | Date | Country | Kind |
|---|---|---|---|
| 2022-000682 | Jan 2022 | JP | national |
