JP6205904B2 - Image reading device - Google Patents

Description

  The present invention relates to a technique for performing skew correction on a read image obtained by reading a document.

  Conventionally, when a read image read by a scanner or the like is inclined with respect to a reference direction, the read image is divided into block units, and rotation correction is performed in block units to perform skew correction of the read image. It has been known. In this apparatus, a technique for switching an image memory that reads a read image in block units according to the tilt angle of the read image is disclosed (for example, Patent Document 1).

JP 2010-205143 A

  By the way, in an apparatus that reads a read image into a storage unit such as an image memory and performs skew correction, an increase in the capacity of the read image stored in the storage unit may be a problem. For example, when skew correction of a read image is performed while a document is being read, skew correction is started until a predetermined amount of the read image necessary for forming the output image after skew correction is stored in the storage unit. Can not do it. Further, even when the predetermined amount of the read image is stored, the skew correction cannot be started until the inclination angle of the read image is calculated using the predetermined amount of the read image. As a result, the capacity of the read image stored in the storage unit at the start of skew correction increases, and accordingly, the capacity allocated for skew correction increases when skew correction is performed. In other words, there has been a problem that the capacity of the storage unit has to be increased in order to secure a capacity that can be allocated to perform other processes.

  In the present specification, a technique for suppressing the capacity of a storage unit used for skew correction is disclosed.

  An image reading device disclosed in the present specification includes an image reading unit, a control unit that corrects a skew of the read image read in the main scanning direction by the image reading unit into an output image along the main scanning direction, A storage unit, wherein the storage unit includes an input area in which the scanned image is stored, and the control unit performs the skew feeding from the scanned image stored in the input area in the skew feeding correction. A setting process for setting a required capacity that is a capacity of the input area necessary for correction, a read process for reading the read image stored in the input area and selecting it as the output image, and a capacity of the input area A release instruction process for instructing the storage unit to release the area in which the read image read in the read process is stored from the input area, if larger than the required capacity; Provided with a configuration that.

  In this image reading apparatus, when correcting the skew feeding of the read image, the capacity of the input area in which the read image is stored is compared with the necessary capacity. If the capacity of the input area is larger than the required capacity, the storage unit is instructed to release the area in which the skew correction has been completed, that is, the area from which the read image has been read out from the input area. . As a result, even if the capacity allocated for the input area temporarily increases at the start of skew correction, the capacity of the input area can be reduced by releasing a part of the input area thereafter. In addition, the freed area can be allocated for other processing, and an increase in the capacity of the storage unit can be suppressed.

  In the image reading apparatus, the storage unit further includes an output area in which the output image is stored, and the control unit further includes the area instructed to be released in the release instruction process to the storage unit. An allocation instruction process for instructing allocation as the output area may be executed.

  As skew correction from the read image to the output image proceeds, the capacity required for the output area increases. In this image reading apparatus, an area released from the input area in the storage unit is allocated as an output area, so that an increase in the capacity of the output area can be compensated by a decrease in the capacity of the input capacity. Can be made substantially constant during skew correction.

  In the image reading apparatus, the read image is obtained by reading an original by an image reading unit, and the required capacity is a capacity capable of storing a read image obtained by reading an area including at least one side of the original. Also good.

  If a read image of an area including at least one side of the document is stored in the input area, an output image on the one side can be selected. Then, by updating the read image stored in the input area, output images parallel to the one side can be sequentially selected, and as a result, the entire read image can be selected as the output image. According to this image reading apparatus, it is possible to correct the skew of the read image to the output image while suppressing the capacity of the input area.

  In the image reading apparatus, the image reading unit reads the read image by repeating reading in the main scanning direction a plurality of times in the sub-scanning direction orthogonal to the main scanning direction, and the control unit performs the release instruction process. In the above, it is confirmed whether or not each line read image read by one reading in the main scanning direction is read, and for each line read image, the area where the line read image is stored is input It may be configured to instruct to release from the area.

  According to this image reading apparatus, a part of the input area can be released for each line read image that is a unit for the image reading unit to read.

  In the image reading apparatus, the read image is obtained by reading an original by an image reading unit, and the control unit further executes an initial setting process for setting an initial capacity of the input area, May be configured from the necessary capacity and a capacity capable of storing a read image obtained by reading one original.

  In this image reading apparatus, since the capacity of the input area is set wide enough to store the read image obtained by reading one original, at the start of the skew correction, the read image has the capacity of the input area. There is no problem that cannot be remembered beyond. In addition, this image reading apparatus instructs to release a part of the input area when correcting the skew of the read image, thereby suppressing an increase in the capacity of the storage unit. Can do.

  In the image reading apparatus, the control unit repeats the reading process and the release instruction process until the capacity of the input area matches the required capacity, and the capacity of the input area is set to the required capacity by the release instruction process. The input area may be used as a ring buffer.

  In this image reading apparatus, when the capacity of the input area becomes equal to the required capacity as a result of repeatedly executing the release instruction process, the input area is used as a ring buffer, thereby reducing the necessary storage capacity. The skewed image can be corrected to the output image.

  In the image reading apparatus, the control unit starts the skew correction during reading of the image reading unit, and further executes an initial setting process for setting an initial capacity of the input area. The auxiliary capacity determined from the required capacity and the processing speed of the control unit and the image reading unit may be used.

  In this image reading apparatus, in order to shorten the skew correction processing time, skew correction is started during reading of the image reading unit, and the control unit sets the initial capacity of the input area at the start of skew correction. . When the required capacity is set by the setting process during the skew correction, it seems that the initial capacity may be set to the same capacity as the required capacity. However, at the start of skew correction, the processing speed of the control unit may decrease, for example, when initial setting must be performed when reading of a read image is started. As a result, an event may occur in which the capacity of the read image read by the image reading unit exceeds the required capacity. In this case, reading of the image reading unit must be interrupted during skew correction, and skew correction is performed. In some cases, the processing time could not be shortened.

  In this image reading apparatus, in addition to the required capacity, an auxiliary capacity set based on the calculation speed of the control section and the reading speed of the image reading section is added. Therefore, at the start of skew correction, the occurrence of an event that requires reading of the image reading unit during skew correction can be suppressed, and the skew correction processing time can be shortened. Further, since the auxiliary capacity is released in the skew correction, even if the initial capacity is set using the auxiliary capacity, it is possible to suppress an increase in the capacity of the storage unit.

  In the image reading device disclosed in this specification, the capacity of the storage unit used for skew correction can be suppressed.

Block diagram showing the configuration of the image reading apparatus Block diagram showing the configuration of the skew correction unit Explanatory drawing which shows the relationship between the coordinate system after correction, the image before correction, and the image after correction Flow chart showing reading control processing Explanatory drawing which shows the image before correction | amendment in a read image The flowchart which shows the management processing in Embodiment 1. Explanatory drawing which shows the relationship between the image before correction | amendment in the buffer for input in Embodiment 1, the image after correction | amendment in the buffer for output, and the execution state of oblique light correction | amendment Explanatory drawing which shows the relationship between the image before correction | amendment in the buffer for input in Embodiment 1, the image after correction | amendment in the buffer for output, and the execution state of oblique light correction | amendment The flowchart which shows the management processing in Embodiment 2. Explanatory drawing which shows the relationship between the image before correction | amendment in the buffer for input in Embodiment 2, the image after correction | amendment in the buffer for output, and the execution state of oblique light correction | amendment Explanatory drawing which shows the relationship between the image before correction | amendment in the buffer for input in Embodiment 2, the image after correction | amendment in the buffer for output, and the execution state of oblique light correction | amendment

<Embodiment 1>
The first embodiment will be described with reference to FIGS.

1. Configuration of Image Reading Apparatus The image reading apparatus 1 has a function of performing skew correction for reading a document (not shown) and correcting a document image included in the acquired scanned image into a document image along a reference direction. Hereinafter, a document image before skew correction is referred to as an uncorrected image Gi, and a document image after skew correction is referred to as a corrected image Go. The pre-correction image Gi is an example of a read image, and the post-correction image Go is an example of an output image. Note that an arbitrary direction can be set as the reference direction, and can be set, for example, in the main scanning direction which is the reading direction of the document sheet.

  As shown in FIG. 1, the image reading apparatus 1 includes an ASIC (application-specific integrated circuit) 2 that controls each unit of the image reading apparatus 1, and a ROM 3, a DRAM 4, a conveyance unit 6, and a drive provided outside the ASIC 2. A circuit 7, a reading unit 8, a device control unit 9, an analog front end (hereinafter referred to as AFE) 10, and the like are included. The DRAM 4 is an example of a storage unit, and the reading unit 8 is an example of an image reading unit. Further, as indicated by a dotted line 5 in FIG. 1, the one including the ASIC 2, the ROM 3, and the DRAM 4 is an example of the control unit.

  The ASIC 2 includes a central processing unit (hereinafter referred to as CPU) 13 and also includes dedicated hardware circuits such as a reading correction circuit 14 and a skew feeding correction circuit 15. The ROM 3 stores various control programs for controlling the operation of the image reading apparatus 1. The CPU 13 controls each unit in accordance with the program read from the ROM 3, and uses image data reading to be described later using a hardware circuit. Processing such as processing is executed.

  The drive circuit 7 is connected to the transport unit 6 and drives a motor (not shown) based on a command from the CPU 13. The conveyance unit 6 includes a roller (not shown), and conveys a document when the roller is rotationally driven by the motor.

  The device control unit 9 is connected to the reading unit 8 and transmits a reading control signal to the reading unit 8 based on a command from the CPU 13. The reading unit 8 reads a document conveyed by the conveyance unit 6 based on a reading control signal from the device control unit 9.

  A plurality of light receiving elements (not shown) are arranged in a row in the reading unit 8. The reading unit 8 reads one line in the main scanning direction along the light receiving elements by one reading, and reads a line read image LB ( (See FIG. 5). When reading the document, the reading unit 8 repeatedly reads the document conveyed in the direction orthogonal to the main scanning direction by the conveying unit 6 a plurality of times, and reads a plurality of lines in the sub-scanning direction orthogonal to the main scanning direction. A read image B (see FIG. 5) configured by collecting the read images LB is acquired.

  The AFE 10 is connected to the reading unit 8, performs AD conversion on the read image B acquired by the reading unit 8, and transmits it to the ASIC 2. The reading correction circuit 14 performs shading correction for each line read image LB on the AD-converted read image B, and stores the read image B subjected to the shading correction in the DRAM 4. The read correction circuit 14 transmits an interrupt signal to the CPU 13 every time each line read image LB is stored in the DRAM 4.

  As shown in FIG. 2, the DRAM 4 has an input line buffer 4A for storing the read image B including the uncorrected image Gi and an output line buffer 4B for storing the corrected image Go. ing. The read correction circuit 14 stores the read image B subjected to the shading correction in the input line buffer 4A. The skew correction circuit 15 performs skew correction on the read image B stored in the input line buffer 4A, and stores the corrected image Go in the output line buffer 4B. The input line buffer 4A is an example of an input area, and the output line buffer 4B is an example of an output area.

  In addition to this, the image reading apparatus 1 includes a display unit 11 that includes an LED and a liquid crystal display and displays the status of the image reading apparatus 1 and various setting buttons, and an operation unit 12 that receives various commands from a user. Is provided.

2. Configuration of Skew Correction Circuit Next, the configuration of the skew correction circuit 15 will be described. As shown in FIG. 2, the skew feeding correction circuit 15 includes a state machine 21, an arithmetic processing unit 22, a read DMA processing unit 25, and a write DMA processing unit 26.

  As shown in the left diagram of FIG. 3, the state machine 21 designates corrected coordinates (Xo, Yo) of the corrected coordinate system in which each pixel of the corrected image Go is to be arranged. As indicated by an arrow Z1 in FIG. 3, the state machine 21 designates one line for sequentially designating the corrected coordinates of each pixel included in the corrected image Go in the direction in which the X coordinate increases. Further, the state machine 21 repeatedly designates one line in the direction in which the Y coordinate increases as indicated by an arrow Z2 in FIG. 3, and sequentially designates the corrected coordinates of each pixel included in the corrected image Go. . The state machine 21 transmits the specified corrected coordinates to the arithmetic processing unit 22. When the state machine 21 finishes specifying the corrected coordinates of each pixel included in the corrected image Go, the state machine 21 transmits an interrupt signal to the CPU 13.

  Upon receiving the corrected coordinates from the state machine 21, the arithmetic processing unit 22 calculates the uncorrected coordinates (Xi, Yi) corresponding to the corrected coordinates using the inclination angle θ of the uncorrected image Gi with respect to the main scanning direction. To do. The calculation method will be described in detail later. When the pre-correction coordinates are calculated, the arithmetic processing unit 22 associates the post-correction coordinates and the pre-correction coordinates and transmits them to the read DMA processing unit 25.

  When the read DMA processing unit 25 receives the corrected coordinates and the uncorrected coordinates from the arithmetic processing unit 22, the read DMA processing unit 25 transmits the uncorrected coordinates to the input line buffer 4 </ b> A of the DRAM 4. Then, as shown in the middle diagram of FIG. 3, pixel data at coordinates before correction of the image Gi before correction is read out from the input line buffer 4A (see arrow Z3). The read DMA processing unit 25 transmits the read pixel data to the write DMA processing unit 26 in association with the corrected coordinates.

  When the write DMA processing unit 26 receives the pixel data from the read DMA processing unit 25, as shown in the right diagram of FIG. 3, the write DMA processing unit 26 uses the pixel data as pixel data of corrected coordinates transmitted in association with the pixel data. And stored in the output line buffer 4B of the DRAM 4 (see arrow Z4). The output line buffer 4B stores the pixel data at each corrected coordinate by the state machine 21 designating the corrected coordinate of each pixel included in the corrected image Go, and as a result, the output line buffer 4B. Is stored the corrected image Go.

  As described above, in the skew correction circuit 15, the pixel data stored in association with the pre-correction coordinates in the pre-correction image Gi is associated with the post-correction coordinates in the post-correction image Go. Is converted into a corrected image Go. At this time, the skew correction circuit 15 calculates the pre-correction coordinates using the inclination angle θ of the pre-correction image Gi with respect to the main scanning direction, so that the inclination of the pre-correction image Gi with respect to the main scanning direction is suppressed. As a result, the pre-correction image Gi is skew-corrected to the post-correction image Go.

  The arithmetic processing unit 22 includes a size management unit 23 and executes management processing for managing the capacities of the input line buffer 4A and the output line buffer 4B of the DRAM 4. The management process is executed in the skew correction from the pre-correction image Gi to the post-correction image Go, and details thereof will be described later.

3. Image Reading Process Next, the image reading process will be described with reference to FIGS. The image reading process includes a process of correcting the skew correction of the uncorrected image Gi included in the acquired read image B to the corrected image Go. The image reading process is an example of skew correction. When an instruction to read a document is input from the user via the operation unit 12, the CPU 13 executes an image reading process.

  As shown in FIG. 4, the CPU 13 first instructs the transport unit 6 to transport a document, and starts transporting the document (S2). Subsequently, the CPU 13 instructs the reading unit 8 to read a document (S4). Accordingly, the reading unit 8 starts reading one line in the main scanning direction with respect to a predetermined reading range including the original.

  The CPU 13 initializes the line number N indicating the number of line read images LB stored in the input line buffer 4A of the DRAM 4 (S6), and waits for an interrupt signal to be transmitted from the read correction circuit 14. When the line read image LB is stored in the input line buffer 4A of the DRAM 4 and an interrupt signal is received from the read correction circuit 14 (S8), the CPU 13 increases the number N of lines by 1 (S10), The number N is compared with the initial storage line number SN stored in the ROM 3 (S12).

  Here, the initial storage line number SN means the number of line read images LB necessary to detect the inclination angle θ of the pre-correction image Gi with respect to the main scanning direction, and specifically, as shown in FIG. In addition, it means the number of line read images LB necessary to include the entire area of the line image L3 among the line images L1 to L4 of the four sides on the outer periphery of the pre-correction image Gi.

  The reading unit 8 is preset with a maximum size of a document that can read an image. Even when the maximum size document is tilted with respect to the main scanning direction, the maximum tilt that can be transported using the transport unit 6. The angle θmax is set in advance. The initial storage line number SN means the number of line read images LB necessary to include the entire area of at least one line image L when the maximum-size document is inclined at the maximum inclination angle θmax.

  When the line number N is smaller than the initial storage line number SN (S12: YES), the CPU 13 may not be able to detect the inclination angle θ from the read image B stored in the input line buffer 4A. To repeat the process from S8. On the other hand, when the line number N is equal to or larger than the initial storage line number SN (S12: NO), the CPU 13 determines the inclination angle θ of the pre-correction image Gi with respect to the main scanning direction from the read image B stored in the input line buffer 4A. An inclination detection process for detecting is performed (S14).

  FIG. 5 shows the entire image of the read image B including the pre-correction image Gi. When the CPU 13 executes the tilt detection process, the input line buffer 4A has a lower side than the boundary line LL. The line read image LB having the initial storage line number SN described in the above is stored. That is, the input line buffer 4A stores the read image B including the entire area of the line image L3 among the line images L1 to L4 on the four sides on the outer periphery of the pre-correction image Gi. In the read image B, the pre-correction image Gi is tilted by the tilt angle θ with respect to the main scanning direction. In the read image B, the lower left corner P1 of the pre-correction image Gi is shifted from the origin (0, 0) of the pre-correction coordinate system (Xi, Yi) by xpm in the X direction and ypm in the Y direction. is there.

  When executing the tilt detection process, the CPU 13 extracts the line image L3 from the read image B stored in the input line buffer 4A by various known techniques, and from the extraction result, xpm, The ypm, the tilt angle θ, and the document size are obtained. Specifically, the CPU 13 detects xpm and ypm from the coordinates of the right end of the extracted line image L3. Further, an angle between the direction along the extracted line image L3 and the main scanning direction is detected as an inclination angle θ.

  Further, the length Xmax of the extracted line image L3 is detected and compared with standardized document sizes such as A4 and B5 stored in the ROM 3. The CPU 13 estimates the length Ymax of the line image L4 from the comparison result between the length Xmax and the standardized document size, and estimates the document size.

  Next, the CPU 13 performs setting processing for setting a minimum storage line number MN, which is the number of line read images LB necessary for storing the entire area of the line image L3, from the obtained inclination angle θ and document size. Execute (S16). That is, the minimum storage line number MN is the line reading necessary for storing the end image RB (see the hatched portion in FIG. 5) obtained by reading the end region on the outer peripheral side of the document corresponding to the line image L3. It is the number of images LB. Specifically, it is set based on the ratio between the sine value at the inclination angle θ of the length Xmax of the line image L3 and the width of the line read image LB in the sub-scanning direction. For this reason, the minimum storage line number MN is usually smaller than the initial storage line number SN. The minimum storage line number MN is an example of the required capacity.

  Further, the CPU 13 sets the total area storage line number AN which is the number of line read images LB necessary for storing the entire area of the corrected image Go from the obtained document size (S18). Specifically, the total area storage line number AN is set based on the ratio between the length Ymax of the line image L4 and the width of the line read image LB in the sub-scanning direction. The total area storage line number AN is an example of a capacity capable of storing a read image B obtained by reading one original.

  When the above values are set, the CPU 13 determines the inclination angles θ, xpm, ypm, document size (Xmax, Ymax) obtained by the inclination detection process, the minimum number of storage lines MN, and the total area storage line number AN. This is transmitted to the skew correction circuit 15, and the skew correction circuit 15 is instructed to perform skew correction (S20).

  After instructing skew correction, the CPU 13 detects the presence or absence of a document read by the reading unit 8 using various sensors (not shown) (S22: NO). When there is no more document to be read by the reading unit 8 (S22: YES), the CPU 13 instructs the reading unit 8 to stop reading the document (S24) and instructs the conveyance unit 6 to stop conveying the document. (S26). Then, the CPU 13 waits for an interrupt signal to be transmitted from the skew feeding correction circuit 15 (S28: NO), and upon receiving the interrupt signal from the skew feeding correction circuit 15 (S28: YES), the image reading process ends. .

(Management process)
The skew correction circuit 15 executes skew correction in accordance with an instruction from the CPU 13, and each unit included in the skew correction circuit 15 performs the above-described processes. The skew correction is executed during the image reading process of the CPU 13 and further executed while the original is being read by the reading unit 8. The arithmetic processing unit 22 included in the skew correction circuit 15 executes management processing in the skew correction.

  As shown in FIG. 6, the arithmetic processing unit 22 waits for an instruction for skew correction to be input from the CPU 13 (S32: NO). The CPU 13 does not obtain values such as the inclination angle θ and the document size and stores the skew correction to the arithmetic processing unit 22 until the line read image LB having the initial storage line number SN is stored in the input line buffer 4A. Do not send instructions. During this time, as shown in the left diagram of FIG. 7, the skew correction circuit 15 is in an off state in which the skew correction is not performed, and the corrected image Go is not stored in the output line buffer 4B. 7 and 8, the hatched areas of the input line buffer 4A and the output line buffer 4B indicate areas in which no image is stored.

  As shown in the middle diagram of FIG. 7, when the line read image LB with the minimum storage line number MN is stored in the input line buffer 4A, the entire line image L3 of the pre-correction image Gi is the input line buffer. Stored in 4A. However, since the minimum storage line number MN is a value smaller than the initial storage line number SN, the skew feeding correction circuit 15 is maintained in the off state even in this case, and the corrected image Go is stored in the output line buffer 4B. Is not remembered.

  7, when the line read image LB having the initial storage line number SN is stored in the input line buffer 4A, the CPU 13 instructs the arithmetic processing unit 22 to perform skew feeding correction. Then, values such as the minimum number of storage lines MN and the total number of storage lines AN are transmitted to the arithmetic processing unit 22. As a result, the skew feeding correction circuit 15 is switched to an on state in which skew feeding correction is performed. When these values are input from the CPU 13 (S32: YES), the arithmetic processing unit 22 receives the initial value of the input storage line number NN indicating the number of line read images LB that can be stored in the input line buffer 4A, and Then, an initial setting process for setting an initial value of the output storage line number TN indicating the number of line read images LB that can be stored in the output line buffer 4B is executed (S34).

  The arithmetic processing unit 22 secures a storage area for the number NN of input storage lines including the area where the line read image LB is already stored in the storage area of the DRAM 4, and stores the storage area in the input line buffer. Used as 4A and assigned not to be used for other purposes. Further, the arithmetic processing unit 22 secures a storage area for the number TN of output storage lines in the storage area of the DRAM 4, uses the storage area as the output line buffer 4B, and does not use it for other purposes. assign.

  In the present embodiment, the initial value of the output storage line number TN is set to 1, and the total value of the initial value of the input storage line number NN and the initial value of the output storage line number TN is the minimum number of storage lines. The number of lines is set by adding the MN and the total area storage line number AN.

Next, the arithmetic processing unit 22 waits for the corrected coordinates to be input from the state machine 21 (S36: NO), and when the corrected coordinates are input (S36: YES), calculates the uncorrected coordinates. (S38). Specifically, the post-correction coordinates are rotated by the tilt angle θ, and further, the pre-correction coordinates are calculated by shifting by xpm in the X direction and ypm in the Y direction. Pre-correction coordinates (Xi, Yi) are expressed as follows using post-correction coordinates (Xo, Yo) and inclination angles θ, xpm, and ypm.
Xi = cos θ * Xo−sin θ * Yo + xpm
Yi = sin θ * Xo + cos θ * Yo + ypm

  When calculating the pre-correction coordinates, the arithmetic processing unit 22 associates the post-correction coordinates and the pre-correction coordinates and transmits them to the read DMA processing unit 25 (S40). Thereby, the read DMA processing unit 25 and the write DMA processing unit read out the pixel data at the coordinates before correction from the image before correction Gi stored in the input line buffer 4A, and the image data at the coordinates after correction of the corrected image Go. Is read out and stored in the output line buffer 4B.

  As indicated by an arrow Z1 in FIG. 3, the arithmetic processing unit 22 performs the processing from S36 until the corrected image Go of one line is stored in accordance with the designation of one line of the state machine 21 (S42: NO). repeat. Then, as shown in the lower right diagram of FIG. 7, when one line of corrected image Go is stored, a determination process for determining whether or not to release a part of the storage area of the input line buffer 4A (S44, S46) is executed.

  In the determination process, the arithmetic processing unit 22 compares the input storage line number NN and the minimum storage line number MN, and the number of lines obtained by subtracting 1 from the input storage line number NN is smaller than the minimum storage line number MN. It is confirmed whether it is larger (S44). When the number of lines obtained by subtracting 1 from the input storage line number NN is larger than the minimum storage line number MN (S44: YES), the arithmetic processing unit 22 further reads each line read stored in the input line buffer 4A. For each image LB, it is determined whether or not the line read image LB is unnecessary (S46).

  In determining whether the line read image LB is necessary or not, the arithmetic processing unit 22 extracts the range of the corrected image Go included in the line read image LB for each line read image LB stored in the input line buffer 4A. To do. Then, when the extracted range exists in each line read image LB, the arithmetic processing unit 22 further confirms whether or not all the corrected images Go within the extracted range have been read.

  The arithmetic processing unit 22 determines that the line read image LB is necessary when each line read image LB includes the corrected image Go that has not been read out within the extracted range (S46: NO). Then, the processing from S36 is repeated. On the other hand, when the extracted range does not exist or when all the corrected images Go within the extracted range are read (S46: YES), the line read image LB can be read in the subsequent processing. Therefore, it is determined that the line read image LB is unnecessary.

  In this case, as shown in black in the upper right diagram of FIG. 7, the arithmetic processing unit 22 stores the storage area in which the line read image LB determined to be unnecessary is stored in the DRAM so as to release it from the input line buffer 4A. The instructing release instruction process is executed (S48).

  Further, the arithmetic processing unit 22 instructs the DRAM to allocate the storage area of the DRAM 4 released from the input line buffer 4A as the output line buffer 4B, as shown in the lower right diagram of FIG. 7 in comparison with the middle lower diagram. The allocation instruction process is executed (S50). As a result, the input storage line number NN decreases by 1, and the output storage line number TN increases by 1 (S52). When executing the assignment instruction process, the arithmetic processing unit 22 repeats the processes from S36.

  The arithmetic processing unit 22 repeats the processes from S36 to S52 while the number of lines obtained by subtracting 1 from the input storage line number NN is larger than the minimum storage line number MN (S44: YES), and the read process and the determination process The release instruction process and the allocation instruction process are repeated. As a result, as shown in the left diagram of FIG. 8, as the uncorrected image Gi is read from the input line buffer 4A, the input line buffer 4A decreases, and the corrected image Go is stored in the output line buffer 4B. As a result, the output line buffer 4B increases.

  As shown in the middle diagram of FIG. 9, when the number of lines obtained by subtracting 1 from the number of input storage lines NN is equal to or less than the minimum number of storage lines MN (S44: NO), the arithmetic processing unit 22 performs S36 to S52. Of these processes, the processes from S36 to S44 are repeated, and the processes from S46 to S52 are not repeated. The arithmetic processing unit 22 does not change the storage capacity of the input line buffer 4A and the output line buffer 4B, and the image of the output line buffer 4B indicated by hatching in the middle and lower diagrams of FIG. 9 is not stored. The storage of the corrected image Go in the area is repeated.

  Then, as shown in the right diagram of FIG. 9, when the storage of the corrected image Go in the output line buffer 4B is completed and the skew feeding correction is completed, an interrupt signal is input from the state machine 21 to the CPU 13, and the arithmetic processing is performed. The skew correction of the skew correction circuit 15 including the management process in the unit 22 ends.

4). Advantages of the present embodiment (1) The image reading apparatus 1 of the present embodiment determines whether or not the line read image LB stored in the input line buffer 4A is necessary during skew correction. For a line read image LB that is determined to be unnecessary, that is, that is not read in the subsequent processing, the storage area in which the line read image LB is stored is stored in the input line buffer 4A. The DRAM 4 is instructed to release the memory.

  In the conventional image reading apparatus, the storage area of the input line buffer 4A is not released during the skew correction. Therefore, for example, a relatively large area is secured as the storage area of the input line buffer 4A at the start of skew correction, for example, the initial value of the input storage line number NN is set to be approximately equal to the total area storage line number AN. There was a problem. That is, in the above-described image reading apparatus, in the above-described case, the storage capacity of the DRAM 4 that can be used for applications other than the input line buffer 4A is reduced through skew correction, and the capacity of the DRAM 4 needs to be increased. It was.

  On the other hand, in the image reading apparatus 1 of the present embodiment, even when a relatively large area is secured as the storage area of the input line buffer 4A at the start of the skew correction, the input line buffer 4A is thereafter set. Some storage space is freed. Therefore, even if the storage capacity of the DRAM 4 that can be used for purposes other than the input line buffer 4A is reduced at the start of skew correction, the storage capacity can be increased thereafter, and the capacity of the DRAM 4 increases. Can be suppressed.

(2) In the image reading apparatus 1 of the present embodiment, the storage area released from the input line buffer 4A is allocated as the output line buffer 4B. In the DRAM 4, as the skew correction from the pre-correction image Gi to the post-correction image Go proceeds, the number of input storage lines NN required for the input line buffer 4A decreases, while that required for the output line buffer 4B. The number of output storage lines TN to be increased. In this image reading apparatus 1, the increase in the number of output storage lines TN required for the output line buffer 4B is compensated by the decrease in the number of input storage lines NN required for the input line buffer 4A. That is, the total number of lines of the input storage line number NN and the output storage line number TN can be made constant through the skew correction. Accordingly, the storage capacity of the DRAM 4 that can be temporarily used for purposes other than the input line buffer 4A and the output line buffer 4B is not reduced at the start of skew correction, and the capacity of the DRAM 4 is increased. Can be suppressed.

(3) In the image reading apparatus 1 of the present embodiment, when determining whether or not to release a part of the storage area of the input line buffer 4A, the input storage line number NN, the minimum storage line number MN, Compare If the number of lines equal to or greater than the minimum storage line number MN is secured as the input storage line number NN, the pixel data of the pre-correction image Gi corresponding to the post-correction image Go of one line is read from the input line buffer 4A. It is possible to prevent the skew correction from being interrupted halfway due to a shortage of the storage area of the input line buffer 4A.

(4) In the image reading apparatus 1 of this embodiment, when determining whether or not to release a part of the storage area of the input line buffer 4A, each line read image LB stored in the input line buffer 4A is determined. Judgment is necessary for each time. If it is determined that the line read image LB is unnecessary, the storage area in which the line read image LB is stored is released from the input line buffer 4A. By performing release in units of line read images LB to be read by the reading unit 8, storage and release in the input line buffer 4A can be performed in units of the same line read image LB.

(5) In the image reading apparatus 1 of the present embodiment, skew correction is executed while the original is being read by the reading unit 8. In this image reading apparatus 1, the initial value of the input storage line number NN is set substantially equal to the total area storage line number AN, and a relatively large area is secured as a storage area of the input line buffer 4A at the start of skew correction. Is done. For this reason, the storage area of the input line buffer 4A is insufficient during skew correction, and the reading of the original by the reading unit 8 is interrupted, thereby preventing an increase in the time required for skew correction. .

<Embodiment 2>
The second embodiment will be described with reference to FIGS. 9 to 11. This embodiment is different from the first embodiment in the method of setting the initial value of the input storage line number NN (see S62 in FIG. 9). Further, the storage area of the DRAM 4 released from the input line buffer 4A is not allocated as the output line buffer 4B (no S52 in FIG. 9), which is different from the first embodiment. In the following description, the same description as that of the first embodiment will not be repeated.

1. Management Process In the management process shown in FIG. 9, when the CPU 13 receives values such as the minimum storage line number MN and the total storage line number AN from the CPU 13 (S32: YES), the input storage line number NN. And the initial value of the output storage line number TN are set (S62).

  In this embodiment, as shown in FIG. 10, the initial value of the output storage line number TN is set to the total storage line number AN. Further, the initial value of the input storage line number NN is set to be the number of lines obtained by adding the minimum storage line number MN and the auxiliary line number HN. Here, the auxiliary line number HN is the number of lines determined from the processing speed of each unit included in the skew correction circuit 15 and the reading speed of the reading unit 8, and is determined in advance and stored in the ROM 3.

  When the processing speed of each unit included in the skew correction circuit 15 is faster than the reading speed of the reading unit 8, the skew correction is performed even if the initial value of the input storage line number NN is set equal to the minimum storage line number MN. Due to the processing delay in the circuit 15, the storage capacity for storing the line read image LB cannot be secured in the input line buffer 4 </ b> A, and it is difficult for the reading unit 8 to stop reading. Therefore, the auxiliary line number HN is determined to be a relatively small value.

  On the other hand, when the processing speed of each unit included in the skew feeding correction circuit 15 is slower than the reading speed of the reading unit 8, when the initial value of the input storage line number NN is set equal to the minimum storage line number MN, skewing is performed. Due to the processing delay in the correction circuit 15, the storage capacity for storing the line-read image LB cannot be secured in the input line buffer 4 </ b> A, and a situation in which reading by the reading unit 8 stops tends to occur. Therefore, the auxiliary line number HN is determined to be a relatively large value.

  As a result, in this embodiment, the initial value of the input storage line number NN is set to a value smaller than the total area storage line number AN. Therefore, in the input line buffer 4A, as shown in the upper middle diagram of FIG. 11, the input storage line number NN and the minimum storage line number MN are equalized by the release instruction process (S48) in the management process, and A state in which there is no area in which no image is stored in the input line buffer 4A is realized.

  In this case, the arithmetic processing unit 22 stores the line read image LB newly stored in the input line buffer 4A in the input line buffer 4A as shown in the upper right diagram of FIG. 11 in comparison with the middle upper diagram. Of the line read images LB having the minimum number of stored lines MN, the line read image LB stored earliest is overwritten and stored in the storage area.

  That is, when the number of lines obtained by subtracting 1 from the number of input storage lines NN is equal to or less than the minimum number of storage lines MN (S44: NO), the arithmetic processing unit 22 performs the input line buffer 4A in the subsequent processing. Is used as a ring buffer having a minimum number of storage lines MN. Thereafter, the arithmetic processing unit 22 uses the input line buffer 4A as a ring buffer as indicated by arrows Z5 and 6 in FIG. 11 until the storage of the corrected image Go in the output line buffer 4B is completed. When the storage of the corrected image Go in the output line buffer 4B is completed, the skew correction of the skew correction circuit 15 including the management process is ended.

2. Advantages of the present embodiment (1) In the image reading apparatus 1 of the present embodiment, the initial value of the input storage line number NN is set to the number of lines obtained by adding the minimum storage line number MN and the auxiliary line number HN. As described above, in the skew feeding correction circuit 15, the storage capacity for storing the line read image LB cannot be secured in the input line buffer 4 </ b> A in relation to the reading speed of the reading unit 8. There may be situations where reading stops.

  In particular, at the start of the skew correction, the CPU 13 calculates the minimum storage line number MN, the arithmetic processing unit 22 receives the minimum storage line number MN, and sets an initial value of the output storage line number TN. A situation in which the processing speed of the skew correction circuit 15 decreases tends to occur.

  On the other hand, in the image reading apparatus 1 of the present embodiment, the number of lines obtained by adding the auxiliary line number HN to the minimum storage line number MN is set as the initial value of the input storage line number NN. Therefore, it is possible to suppress the occurrence of an event that requires reading of the reading unit 8 to be interrupted during the skew correction at the start of the skew correction, and to shorten the skew correction processing time. Further, since the auxiliary line number HN is released from the input line buffer 4A in the release instruction processing in the skew feeding correction, even if the input storage line number NN is set using the auxiliary line number HN, the DRAM 4 An increase in capacity can be suppressed.

(2) In the image reading apparatus 1 of the present embodiment, the initial value of the input storage line number NN is set to a value smaller than the total storage line number AN, and stored in the input line buffer 4A at the start of skew correction. The area is suppressed to a relatively small area. Further, after the skew correction is started, when the input storage line number NN and the minimum storage line number MN become equal, the input line buffer 4A is used as a ring buffer, and after the skew correction is started. The storage area of the input line buffer 4A is suppressed to a relatively small area. That is, the storage area of the input line buffer 4A can be suppressed through the skew correction, and the increase in the capacity of the DRAM 4 can be suppressed.

<Other embodiments>
The technology disclosed in the present specification is not limited to the embodiments described with reference to the above description and drawings, and includes, for example, the following various aspects.
(1) In the above embodiment, the image reading apparatus 1 having only the scan function has been described. However, the present invention is not limited to this, and a copying machine having a scanning function and a printing function, a facsimile machine having a facsimile function, a multifunction machine having a plurality of functions such as a scanner function, a printing function, and a facsimile function may be used.

(2) In the above embodiment, the image reading apparatus 1 has the CPU 13 and the ASIC 2, and the CPU 13 executes image reading processing using a hardware circuit such as the skew correction circuit 15 in the ASIC 2 as necessary. Although the description has been given using an example in which the control unit is configured, the present invention is not limited to this. For example, if the CPU exists separately from the ASIC and the processing executed by the control unit may be executed by the CPU, the CPU does not exist in the ASIC, and the control unit is installed by a hardware circuit provided in the ASIC. Processing to be executed may be executed. Furthermore, the process which a control part performs may be performed by one or several CPU and ASIC.

(3) The program executed by the CPU 13 is not necessarily stored in the ROM 3, and may be stored in the ASIC 2 or may be stored in another storage device.

(4) In the above embodiment, the output line buffer 4B has been described using an example secured in the DRAM 4. However, the present invention is not limited to this. For example, when the image reading device 1 is further connected to an external device such as a USB or an external PC, the output line buffer 4B may be secured in the storage unit included therein.

(5) Although the first embodiment has been described using the example in which the storage area of the DRAM 4 released from the input line buffer 4A is assigned as the output line buffer 4B, the present invention is not limited to this. The storage area of the DRAM 4 released from the input line buffer 4A may be allocated for uses other than the output line buffer 4B.

4A: input line buffer, 4B: output line buffer, 15: skew feeding correction circuit, 21: state machine, 22: arithmetic processing unit, 23: size management unit, 25: read DMA processing unit, 26: write DMA processing Part, Gi: pre-correction image, Go: post-correction image, MN: minimum number of storage lines, AN: total number of storage lines, HN: number of auxiliary lines, θ: inclination angle

Claims (6)

An image reading unit;
A control unit that corrects a skew of the read image read by the image reading unit in the main scanning direction into an output image along the main scanning direction;
A storage unit;
With
The image reading unit reads the read image by repeating reading in the main scanning direction a plurality of times in a sub-scanning direction orthogonal to the main scanning direction,
The storage unit includes an input area in which the read image is stored,
The controller is
In the skew correction,
A setting process for setting a required capacity, which is a capacity of the input area necessary for the skew correction, from the read image stored in the input area;
A read process for reading the read image stored in the input area and selecting it as the output image;
When the capacity of the input area is larger than the required capacity, release that instructs the storage unit to release the area in which the read image read in the reading process is stored from the input area In the instruction processing , it is confirmed whether or not each line reading image read by one reading in the main scanning direction is read, and the line reading image is stored for each line reading image. A release instruction process for instructing to release an area from the input area ;
An image reading apparatus that executes The image reading apparatus according to claim 1,
The storage unit further includes an output area in which the output image is stored,
The controller is
Furthermore,
An allocation instruction process for instructing the storage unit to allocate the area instructed to be released in the release instruction process as the output area;
An image reading apparatus that executes The image reading apparatus according to claim 1, wherein:
The read image is an image read by the image reading unit,
The image reading apparatus, wherein the required capacity is a capacity capable of storing a read image obtained by reading an area including at least one side of the document. An image reading apparatus according to any one of claims 1 to 3 ,
The read image is an image read by the image reading unit,
The controller is
Furthermore,
An initial setting process for setting an initial capacity of the input area;
Run
The initial capacity is set from the necessary capacity and a capacity capable of storing a read image obtained by reading one original. An image reading apparatus according to any one of claims 1 to 3 ,
The controller is
Until the capacity of the input area matches the required capacity, the reading process and the release instruction process are repeated,
An image reading apparatus that uses the input area as a ring buffer when it is determined by the release instruction process that the capacity of the input area is equal to the required capacity. The image reading apparatus according to claim 5 ,
The controller is
The skew correction is started during reading by the image reading unit,
Furthermore,
An initial setting process for setting an initial capacity of the input area;
Run
The initial capacity is an image reading apparatus set from the necessary capacity and an auxiliary capacity determined from a processing speed of the control unit and the image reading unit.

Images