JP4602913B2 - Color image reader - Google Patents

Description

  The present invention relates to a color image reading apparatus.

  A color image reading apparatus, for example, a color image reading apparatus included in an image forming apparatus such as a copying machine or a facsimile, uses a line sensor to scan and read an image of a document. An image reading apparatus that reads a color document generally reads an image of a document by separating the image into RGB using three line sensors that respectively read three colors of RGB (Red-Green-Blue).

  FIG. 6 is an explanatory diagram showing, as a block, an outline of the circuit configuration of the color image reading device 9 according to the first conventional technique. The color image reading device 9 includes a document reading unit 90, a main body unit 20, and a document table 31. The document 30 is conveyed by a document conveying device (not shown) and moves on the document table 31 at a predetermined speed. The surface of the document 30 that moves on the document table 31 that is in contact with the document table 31 is irradiated with an irradiation lamp (not shown) disposed below the document table 31. Irradiation light reflected from the document surface is guided onto a CCD (Charge Coupled Device) 11 which is an example of an image sensor having the line sensor by a reduction optical system (not shown). In this way, an image of the original is formed on the CCD 11. The CCD 11 reads the imaged original image in units of lines for each RGB color, and converts it into a signal corresponding to the image. When the document 30 moves on the document table 31, the CCD 11 repeats reading in units of lines and reads the entire area of the document 30.

  The CCD 11 also includes means (not shown) for A / D converting the outputs of the line sensors 11a to 11c, and shading correction means (not shown) for correcting variations in reading sensitivity between pixels. A digital signal (image information) subjected to A / D conversion and shading correction is output as a line sensor output.

  The document reading section 90 sends the image information obtained by reading the document to the main body section 20 in synchronization with the RGB timing. The main body unit 20 processes the image information received from the document reading unit 90 by the image processing unit 22 and stores it in the image memory 23.

  The document reading unit 90 includes a CCD 11, a line memory unit G13, and a line memory unit R14 as its circuit configuration. As a hardware configuration, the CCD 11 constitutes a scanner unit together with the irradiation lamp and the reduction optical system described above. In the color image reading device 9 shown in FIG. 6, the scanner unit is fixed in position, and the document 30 is moved in the direction (sub-scanning direction) perpendicular to the line direction (main scanning direction) by the document conveying device. The image is to be read. However, at the time of reading, the document 30 may be kept still on the document table 31 without moving, and the scanner 30 may be moved horizontally along the document table 31 to scan and read the document 30.

  Here, the CCD 11 includes three line sensors 11a to 11c for three colors of RGB. The three line sensors 11a to 11c are arranged in parallel with each other at a predetermined interval in the sub-scanning direction. In the main scanning direction, each of the line sensors 11a to 11c has the number of pixels corresponding to the width of the document to be read. The reading portion A of the original 30 sequentially passes through the reading positions of the line sensors 11a, 11b, and 11c, and is read by each line sensor. That is, the time during which the image of the reading position A, which is the point of interest on the document 30, is read by the line sensor 11a, the line sensor 11b, and the line sensor 11c is different from each other. The line memory unit G13 and the line memory unit R14 are provided to delay the previously read line sensor output and to synchronize the timing of RGB signals with respect to the point of interest. The synchronized RGB signals are sent to the image processing unit 22.

  In FIG. 6, the line memory unit G13 includes a line memory for four lines, and can delay the output of the line sensor 11b by a maximum of four lines. The line memory unit R14 includes a line memory for 8 lines, and can delay the output of the line sensor 11a by a maximum of 8 lines. The delay time by the line memory unit G13 and the line memory unit R14 can be changed in units of one line in accordance with an instruction from the main control unit 21. That is, the line memory unit G13 can function as a buffer having any number of lines 1 to 4 in accordance with an instruction from the main control unit 21. Similarly, the line memory unit R14 can function as a buffer having any number of lines 1 to 8. The output of the line memory unit G13, the output of the line memory unit R14, and the output from the B line sensor 11c are sent to the image processing unit 22. Since the B line sensor 11c reads the image latest, the output from the line sensor 11c does not need to be delayed. Therefore, no line memory unit is provided at the subsequent stage of the line sensor 11c.

  Further, MTF filters 40, 41, and 42 are arranged as filter means for correcting the MTF of the image information at the subsequent stage of each RGB line memory unit, and the sharpness of the read image is increased using the MTF filter. There is something that processes to lower or lower. That is, when the read image is blurred, the image is sharpened. On the contrary, when the image is rough, the MTF filter is provided to smooth the image.

  The main body unit 20 includes a main control unit 21 that controls the document reading unit 90, an image processing unit 22 that receives image information from the document reading unit 90 and an image memory 23 described later, and performs image processing, a semiconductor memory, and the like. An image memory 23 for storing image information processed by the unit 22, an operation button for giving an instruction to the color image reading device 9 by a user, and an operation display unit 24 including a display unit for displaying information to the user such as a liquid crystal display; A printing unit 25 that outputs the image information processed by the image processing unit 22 to printing paper is included.

  FIG. 7 is an explanatory diagram showing the positional relationship between the line sensors 11a to 11c for each of the RGB colors shown in FIG. 6 and the relative output timing of each of the RGB colors. Fig.7 (a) shows the positional relationship of line sensor 11a-11c. The line sensors 11a to 11c are arranged in parallel with each other at an interval corresponding to the time difference T at a predetermined document moving speed. The leading edge of the document 30 reaches the reading position of the R line sensor 11a at a certain time. Then, after the time T has elapsed, the document 30 advances from the position 30a to the position 30b, and the leading end reaches the reading position of the G line sensor 11b. Further, after the time T elapses, the document 30 advances from the position 30b to the position 30c, and the leading end reaches the reading position of the B line sensor 11c. 7A shows the positional relationship between the original and the CCD sensor in the moving direction of the original, but it does not correspond in the main scanning direction orthogonal thereto. That is, actually, a projected image of the full width of the original 30 in the main scanning direction is projected between both ends of each line sensor (hereinafter also referred to as a CCD line sensor) through the reduction optical system.

  FIG. 7B is a time chart schematically showing the timing at which the line sensors 11a to 11c start reading the leading edge of the image 30 in units of lines. The head of the reading area of the document 30 is first read by the line sensor 11a, that is, the line sensor that reads the color of R. This corresponds to the case where the document is at the position 30a. Next, it is read by the line sensor 11b, that is, the line sensor that reads the color G with a delay of 4 lines. Thereafter, the line sensor 11c, that is, the line sensor that reads the color of B, is further read four lines later.

  The period t shown in FIG. 7B is one period of the clock indicating the reading timing for each line, that is, from the time when the reading CCD line sensor reads an image for one line to the time when the image for the next one line is read. It's time. At a predetermined document moving speed, the time difference between the line sensor 11a and the line sensor 11b and the time difference between the line sensor 11b and the line sensor 11c are both T = 4t. Here, the predetermined document moving speed is a document moving speed in the same magnification mode, that is, a mode in which the image is not reduced or enlarged. Accordingly, the difference in reading time between the R line sensor 11a and the B line sensor 11c (hereinafter abbreviated as RB) is 8t, and between the G line sensor 11b and the B line sensor 11c (hereinafter referred to as between GB). The difference in reading time is 4t. In other words, based on the distance between the line sensor 11a and the line sensor 11b and the distance between the line sensor 11b and the line sensor 11c, the document 30 is moved so that the time for the document 30 to move to the adjacent line sensor is 4t. The speed has been determined.

  There is a color image reading apparatus that can enlarge or reduce the size of an image to be formed with respect to the document size. For example, a color image reading apparatus is known in which the reduction magnification can be selected in 1% units from 25 to 99% and the enlargement magnification can be selected in 1% units from 101 to 400%. In such a color image reading apparatus, when an image is reduced, an image is read by increasing a relative moving speed of a scanner including a line sensor with respect to an original according to a reduction ratio. In this case, the reading time difference between the line sensors that read the target point on the document is inversely proportional to the moving speed. Therefore, in order to synchronize the image information of each color, it is necessary to change the delay time of the output of the line sensor according to the moving speed. For example, when the copying magnification is 50%, that is, the moving speed of the scanner unit is twice that of the normal magnification mode, the reading time difference between RBs is 1/2 of the reading time difference in the normal magnification mode, that is, 8t × 0.5 = 4t. . Similarly, the reading time difference T between GBs is 2t, which is ½ of the normal magnification mode. In order to synchronize the image information of each color, the delay time of the R and G line sensor outputs must also be halved in the same magnification mode.

  Furthermore, as described above, if the size to be reduced can be changed in increments of 1%, the moving speed needs to be changed in increments of 1%. At this time, the delay time also needs to be changed in accordance with the reduction size. For example, when the reduction ratio is 88%, the delay time needs to be a delay time obtained by multiplying the delay time in the normal magnification mode by 0.88. At this time, the delay time between RBs is 8t × 0.88 = 7.04t, and the delay time between GBs is 4t × 0.88 = 3.52t. That is, it is necessary to delay the output of the R line sensor 11a by 7.04t and delay the output of the G line sensor 11b by 3.52t.

  As for the R image information, the image information delayed by 7 lines can be obtained from the line memory unit R14, so that the image information substantially equal to the delay amount of 7.04t is obtained. However, for the G image information, it is not possible to obtain image information delayed by 3.5 lines from the line memory unit G13. This is because the image of the document 30 is read in units of lines, so that information can be obtained only in increments of the period t. Image information of 3t or 4t whose delay time is an integral multiple of the period t is obtained from the line memory unit G13. Therefore, for the delay time that is not an integral multiple of the reading cycle t, the image information is obtained by interpolating by calculation from image information whose delay time is an integral multiple of the reading cycle t before and after that. In this example, image information with a delay time of 3.5 t is calculated as interpolation data from image information with a delay time of 3 t and image information with a delay time of 4 t.

  FIG. 8 is a block diagram showing a more detailed configuration of the line memory unit R14 shown in FIG. The line memory unit R14 includes eight line memories 171, a selector A 172, a selector B 173, and an interpolation data creation unit 174.

  The eight line memories 171 are connected in series with each other, and delay the output from the CCD line sensor 11a input to the first stage line memory by one line, that is, the reading cycle t. The output of the first line memory is input to the next line memory. Thereafter, the output of the previous line memory is input up to the eighth line memory. In both line memories, the input is delayed by one line, that is, by the reading cycle t.

  The output from the CCD line sensor 11a and the outputs from the eight line memories 171 can be selectively input to the selector A172 and the selector B173. The selector A 172 and the selector B 173 are respectively selected from a total of nine inputs of the CCD line sensor 11 a and the eight line memories 171 according to “select value 1” and “select value 2” instructed from the main control unit 21. One input is selected and output. The outputs of the selector A 172 and the selector B 173 are input to the interpolation data creation unit 174.

  The interpolation data creation unit 174 generates and outputs interpolation data of the output “value A” of the selector A 172 and the output “value B” of the selector B 173 based on the “weighting value” instructed from the main control unit 21. . Specifically, “value A” is multiplied by (1− “weighting value”), “value B” is multiplied by “weighting value”, and the sum thereof is obtained as interpolation data. Here, the “delay amount” is a value obtained by dividing the delay time by the period t. In this case, the “weighting value” is a decimal part of the delay amount. The unit of “delay amount” is a line. That is, the weighting value corresponds to a value less than the decimal point in the coefficient of t of the delay time between RBs or the delay time between GBs described above. The “weighting value”, “select value 1”, and “select value 2” corresponding to R and G at each magnification are calculated in advance at the design stage of the color image reading apparatus. Therefore, the value obtained by the calculation is stored in the main body 20 as a data table that can be referred to by the main control unit 21, and the main control unit 21 refers to the table according to the enlargement / reduction ratio, “Value”, “select value 1”, and “select value 2” may be indicated.

  For example, when the delay amount is 3.7 lines, that is, the delay time is 3.7 t, the main control unit 21 indicates that the selector A 172 selects the output of the line memory delayed by 3 lines. “Value 1” is instructed, and “select value 2” indicating that the output of the line memory delayed by 4 lines is selected to the selector B 173. The selector A 172 selects the output of the third line memory and outputs it as “value A”, and the selector B 173 selects the output of the fourth line memory and outputs it as “value B”.

  Further, the main control unit 21 instructs the interpolation data creation unit 174 to “weight”, for example, “0.7”. Assuming that the output “value A” of the selector A 172 and the output “value B” of the selector B 173 are, for example, “100” and “50”, respectively, the interpolation data creation unit 174 selects 30% of the output “100” of the selector A 172 and the selector Interpolation data is obtained by adding 70% of the output “50” of B173. That is, the interpolation data becomes “65” from the calculation formula 100 × 0.3 + 50 × 0.7. The interpolation data creation unit 174 outputs the R interpolation data obtained as described above.

  The above description relates to the line memory unit R14. The configuration of the line memory unit G13 is the same as that of the line memory unit R14 except that the number of line memories is four, and “select value 1” and “ In accordance with the select value 2 ”, an output selected from a total of five inputs of the CCD line sensor 11b and the four line memories is interpolated to obtain an interpolation output for G.

  Further, the above explanation is about the time of reduction, but there are also things that change the reading speed according to the enlargement magnification even at the time of enlargement. The document moving speed at the time of enlargement is slower than that at the same magnification. In such a case, interpolation processing is required even during enlargement.

Further, as a second conventional technique, the relative positional shift of the three CCD line sensors that convert the light separated into the R component, the G component, and the B component into an electrical signal is determined according to the positional shift amount. There is a color image forming apparatus capable of correcting an electrical signal from each CCD line sensor by delaying and forming an output image without color misregistration (see, for example, Patent Document 1).
Japanese Patent Laid-Open No. 5-75792

  The first conventional technique is to delay the output of the R and G color line sensors based on the B color line sensor that reads the image the latest. When the delay amount is not an integer, interpolation is performed on the outputs of the R and G line sensors. That is, interpolation is performed according to the reduction magnification. However, when interpolation is performed, the sharpness of each color image decreases. The decrease in sharpness is greatest when the weighting value is 0.5. This is because the above-described interpolation processing obtains the weighted average of the image information of the preceding and succeeding lines, so that the steepness of the edge portion across the lines is lost. In particular, a thin line with a narrow interval between both edges, a flat portion with a high density portion, and a peak has a peak that has a different degree of sharpness. Recognized as a change. For example, in a small black character, the thin line portion appears to be colored, and conversely, the thin line portion of the color character becomes black. These are recognized as image quality degradation. In the first prior art, the latest reading sensor, for example, the B line sensor 11c in the example of FIG. 6, does not have a line memory unit, so that the output from the line sensor is not interpolated. On the other hand, since the R and G line sensor outputs are subjected to interpolation processing with respective weight values, the degree of reduction in sharpness due to the interpolation processing is different for each color. As a result, there is a problem that image quality deterioration is easily recognized.

  For example, in the table of FIG. 3 showing the delay amount with respect to each reduction magnification, the delay amount of G in the prior art at a copy magnification of 88% is 3.52, and its weight value is 0.52. The delay amount of R is 7.04, and its weight value is 0.04, which is close to zero. Further, since the delay amount of B is always zero, the weighting value is always zero. In this case, the weight values for R and B are zero or close to them, while the weight value for G is approximately halfway between the values of the previous line and the subsequent line, and interpolation takes the average value of the previous and subsequent lines. Therefore, only the sharpness of G is lost and the image quality is deteriorated.

  The second conventional technique reduces the occurrence of color misregistration due to misregistration during the manufacture of the CCD line sensor by delaying the output of each CCD line sensor in accordance with the misregistration amount. This delays the output in accordance with the positional deviation between the CCD line sensors, and does not adjust the reading time difference between the CCD line sensors. Further, the delay amount of each line sensor with respect to the reference line sensor is not obtained according to the reduction magnification.

  The present invention has been made in view of such circumstances, and even when interpolation processing is performed, such as during zooming, suppresses the difference in the degree of sharpness reduction in each color and recognizes the reduction in image quality due to this. It is an object of the present invention to provide a color image reading apparatus that is difficult to perform.

  The present invention includes a parallel line sensor for each color component for reading an image of a color document including a plurality of color components, a drive unit that causes each line sensor to read an image for each line at a predetermined period, and each line sensor; A position moving unit that relatively moves the color original in a direction orthogonal to the line sensor at a predetermined moving speed, a moving speed determining unit that determines the moving speed, and an output of a signal read by each line sensor are set. A line delay unit that delays according to the delay amount, and outputs from each image sensor are delayed and synchronized with each other according to the arrangement interval of each line sensor and the moving speed, and the delay amount from reading each output is read A control unit that determines a delay amount including a fraction of a cycle so as to be substantially uniform with respect to an integral multiple of the cycle, and sets the determined delay amount in each line delay unit; To provide a color image reading apparatus characterized by obtaining.

  According to the color image reading apparatus of the present invention, the outputs from the image sensors are delayed and synchronized with each other according to the arrangement interval of the line sensors and the moving speed, and the delay amount from the reading of each output is an integer of the reading cycle. Since it includes a control unit that determines a delay amount including a fraction of the period so as to be substantially evenly shifted from each other and sets the determined delay amount in each line delay unit, from an integral multiple of the read cycle in line units The reduction in the sharpness of the image depending on the shift amount is uniformized for each color component, so that the deterioration in image quality such as a color change of a thin line can be made inconspicuous.

  The control unit may determine each delay amount so that a fraction with respect to the period does not become a half of the period. When the fraction is ½ of the period, the amount of deviation from the integral multiple of the period is the largest, and the sharpness of the image is the most reduced. According to the present invention, the condition that the sharpness of the image is the lowest is avoided. can do.

  The moving speed is determined according to the setting of the reading magnification, and further includes a delay amount table in which the delay amount of each output with respect to each reading magnification is stored in advance, and the control unit sets the delay amount table according to the set reading magnification. The delay amount of each output may be determined with reference to FIG.

  In this way, since the delay amount of each output can be obtained by referring to the table, it is not necessary to calculate the delay amount based on the magnification every time the reading magnification is set, and the processing load on the control unit is reduced. Is done. In other words, high performance is not required for the control unit, and the cost of a CPU or the like that realizes the processing unit can be reduced.

In order to change the sharpness of the read image, an MTF correction filter that performs MTF correction on the output from each line sensor is further provided, and the delay unit uses the MTF filter to delay the fractional delay with respect to the reading cycle for each line. You may do it.
In this way, since the delay process finer than the reading cycle of each line can be realized using the MTF filter provided for each output in order to adjust the sharpness of the image, the circuit is simplified and the cost is reduced. Can be reduced. In particular, since the output from the line sensor that reads the document reading position last does not need to be delayed in order to synchronize with other outputs, the amount of deviation from the reading cycle for each line can be realized only by the MTF filter. it can.
Other outputs may use an MFT correction filter, but a line memory for delaying a plurality of lines in units of lines and a delay unit having a function of interpolating a fraction of the reading cycle are provided separately from the MTF correction filter. May be.

  Hereinafter, the present invention will be described in more detail with reference to the drawings. The following description will provide a better understanding of the present invention. In addition, the following description is an illustration in all points, Comprising: It should be thought that it is not restrictive.

  FIG. 1 is an explanatory diagram showing, as a block, an outline of a circuit configuration of a color image reading apparatus 1 according to an embodiment of the present invention. A color image reading apparatus 1 shown in FIG. 1 is a single color image reading apparatus such as a scanner apparatus or a color image reading apparatus included in an image forming apparatus such as a copying machine or a facsimile. The parts corresponding to those of the color image reading device 9 in FIG. The color image reading apparatus 1 in FIG. 1 is different from the conventional color image reading apparatus 9 in FIG. 6 in order to delay the output from the line sensor 11c that reads the target point of the original 30 the latest within one line period. The MTF filter 42 is provided after the line sensor B. The MTF filter 42 functions as an interpolation processing unit for the line currently being read and the previous line. Here, when the MTF filters 40, 41, and 42 are arranged for the respective RGB colors in order to change the sharpness of the image, the B MTF filter 42 may be used as the storage processing unit. In FIG. 1, since the line memory unit R14 and the line memory unit G13 include an interpolation processing unit, the R and G MTF filters 40 and 41 are not essential elements. Accordingly, these are indicated by dotted lines in FIG.

  The CCD 11 and a scanner unit (not shown) correspond to the image sensor unit described in the claims. A document conveying device (not shown) corresponds to the position moving unit described in the claims. The line memory unit R14, the line memory unit G13, and the MTF filter B42 correspond to the line delay unit described in the claims. The main control unit 21 corresponds to the control unit described in the claims.

  As described above, when the image is reduced, reading is performed by increasing the relative moving speed between the document 30 and the CCD 11. Therefore, the delay time of the R and G line sensors with respect to the B line sensor 11c is the read timing for reading the image. , Which is an integral multiple of the read cycle t, that is, a time that is an integral multiple of the time from when the line sensor reads an image for one line until the next image for one line is read. The output of the line sensor whose delay time is not an integral multiple of the reading cycle t needs to be interpolated.

  Each of the line sensor unit G13 and the line sensor unit R14 has a configuration corresponding to FIG. 6, and includes the selector A, the selector B, and the interpolation data generation unit illustrated in FIG. Therefore, the output of any line sensor can be interpolated.

  Assuming that the reference line sensor serving as a reference for the delay time, that is, the B line sensor 11c that reads the image latest, interpolation is performed by the R and G line sensors. Since the delay time of the reference line sensor is a time that is an integral multiple of the reading cycle t, no interpolation is necessary, and the interpolation is performed because the delay time for the reference line sensor is not an integral multiple of the reading cycle t. It is. In this description, it is the B line sensor that reads the image the latest, but the present invention is not limited to this, and an R or G line sensor may be used. In this case, an 8-line line memory unit is arranged after the line sensor that reads the image first, and a 4-line line memory unit is arranged after the line sensor that reads the image.

  In this embodiment, the delay amount for each reduction ratio for each RGB color is determined as follows. First, the delay amount of each line sensor is calculated according to the reduction magnification of the image based on the line sensor B that reads the image the latest. This process is the same as that of a conventional color image reading apparatus. Specifically, first, the delay amount 8 of the R line sensor 11a and the delay amount 4 of the G line sensor 11b in the equal magnification mode are respectively multiplied by the reduction ratios to calculate the delay amounts for R and G. The delay amount of B is zero. From the obtained weight values, the weight values for each of the RGB colors are obtained. As described above, the weighting value corresponds to the decimal part of the delay amount.

  Next, for each of the RGB colors, a difference from 0.5 of the weighted value is obtained, and a certain offset is added to the delay amount of each color so that the minimum value of each difference becomes the largest. Here, the offset amounts of the respective RGB colors are made equal so that the relative delay amounts of the respective colors are not changed. The purpose of increasing the difference of the weighting value from 0.5 is to suppress the reduction in sharpness as much as possible. However, the weighting value of B is set to be in the range of 1 or less. B is the line that reads the image the latest, and is therefore the line currently being read.

  FIG. 3 and FIG. 4 are tables showing the reduction amounts and weighting values of the respective colors with respect to each reduction magnification and equal magnification (100%) that can be set in units of 1% from 50 to 99% in the color image reading apparatus of the present invention. It is an example of (delay amount table). 3 shows a case of 50 to 74%, and FIG. 4 shows a case of 75 to 100%. Note that FIGS. 3 and 4 are examples, and a table may be created as necessary for a reduction ratio or an enlargement ratio smaller than this. 3 and 4 show each magnification in the vertical direction, and each column in the horizontal direction indicates the delay amount of each RGB color according to the prior art, each delay amount according to the embodiment of the present invention, and each color according to the embodiment of the present invention. The weighting value, the difference of each weighting value from 0.5, and the minimum value of the difference are shown.

  For R and G, interpolation processing is performed based on the determined weight value. The details of the interpolation processing are as already described. On the other hand, for B, smoothing processing by the MTF filter 42 is performed based on the determined weighting value.

  A case where the reduction ratio is 88% will be described as an example. In the conventional technique, the delay amount between RBs is 7.04, the delay amount between GBs is 3.52, and the delay amount of B is zero. Therefore, the weighting values according to the prior art are 0.04 for R, 0.52 for G, and zero for B. Then, the smallest difference from 0.5 is 0.52-0.5 = 0.02 of G. This value is very small compared to the difference of other colors, that is, the difference of R 0.5-0.04 = 0.46 and the difference of B 0.5-0 = 0.5. Therefore, at a reduction ratio of 88%, only the sharpness of G is greatly reduced, and the above-described deterioration in image quality such as the color change of the thin line is conspicuous.

  FIG. 2 is an explanatory diagram showing an output transition in the sub-scanning direction of the output value of each line memory unit shown in FIG. Indicates the output transition of each line memory unit when a black thin line or rectangular image (hereinafter abbreviated as BK image) whose width in the sub-scanning direction is several pixels is read by each RGB line sensor. ing. FIG. 2A shows an ideal case where the sharpness of all RGB, in other words, the spatial frequency characteristics are equal.

  The width of the BK image is set such that each RGB output value shows the lowest peak near the center of the BK image. In such an image, a color change due to variations in sharpness appears remarkably.

  FIG. 2B shows the characteristics corresponding to FIG. 3 when the copy magnification is 88%. In this case, only the sharpness of the output of the G line sensor 11b is reduced by the interpolation, and as a result, the depth to the valley of the G image information becomes shallower than R and B.

  Since the output of the R line sensor has a small weighting value, there is almost no decrease in sharpness due to interpolation. Since the output of the B line sensor is not interpolated, there is no reduction in sharpness due to the interpolation. Then, an output value as if there are many G components is obtained near the center of the BK image. When developing and printing with RGB complementary color toner based on the output value, the M (Magenta) component is developed less than the C (Cyan) component and the Y (Yellow) component. As a result, the printed image is not black, and the image is biased to green because the M component is smaller than the C component and the Y component.

  As shown in FIG. 3, in this embodiment, an offset of 0.22 is added to the weight values of each RGB color, so that the weight values of each color are 0.22 for B and 0.74 for G. , R is 0.26. At this time, the difference of each weighting value from 0.5 is 0.28 for B and 0.24 for G and R. Therefore, the minimum value of the difference is 0.24 and the maximum value is 0.28.

  Comparing this with the prior art, the difference from the weighting values of 0.5 for each color is 0.5 for B, 0.02 for G, and 0.46 for R, and the variation in the difference is between B and G. In this embodiment, the difference from 0.5 is 0.28 for B, G and R are 0.24. The difference variation is suppressed to a difference between B and G or R of 0.28−0.24 = 0.02. Therefore, the color change of the BK image is not noticeable.

  In addition, the color with the lowest sharpness, that is, the color with the weight value closest to 0.5, is G in the prior art, and the difference from the weight value of 0.5 is only 0.02. On the other hand, in this embodiment, G or R, the difference of the weighting value from 0.5 increases to 0.24, and the reduction in sharpness is suppressed for the worst condition color.

  FIG. 2C shows the output transition of the BK image according to this embodiment. As shown in FIG. 2C, the output of each color of RGB is substantially equal near the center of the BK image.

  In this embodiment, the change in the sharpness with respect to the weighting value is caused by the interpolation processing by the line memory unit R13 and the line memory unit G14 for the R and G line sensor outputs, as in the prior art. On the other hand, for the line sensor output of B that does not have a line memory unit, the MTF filter 42 is used to perform a smoothing process commensurate with the sharpness reduction of the R and G lines, and the RGB output values are near the center of the BK image. The values should be approximately the same at the position. Therefore, a decrease in the sharpness of the G color is suppressed, and variation in the sharpness of each color is suppressed by bringing the sharpness of the R and B colors close to the sharpness of the G color. It is possible to suppress image quality deterioration, particularly changes in the color of thin lines.

  The main control unit 21 may calculate a weight value for each of the RGB colors according to the magnification set before reading the image. Alternatively, the main control unit 21 includes a semiconductor memory such as a ROM (Read Only Memory) (not shown), the delay amount tables of FIGS. 3 and 4 are stored in the semiconductor memory, and the table is set according to the magnification set by the user. The weight value of each color may be acquired with reference to FIG.

  When a copy magnification instruction is given from the user using the operation button of the operation display unit 24, the main control unit 21 reduces the reduction when the designated copy magnification is a reduction magnification registered in the delay amount table. Based on the magnification, the delay amount table is referred to obtain a weighting value for each line sensor and instruct the line memory unit of each color. In this way, it is not necessary for the main control unit 21 to calculate the delay amount each time the copy magnification is instructed.

  FIG. 5 is an explanatory diagram showing details of the function of the MTF filter 42 according to the present invention. As shown in FIG. 5A, the MTF filter has one pixel before and after in the main scanning direction (previous pixels a1, a4, a7, subsequent pixels a3, a6, a9) with respect to the target pixel a5. This is a 3 × 3 filter that operates on the front and rear one pixel (previous pixels a1, a2, and a3, and subsequent pixels a7, a8, and a9) in the sub-scanning direction. The configuration of the MTF filter may be realized by applying a known one. The MTF filter can set the sum of products obtained by multiplying the pixel values a1 to a9 to be calculated by the filter coefficients k1 to k9, respectively, as an output value for the target pixel. Each filter coefficient is set by the main control unit 21.

That is, when the output value of the pixel of interest a5 is a5 ′, a5 ′ is obtained from the following product-sum operation.

In this embodiment, the MTF filter 42 is used for delay processing. In order for the MTF filter 42 to function as a delay unit, a filter coefficient may be set so as to function as a smoothing filter. That is, the main control unit 21 sets the filter coefficients k1, k3, k4, and k6 to k9 to zero, and sets k2 and k5 to values according to the B weighting value. That is, a weighting value of B is set for k2, and (1- (B weighting value)) is set for k5. In this case, the output value a5 ′ of the target pixel a5 is

It is. This expression corresponds to an arithmetic expression for performing interpolation processing from the preceding and succeeding lines in the line memory unit R14 or the line memory unit G13. That is, since the B line is interpolated between the line being read and the line read before, the filter coefficient of the MTF filter is the line being read and the line read before that so as to satisfy the condition. The arithmetic processing is performed by setting the coefficient between the two to a non-zero value. a2 is the previous line of the interpolation process, a5 is the line being read, and the filter coefficient k2 corresponds to the weighting value of the interpolation process.

  FIG. 5B shows, as an example, each filter coefficient when the weighting value of B is zero. In this example, the main control unit 21 sets k2 to zero and k5 to (1-0) = 1.

  FIG. 5C shows each filter coefficient when the weighting value of B is 0.2 as a different example. In this example, the main control unit 21 sets 0.2 for k2 and (1-0.2) = 0.8 for k5.

  As described above, the value of the filter coefficient set in the MTF filter 42 is determined based on the weighting value for B.

  Finally, it is apparent that there can be various modifications of the present invention in addition to the above-described embodiment. Such variations are not to be construed as not belonging to the features and scope of the invention. The scope of the present invention is intended to include all modifications within the meaning and range equivalent to the scope of the claims.

1 is an explanatory diagram showing an outline of a circuit configuration of a color image reading apparatus according to an embodiment of the present invention as a block; FIG. FIG. 2 is an explanatory diagram illustrating output transition in the sub-scanning direction of output values of each line memory unit illustrated in FIG. 1. In the color image reading apparatus of the present invention, it is an example of a table showing a delay amount and a weight value of each color for each reduction magnification that can be set in units of 1% from 50 to 74%. In the color image reading apparatus of the present invention, it is an example of a table showing the reduction amount and weight value of each color for each reduction magnification and equal magnification that can be set in units of 75% to 99%. It is explanatory drawing which shows the detail of the function of the MTF filter which concerns on this invention. It is explanatory drawing which shows the outline | summary of the circuit structure of the color image reading apparatus by a 1st prior art as a block. It is explanatory drawing which shows the positional relationship of the line sensors 11a-11c shown in FIG. 6, and the time relationship of an output. FIG. 7 is a block diagram showing a more detailed configuration of the line memory unit R14 shown in FIG. 6.

Explanation of symbols

1,9 Color image reading device 90 Document reading unit 11 CCD
12 Line sensor unit B
13, 16 Line sensor unit G
14, 17 Line sensor unit R
18 MTF filter 15 Setting unit 20 Main unit 21 Main control unit 22 Image processing unit 23 Image memory 24 Operation display unit 25 Printing unit 30 Document 31 Document table 40 MTF filter R
41 MTF filter G
42 MTF filter B
171 Line memory 172 Selector A
173 Selector B
174 Interpolation data creation unit

Claims (1)

A plurality of line sensors that extend in the main scanning direction and are arranged in parallel in the sub-scanning direction orthogonal to the main scanning direction, scan the document at a predetermined cycle for each color component by scanning at a predetermined speed in the sub-scanning direction; A correction filter that corrects the MTF of the reference signal output from the reference line sensor that reads the same position of the document after any other line sensor among the line sensors, and a line delay that delays signals from the other line sensors And
MTF correction filters that respectively correct MTFs related to signals from other line sensors in order to change the sharpness of the read image;
A control unit for setting a correction amount of the correction filter and a delay amount for each color component of the line delay unit,
The control unit determines a delay amount including a fraction of the period so that the reference signal and signals from other line sensors are substantially uniformly shifted with respect to an integral multiple of the period, and the determined delay amount Based on this, an integral part of the delay amount of the signal from the other line sensor is set in the line delay unit, and a correction amount corresponding to a fraction of the delay amount of the reference signal is set in the MTF correction filter . A color image reading apparatus , wherein a correction amount corresponding to a fraction of the delay amount is set in the correction filter .

Images