JP3698196B2 - Image processing apparatus and image input apparatus - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to an image processing apparatus that scales image data at an arbitrary scaling ratio, and an image input apparatus that uses the image processing apparatus.
[0002]
[Prior art]
Conventionally, many image processing apparatuses that perform image data enlargement processing in the sub-scanning direction store image data once in a page memory and then sequentially read out necessary lines. For example, in the image processing apparatus described in Japanese Patent Application Laid-Open No. 11-122481, an enlargement processing unit is provided at a subsequent stage of the memory unit, and an image signal read from the memory unit is enlarged, and an enlarged image is output. In such an apparatus, an operation of once writing an image signal for one page written in a memory and then reading out occurs, so that it takes a long time to obtain an output image. Furthermore, a memory that can store an image for one page must be provided, which causes an increase in cost.
[0003]
On the other hand, image scaling is realized by using a line memory that operates in synchronization with clocks having different frequencies for writing and reading, so that an output image can be obtained in real time without requiring a page memory. There is an image processing apparatus. For example, in the pixel number conversion device described in Japanese Patent Application Laid-Open No. 10-207432, writing to the line memory is performed in synchronization with the write clock, and the read clock has a frequency based on the conversion ratio of the number of pixels corresponding to the frequency of the write clock. Image data scaling is realized by synchronously reading from the line memory. However, in order to realize scaling with an arbitrary and fine scaling factor as in a digital copying machine by using such a method, a clock generating unit capable of generating a clock with an arbitrary frequency is provided, or a plurality of clock generating units are provided. There was a problem that it was necessary to prepare and the cost was high.
[0004]
[Problems to be solved by the invention]
The present invention has been made in view of the above-described circumstances, and does not require a page memory when scaling image data, and is capable of realizing scaling at an arbitrary and fine scaling ratio in real time. Is intended to provide. It is another object of the present invention to provide an image input apparatus using such an image processing apparatus.
[0005]
[Means for Solving the Problems]
The present invention relates to an image processing apparatus for performing a scaling process of image data, wherein the image data is written to the first line memory with a first clock and is determined based on at least the frequency and the maximum magnification of the first clock. Read from the first line memory and output with the second clock. Further, the image data output from the first line memory is written into the second line memory with the second clock, read out with the second clock, and output. For example, when the maximum magnification is an integer multiple, the frequency of the second clock may be an integer (N) times the frequency of the first clock. In this case, while writing one line of image data into the first line memory, the image data is output once from the first line memory and (N−1) times from the second line memory. Therefore, a line for performing N-times enlargement processing in the sub-scanning direction can be supplemented. The frequency of the second clock may be determined in consideration of the scaling factor in the main scanning direction.
[0006]
The processing means receives the image data supplemented with the lines from the first line memory and the second line memory based on the maximum magnification (or obtained in consideration of the variable magnification) in this way, A scaling process in the scanning direction is performed. For example, by alternately referring to the output from the first line memory and the output from the second line memory, for example, by thinning out unnecessary lines, image data that has been scaled in the sub-scanning direction at an arbitrary scaling factor is obtained. Can do. At this time, a new scaling factor is generated based on the set scaling factor and the frequency ratio between the first clock and the second clock, and scaling processing is performed according to the generated scaling factor. .
[0007]
For example, the configuration may include a plurality of second line memories. In this case, the output of the first line memory is written to one of the second line memories, and the other second line memories are used. Are connected in series so as to write outputs of different second line memories. Then, the processing means may perform a scaling process such as an interpolation calculation based on the output of the first line memory and the outputs of the plurality of second line memories. In the scaling process, the scaling process can be performed by selectively referring to one or more outputs from the output of the first line memory or the outputs of the plurality of second line memories. .
[0008]
With respect to the main scanning direction, a pre-stage processing means for performing a scaling process in the main scanning direction based on one or a plurality of pixel data in the image data is provided, and the output of the pre-stage image processing means is input to the first line memory. It can be constituted as follows. At this time, the pre-processing means can perform the scaling process in the main scanning direction only at the time of reduction, so that the capacity of the first line memory and the second line memory can be minimized. . The enlargement process in the main scanning direction may be performed by a processing unit, or a means for performing enlargement processing in the main scanning direction may be provided at a later stage.
[0009]
Using such a scaling process function, for example, the scaling process may be performed on image data read by an original reading apparatus that reads an image by relatively moving the original reading unit and the original. it can. This makes it possible to input an enlarged image without reducing the relative movement speed when reading a document. Further, the magnification can be changed even if the moving speed of the document reading device is fixed, and the structure of the document reading device can be simplified.
[0010]
In addition, an image processing apparatus provided with an image recognition unit that receives image data after scaling and recognizes a predetermined image, and an image output control unit that controls output of the image data according to a recognition result by the image recognition unit are provided. It can also be configured as an image input device. In this case, the image recognizing unit depends on the deformation of the input image data and the like, but the image recognizing accuracy depends on the processing. Recognition accuracy in the recognition means can be improved.
[0011]
DETAILED DESCRIPTION OF THE INVENTION
FIG. 1 is a block diagram showing a first embodiment of the image processing apparatus of the present invention. In the figure, 1 is a first line memory, 2 is a second line memory, 3 is a first image processing unit, 4 is a control unit, 11 is input image data, 12 is a first clock signal, 13 is a second clock signal, Reference numeral 14 denotes output image data. The first line memory 1 writes the input image data 11 in accordance with the first clock signal 12 and reads out and outputs the image data in accordance with the second clock signal 13. The second line memory 2 writes image data read from the first line memory 1 according to the second clock signal 13 in accordance with the second clock signal 13, and reads out and outputs the image data in accordance with the second clock signal 13. is there.
[0012]
The frequency of the second clock signal 13 is determined based on the frequency of the first clock signal 12 and the maximum magnification. For example, when the maximum magnification is N times, the frequency of the second clock signal 13 can be N times that of the first clock signal 12. In the following description, it is assumed that N = 2 for simplicity.
[0013]
The first image processing unit 3 receives the image data output from the first line memory 1 and the second line memory 2 and refers to these image data alternately to perform a scaling process in the sub-scanning direction. At this time, a new scaling factor is generated based on the set scaling factor and the frequency ratio between the first clock signal 12 and the second clock signal 13, and scaling processing is performed according to the generated scaling factor. The scaling process in the sub-scanning direction can be performed by appropriately thinning out the image data output from the first line memory 1 and the image data for each line output from the second line memory 2 in units of lines. . The image data after the scaling process is output as output image data 14.
[0014]
The control unit 4 controls the timing of image data writing start and reading start with respect to the first line memory 1 and the second line memory 2, or is synchronized with the read timing of the first line memory 1 and the second line memory 2. The operation of the first image processing unit 3 is controlled. Although not shown, the first clock signal 12 and the second clock signal 13 may be used for control.
[0015]
Next, the operation of the image processing apparatus according to the first embodiment of the present invention will be described. FIG. 2 is a timing chart showing an example of the timing of writing to and reading from the first line memory 1 and the second line memory 2 in the first embodiment of the image processing apparatus of the present invention, and FIG. FIG. In the figure, a is the first clock signal 12, b is the second clock signal 13, c is the image data (that is, input image data 11) written to the first line memory 1, and d is the image read from the first line memory 1. Data, e is image data written to the second line memory 2 (same data as d), and f is image data read from the second line memory 2. In this example, the second clock signal 13 has a frequency twice that of the first clock signal 12, and the image data can be enlarged up to 200% in the sub-scanning direction. FIG. 3 shows an enlarged timing chart for one line with eight pixels per line.
[0016]
First, input image data 11 is input as indicated by c and written to the first line memory 1 in synchronization with the first clock signal 12 indicated by a. After the image data for one line is written, the image data is read from the first line memory 1 in synchronization with the second clock signal 13 indicated by b. For example, when the input image data 11 of line 0 is written in the first line memory 1, the already written image data of line 0 is read at the timing when the input image data 11 of the next line 1 is written. . As described above, since the second clock signal 13 has twice the frequency of the first clock signal 12, as shown in d, the first line memory 1 can read image data in half the time of writing. it can.
[0017]
The image data read from the first line memory 1 is input to the first image processing unit 3 and written to the second line memory 2 in synchronization with the second clock signal 13 (b) as indicated by e. It is. When the image data for one line is written, the image data is read from the second line memory 2 as shown by f. Reading of the image data from the second line memory 2 is performed in synchronization with the second clock signal 13 shown in b.
[0018]
In general, there is an invalid period of several clocks between lines of image data. Therefore, image data is written to and read from the second line memory 2 during the time required to write the input image data 11 to the first line memory 1 and the invalid period between lines. Thus, by using the image data (d) read from the first line memory 1 and the image data (f) read from the second line memory 2, a 200% enlarged image is always obtained in the sub-scanning direction. Will be obtained.
[0019]
The image data (d and f) read from the first line memory 1 and the second line memory 2 are sent to the first image processing unit 3. In the first image processing unit 3, the image data (d) read from the first line memory 1 and the image data (f) read from the second line memory 2 that are output with a time shift are alternately displayed. Refer to and perform scaling. When referring to the image data, the image data (d) and the image data (f) may be switched. Here, the image data (d) read from the first line memory 1 and the second line memory 2 are used. The logical sum of the image data (f) read out from is obtained, and a 200% enlarged image is obtained in the sub-scanning direction.
[0020]
Then, a reduction process in the sub-scanning direction is performed according to a preset scaling factor. At this time, an enlarged image up to 200% which is the maximum magnification is obtained, and therefore reduction processing is always performed. For example, suppose that it is desired to obtain 120% of image data in the sub-scanning direction as output image data 14 with respect to input image data 11. At this time, since the image data before the scaling process in the first image processing unit 3 is enlarged to 200% in the sub-scanning direction, the reduction process in the sub-scanning direction of 120% / 200% = 60% is performed. Just do it. Thus, by newly calculating the scaling factor and performing the reduction process, it is possible to perform enlargement or reduction in the sub-scanning direction at any scaling factor up to the maximum magnification (200% in this case).
[0021]
FIG. 4 is a timing chart showing an example at the time of scaling processing in the first image processing unit 3 in the first embodiment of the image processing apparatus of the present invention. In the figure, d and f are the same as those in FIGS. g is image data obtained by ORing d and f. Further, h is a scaling coefficient addition result. Here, the scaling factor is a value represented by the reciprocal of the scaling factor. In the following description, the newly calculated scaling factor is set to 60% (= 120% / 200%) as in the above example, and the scaling factor in this case is 1/60% = 1.667. i is the previous result of the scaling coefficient addition result. j is output image data.
[0022]
As described above, the first image processing unit 3 calculates the logical sum of the image data d output from the first line memory 1 and the image data f output from the second line memory 2 in this example, and the image data g.
[0023]
The scaling factor addition result h is initialized with an initial value of 0, and the scaling factor (in this example, 1.667) is added and updated after the output of the output image data 14 is completed for one line. The previous result i of the scaling coefficient addition result has an initial value of 0, and the data of the scaling coefficient addition result h is output synchronously after the end of one line of the logical sum image data g.
[0024]
As the scaling process, here, the difference between the integer part of the scaling coefficient addition result h and the previous result i of the scaling coefficient addition result is compared, and if the difference is 2 or more, the line at that timing is output as an invalid line. If the difference is less than 2, the line at that timing is output as an effective line. In this way, line thinning (reduction processing) is performed according to the newly calculated scaling factor, and scaling processing is realized in the sub-scanning direction. Of course, various other methods may be applied as the scaling process performed by the first image processing unit 3.
[0025]
For example, in FIG. 4, the first line 0 in the logical sum image data g is effective because the difference between the integer parts of the scaling coefficient addition result h and the previous result i of the scaling coefficient addition result is not 2 or more. This is a line and is output as output image data 14 as indicated by j. At this time, the scaling coefficient addition result h and the previous result i of the scaling coefficient addition result are both updated, the scaling coefficient addition result h is 1.667, and the previous result i of the scaling coefficient addition result is 0.0. .
[0026]
Next, the second line 0 in the logical sum image data g is not an effective line because the difference between the integer parts of the scaling coefficient addition result h and the previous result i of the scaling coefficient addition result is not 2 or more. Yes, and output as output image data 14 as indicated by j. At this time, the scaling coefficient addition result h and the previous result i of the scaling coefficient addition result are both updated, the scaling coefficient addition result h is 3.333, and the previous result i of the scaling coefficient addition result is 1.667. .
[0027]
The first line 1 is an invalid line because the integer part of the scaling coefficient addition result h and the previous result i of the scaling coefficient addition result is a difference of 2 or more, and is not output to the output image data 14. At this time, the scaling coefficient addition result h is not updated and remains at 3.333. Also, the previous result i of the scaling coefficient addition result is updated to 3.333.
[0028]
As the scaling process, here, the difference between the integer part of the scaling coefficient addition result h and the previous result i of the scaling coefficient addition result is compared, and if the difference is 2 or more, the line at that timing is output as an invalid line. If the difference is less than 2, the line at that timing is output as an effective line. In this way, line thinning (reduction processing) is performed according to the newly calculated scaling factor, and scaling processing is realized in the sub-scanning direction.
[0029]
In the above example, the case where the second clock signal 13 has a frequency twice as high as that of the first clock signal 12 has been described. However, the case where the second clock is a multiple of three times, four times or more times the first clock. 300% (3 times), 400% (4 times) or more is possible. For example, when enlargement by N times is possible, image data output from the first line memory 1, image data output by reading the second line memory 2 (N−1) times according to the second clock signal 13, and Thus, image data enlarged N times in the sub-scanning direction can be obtained. The image data enlarged N times in the sub-scanning direction may be reduced with a new scaling factor calculated according to the set scaling factor.
[0030]
In this way, it is possible to change at an arbitrary and fine scaling factor by a very simple circuit without providing a memory for storing image data for one page and without changing the clock frequency according to a set scaling factor. Double can be realized.
[0031]
FIG. 5 is a block diagram showing a second embodiment of the image processing apparatus of the present invention. In the figure, the same parts as those in FIG. Reference numeral 5 denotes a second line memory. This second embodiment shows an example in which a plurality of second line memories are connected in series to enable a scaling process based on a plurality of lines of image data.
[0032]
Similar to the second line memory 2, the second line memory 5 writes image data in accordance with the second clock signal 13, and reads out and outputs it in accordance with the second clock signal 13. 2 output. With this configuration, the second line memory 2 and the second line memory 5 are connected in series.
[0033]
The first image processing unit 3 receives the image data output from the first line memory 1, the second line memory 2, and the second line memory 5. Among these image data, one or a plurality of image data are selectively referred to, and a scaling process is performed by, for example, an interpolation calculation to generate and output output image data 14 having a predetermined scaling ratio. Of course, the scaling process may be performed by a method other than the interpolation calculation. The control unit 4 also controls the timing of writing and reading the second line memory 5.
[0034]
Next, the operation of the image processing apparatus according to the second embodiment of the present invention will be described. FIG. 6 is a timing chart showing an example of writing and reading timings to the first line memory 1 and the second line memories 2 and 5 in the second embodiment of the image processing apparatus of the present invention, and FIG. It is a partial enlarged view. In the figure, a to f are the same as those in FIGS. k is image data written to the second line memory 5, and m is image data read from the second line memory 5. In this example as well, the second clock signal 13 has a frequency twice that of the first clock signal 12, and the image data can be enlarged up to 200% in the sub-scanning direction. FIG. 7 is an enlarged timing chart for one line with eight pixels per line.
[0035]
First, input image data 11 is input as shown in c, and the input image data 11 is written into the first line memory 1 in synchronization with the first clock signal 12 shown in a. After the image data for one line is written, the image data d is read from the first line memory 1 in synchronization with the second clock signal 13 shown in b. For example, when the input image data 11 of line 0 is written to the first line memory 1, the image data of line 0 already written is written to the first line memory 1 at the timing when the input image data 11 of the next line 1 is written. Will be read from. As described above, since the second clock signal 13 has twice the frequency of the first clock signal 12, as shown in d, the first line memory 1 can read image data in half the time of writing. it can.
[0036]
The image data d read from the first line memory 1 is written to the second line memory 2 as shown by e in synchronization with the second clock signal 13 shown by b. When image data for one line is written to the second line memory 2, the image data f is read from the second line memory 2 in synchronization with the second clock b. The image data f read from the second line memory 2 is written to the second line memory 5 as indicated by k in synchronization with the second clock signal 13 indicated by b.
[0037]
Next, image data is read from the second line memory 5. As described above, reading of image data from the second line memory 5 is performed in synchronization with the second clock, and the second clock has a frequency twice that of the first clock. Accordingly, as indicated by m, the same image data for one line can be read twice from the second line memory 5 during the time required to write one line of image data to the first line memory 1. Thus, a 200% enlarged image is always obtained from the second line memory 5 in the sub-scanning direction. In addition, image data one line before the first line memory 1 and the second line memory 2 can be output.
[0038]
In this manner, image data enlarged by 200% in the sub-scanning direction is obtained from the first line memory 1 and the second line memory 2 in the same manner as in the first embodiment described above, and the image data For the image data one line before, image data obtained by enlarging 200% in the sub-scanning direction is obtained from the second line memory 5.
[0039]
The image data output from the first line memory 1 and the second line memories 2 and 5 are input to the first image processing unit 3. The first image processing unit 3 uses the image data output from the second line memory 2 and either the image data output from the first line memory 1 or the image data output from the second line memory 2. A scaling process is performed by interpolation or the like. The image data m supplied from the second line memory 5 is supplied with image data of the same line first, second and twice, but the first image data is used for the first line memory 1 for interpolation calculation. The second image data is used together with the image data f from the second line memory 2 for interpolation calculation. In FIG. 6, the image data used for the interpolation calculation is hatched.
[0040]
The first image processing unit 3 performs a reduction process in the sub-scanning direction at a preset scaling factor. For example, assume that 120% of image data is obtained as output image data 14 in the sub-scanning direction with respect to the input image data 11 shown in c. At this time, since the image data m input to the first image processing unit 3 is an image enlarged to 200% in the sub-scanning direction, a reduction process in the sub-scanning direction of 120% / 200% = 60% is performed. Just do it. Thus, by newly calculating the scaling factor and performing the reduction process, it is possible to perform enlargement or reduction in the sub-scanning direction at any scaling factor up to the maximum magnification (200% in this case).
[0041]
FIG. 8 is a timing chart showing an example of the scaling process in the first image processing unit 3 in the second embodiment of the image processing apparatus of the present invention. D, f, and m in the figure are the same as those in FIGS. 6 and 7, and h, i, and j are the same as those in FIG. In this example as well, the newly calculated scaling factor is 60% (= 120% / 200%). In this case, the magnification coefficient is 1/60% = 1.667.
[0042]
Also in this example, as the scaling process performed in the first image processing unit 3, the difference between the integer part of the scaling coefficient addition result h and the previous result i of the scaling coefficient addition result is compared, and the difference is 2 or more. If it is, the line at that timing is not output as an invalid line, and if the difference is less than 2, image data for one line is generated as an effective line by interpolation processing and output as output image data 14. The pixel density interpolation calculation at this time is calculated using the image data d read from the first line memory 1, the image data m read from the second line memory 5, and the fractional part of the scaling coefficient addition result h. To do. For example,
(1-h decimal part) × m pixel density + h decimal part × d pixel density
Can be calculated with Alternatively, using the image data f read from the second line memory 2, the image data m read from the second line memory 5, and the fractional part of the scaling coefficient addition result h, for example,
(1-h decimal part) × m pixel density + h decimal part × f pixel density
Can be calculated with Any one of these may be used to calculate the density of the output image data 14. Of course, various other methods may be applied as the scaling process performed by the first image processing unit 3.
[0043]
The scaling coefficient addition result h has an initial value of 0, and the scaling coefficient is added and updated synchronously after the end of one line output in the first image processing unit 3. The previous result i of the scaling coefficient addition result has an initial value of 0, and the data of the scaling coefficient addition result h is output after the end of one line of the image data m read from the second line memory 5. In the first line 0 of the image data m read from the second line memory 5, the difference between the integer parts of the scaling coefficient addition result h and the previous result i of the scaling coefficient addition result is not 2 or more. Therefore, it is an effective line, and output image data 14 is generated and output as indicated by j. Since the scaling coefficient addition result h at this time is 0.0, the pixel density of the output image data 14 is (1-0) × m pixel density + 0 × d pixel density = m pixel density. That is, the image data m read from the second line memory 5 is output. After the output of the output image data 14 for one line, the scaling coefficient addition result h and the previous result i of the scaling coefficient addition result are both updated, the scaling coefficient addition result h is 1.667, and the scaling coefficient addition result The previous result i is 0.0.
[0044]
Next, in the second line 0 of the image data m read from the second line memory 5, the difference between the integer parts of the scaling coefficient addition result h and the previous result i of the scaling coefficient addition result is 2 <2>. Since it is not above, it is an effective line, and the output image data 14 is output as indicated by j. The pixel density of the output image data 14 is obtained by the pixel density of (1−0.667) × m + 0.667 × f because the scaling factor addition result h is 1.667. After the output of the output image data 14 for one line, the scaling coefficient addition result h and the previous result i of the scaling coefficient addition result are both updated, and the scaling coefficient addition result h is 3.333, The previous result i is 1.667.
[0045]
In the first line 1 of the image data m read from the second line memory 5, the difference between the integer parts of the scaling coefficient addition result h and the previous result i of the scaling coefficient addition result is 2 or more. Therefore, it is an invalid line, and the output image data 14 is not output as indicated by j. At this time, the scaling coefficient addition result h is not updated and remains at 3.333, and only the previous result i of the scaling coefficient addition result is updated to 3.333.
[0046]
In this way, by sequentially adding scaling factors and comparing the integer part with the previous result, a line to be thinned out in the sub-scanning direction is selected, and for the line outputting output image data, the scaling factor addition result is Using the decimal part and reference data for pixel density interpolation calculation, reduction processing is performed while performing pixel density interpolation calculation. In this way, image data enlarged to 200% in the sub-scanning direction is used, and reduction processing using interpolation processing is performed from image data of a plurality of lines to obtain image data with a predetermined scaling factor. Scaling can be performed at any scaling factor up to 200% in the sub-scanning direction.
[0047]
In this embodiment, the second clock signal 13 has a frequency twice that of the first clock signal 12, and an example in which the magnification in the sub-scanning direction can be up to 200% is shown. However, the present invention is not limited to this, and enlargement of 300%, 400%, or more can be performed. In that case, the second clock signal 13 is set to be a multiple of three, four or more times the first clock signal 12, and the number of second line memories is further connected in series by one, two or more. Can be realized.
[0048]
In the timing charts shown in FIGS. 6 and 7, when reading image data from the first line memory 1 and writing the image data into the second line memory 2, Writing to the two-line memory 5 is not performed. However, at this timing, there is no problem even if reading from the second line memory 2 and writing to the second line memory 5 are performed. In this case, since the image data is output from the second line memory 2 together with the first line memory 1 and the second line memory 5, the first image processing unit 3 selects and references these outputs. do it. In this case, it is also possible to perform an interpolation operation from the output of the first line memory 1 and the output of the second line memory 2.
[0049]
FIG. 9 is a block diagram showing a third embodiment of the image processing apparatus of the present invention. In the figure, the same parts as those in FIG. Reference numeral 6 denotes a second image processing unit. In the third embodiment, the second image processing unit 6 is provided in the preceding stage of the first line memory 1 of the first embodiment described above, and the scaling process is also performed in the main scanning direction. Is shown. In the third embodiment, regarding the scaling in the sub-scanning direction, the same processing as in the first embodiment described above is performed, and thus the description thereof is omitted here.
[0050]
The second image processing unit 6 performs a scaling process in the main scanning direction based on one or a plurality of pixel data in the input image data 11 and inputs a scaling result in the main scanning direction to the first line memory 1. To do. At this time, the second image processing unit 6 performs only the reduction process in the main scanning direction. The first image processing unit 3 performs enlargement processing in the main scanning direction in addition to scaling in the sub-scanning direction. Alternatively, a third image processing unit that performs enlargement processing in the main scanning direction may be provided after the first image processing unit 3.
[0051]
That is, for scaling in the main scanning direction, the reduction process is performed in the second image processing unit 6, and the enlargement process is performed in the first image processing unit 3. By performing the enlargement process in the subsequent stage in this way, it is possible to suppress the increase in the capacity of the first line memory 1 and the second line memory 2 and perform the scaling process with a small capacity.
[0052]
FIG. 10 is an explanatory diagram showing an example of the data width of the image data between the respective units when the scaling process is performed in the main scanning direction in the third embodiment of the image processing apparatus of the present invention. In FIG. 10, the data width is represented by the number of rectangles. FIG. 10A shows the time of reduction processing in the main scanning direction. In the case of performing reduction in the main scanning direction, when the input image data 11 is input, the second image processing unit 6 performs reduction processing in the main scanning direction, and as shown in FIG. The output of the second image processing unit 6 decreases the data width of one line. Next, the first line memory 1, the second line memory 2, and the first image processing unit 3 perform a scaling process in the sub-scanning direction, and output image data 14 is output. In this case, the first image processing unit 3 does not perform scaling processing in the main scanning direction.
[0053]
FIG. 10B shows the time of enlargement processing in the main scanning direction. When enlarging in the main scanning direction, the second image processing unit 6 does not perform processing on the input image data 11. Therefore, as shown in FIG. 10B, the data width of one line of the image data output from the second image processing unit 6 does not change. Next, the first line memory 1, the second line memory 2, and the first image processing unit 3 perform a scaling process in the sub-scanning direction, and the first image processing unit 3 performs an enlargement process in the main scanning direction. Accordingly, as shown in FIG. 10B, the output image data 14 has a wide data width of one line.
[0054]
With this configuration, the amount of data written to the first line memory 1 and the second line memory 2 can be a maximum amount of data corresponding to the line width at the same magnification processing. The first line memory 1 and the second line memory 2 Cost reduction. At this time, enlargement in the main scanning direction can replenish pixels only up to a data amount corresponding to an invalid period between lines. Therefore, the frequency of the second clock signal 12 is further increased in consideration of the maximum magnification in the sub-scanning direction and the magnification in the main scanning direction. As a result, it is possible to increase the invalid period between each line after scaling in the sub-scanning direction and increase the maximum scaling ratio in the main scanning direction.
[0055]
FIG. 11 is a timing chart showing an example of writing and reading timings to the first line memory 1 and the second line memory 2 in the third embodiment of the image processing apparatus of the present invention. In the figure, a to f are the same as those in FIG. FIG. 11A is a timing chart when the frequency of the second clock signal 13 is twice the frequency of the first clock signal 12, and FIG. 11B shows the frequency of the second clock signal 13 being the first frequency. 3 is a timing chart when the frequency of one clock signal 12 is three times. 11A and 11B, the maximum magnification in the sub-scanning direction is 200%, and the first line memory 1 and the second line memory are written while image data is being written to the first line memory 1. Image data for 2 to 2 lines is read.
[0056]
As shown in FIG. 11A, when the frequency of the second clock signal 13 is twice the frequency of the first clock signal 12, the image data d read from the first line memory 1 and the second line The period between the image data f read from the memory 2, that is, the period shown as the invalid period in FIG. 11A is short. For this reason, there is only a short period in which pixel data can be increased in the main scanning direction, and the enlargement rate in the main scanning direction is extremely limited.
[0057]
As shown in FIG. 11B, when the frequency of the second clock signal 13 is three times the frequency of the first clock signal 12, the image data d read from the first line memory 1 and the second line The invalid period between the image data f read from the memory 2 is longer than that shown in FIG. Since a large amount of pixel data can be output during this long invalid period, it is possible to cope with the case where the number of pixels is increased by enlarging with a large scaling factor in the main scanning direction, and the maximum scaling factor in the main scanning direction is set. It becomes possible to give.
[0058]
In FIG. 11, the case where the frequency of the second clock signal 13 is twice and three times that of the first clock signal 12 has been described. However, the frequency of the second clock signal 13 is four times, five times that of the first clock signal 12 or higher. With the above, it is possible to further increase the maximum magnification in the main scanning direction. Thus, the frequency of the second clock signal 13 may be determined in consideration of the maximum magnification in the main scanning direction as well as the maximum magnification in the sub-scanning direction and the frequency of the first clock signal 12.
[0059]
In the third embodiment, the description has been made based on the configuration based on the above-described first embodiment. Of course, the configuration based on the above-described second embodiment is also possible. is there.
[0060]
Hereinafter, some application examples of the image processing apparatus of the present invention will be described. FIG. 12 is a block diagram showing a first application example of the image processing apparatus of the present invention, and FIG. 13 is a block diagram showing an example of a scanning apparatus. In the figure, the same parts as those in FIG. Reference numeral 21 denotes a scanning device, 22 denotes an image processing device, 31 denotes a platen, 32 denotes a document, and 33 denotes a document reading unit. In the first application example, an image read by the scanning device 21 is input as input image data 11 to the image processing device 22 of the present invention, and a scaling process is performed.
[0061]
The scanning device 21 reads an image on a document and outputs image data for each line. Although there are several types of configurations of the scanning device 21, FIG. 13 shows a flatbed scanner. The document 32 is placed on a platen 31 fitted with, for example, transparent glass. Then, an image on the document 32 is optically read by the document reading unit 33 facing the transparent glass and outputted as image data. When reading an image on the document 32, the image is read in a direction substantially orthogonal to the arrow while moving the document reading unit 33 in the direction indicated by the arrow in FIG.
[0062]
In such a configuration, the scanning device 21 may move the document reading unit 33 at a constant speed even when there is a magnification change instruction in the sub-scanning direction. The magnification change processing is performed in the image processing device 22 of the present invention. Just do it. In particular, during the enlargement process in the sub-scanning direction, it is not necessary to slow down the moving speed of the document reading unit 33, and the reading process can be speeded up. In addition, since the moving speed of the document reading unit 33 is constant, the control mechanism of the document reading unit 33 can be simplified.
[0063]
FIG. 12 shows the configuration of the image processing apparatus 22 according to the present invention based on the above-described third embodiment. For example, based on the first embodiment and the second embodiment. Other configurations are possible.
[0064]
FIG. 14 is a block diagram showing a second application example of the image processing apparatus of the present invention. In the figure, 41 is a scanning device, 42 is an image processing device, 43 is an image recognition unit, 44 is an image output control unit, 45 is read image data, 46 is a recognition result, and 47 is output image data. In this second application example, an example in which the image processing apparatus of the present invention is incorporated in an image input apparatus having an image recognition unit is shown. Such a configuration can be applied, for example, to a part that performs image input and image processing of a digital color copying machine.
[0065]
The scanning device 41 has, for example, the configuration shown in FIG. 13 and outputs read image data 45. In this configuration, scaling in the sub-scanning direction is performed by changing the speed at which the image reading unit 33 in the scanning device 41 is moved. In the main scanning direction, image data with the highest resolution of the image reading unit 33 is output.
[0066]
The image output control unit 44 performs a scaling process in the main scanning direction based on the read image data 45 output from the scanning device 41 and outputs the result as output image data 47. At this time, output control is performed as described later according to the recognition result 46 from the image recognition unit 43.
[0067]
The image processing apparatus 42 is the above-described image processing apparatus of the present invention, and performs the scaling process in the sub-scanning direction so that the scaling process in the sub-scanning direction performed by the scanning apparatus 41 is restored. As the image processing device 42, any configuration such as the configuration shown in the above-described first to third embodiments or a modified example thereof can be applied.
[0068]
The image recognition unit 43 recognizes whether or not the copy image is prohibited from being included in the read image data 45 and outputs a recognition result 46. Recently, with the digitization, colorization, and higher resolution of copying machines, illegal acts of copying banknotes or securities are becoming more serious. As a countermeasure against such an action, the image recognition unit 43 determines whether or not a document to be copied is a target that should not be copied such as banknotes or securities. When the image recognition unit 43 recognizes the document as a bill or a securities, the image output control unit 44 performs a process such as not outputting an image.
[0069]
In such a configuration, in order for the image recognizing unit 43 to accurately perform the image recognizing process, it is desirable that an equal-size image is input to the image recognizing unit 43. However, when zooming is designated, zooming processing in the sub-scanning direction is performed in the scanning device 41 as described above. Therefore, the read image data 45 output from the scanning device 41 may be an image extending in the sub-scanning direction with respect to the original image or a collapsed image. When image recognition processing is performed based on such read image data 45, a case where banknotes or securities cannot be recognized occurs.
[0070]
However, in the configuration shown in FIG. 14, an image processing device 42 for returning to an equal-magnification image in the sub-scanning direction is provided before the image recognition unit 43. Therefore, even if the scanning device 41 performs scaling in the sub-scanning direction, the image recognition unit 43 can perform image recognition processing using the same-size image, and can successfully copy banknotes and securities. Images that should not be recognized can be recognized.
[0071]
In order to generate the output image data 47, the recognition result 46 must be input from the image recognition unit 43 to the image output control unit 44. Therefore, high-speed processing is required in the image processing device 42 and the image recognition unit 43 in order to output the output image data 47 quickly. As described above, in the image processing apparatus 42 of the present invention, the image scaling process can be performed at high speed, so the scaling process in the image processing apparatus 42 does not hinder the speeding up. Furthermore, since the image processing apparatus 42 of the present invention has a simple configuration, the cost of a copying machine can be suppressed.
[0072]
Although two application examples are shown here, of course, the image processing apparatus of the present invention can be applied to various other purposes.
[0073]
【The invention's effect】
As is apparent from the above description, according to the present invention, a page memory is not required as in the prior art, and a complicated configuration such as changing the frequency in accordance with the variable magnification is not required, and it is simple in real time. With such a configuration, there is an effect that it is possible to provide an image processing apparatus that can realize zooming with an arbitrary and fine scaling factor. Further, by applying such an image processing device to an image input device, for example, the configuration of the image reading device can be simplified and the reading speed can be increased, and the image recognition unit can provide an image in real time. There is also an effect that the recognition accuracy can be improved by normalization.
[Brief description of the drawings]
FIG. 1 is a block configuration diagram showing a first embodiment of an image processing apparatus of the present invention.
FIG. 2 is a timing chart showing an example of writing and reading timings to the first line memory 1 and the second line memory 2 in the first embodiment of the image processing apparatus of the present invention.
FIG. 3 is a partially enlarged view of an example of timing of writing to and reading from the first line memory 1 and the second line memory 2 in the first embodiment of the image processing apparatus of the present invention;
FIG. 4 is a timing chart illustrating an example of a scaling process in the first image processing unit 3 in the first embodiment of the image processing apparatus of the present invention.
FIG. 5 is a block diagram showing a second embodiment of the image processing apparatus of the present invention.
FIG. 6 is a timing chart showing an example of writing and reading timings to the first line memory 1 and the second line memories 2 and 5 in the second embodiment of the image processing apparatus of the present invention.
FIG. 7 is a partially enlarged view of an example of timing of writing to and reading from the first line memory 1 and the second line memories 2 and 5 in the second embodiment of the image processing apparatus of the present invention;
FIG. 8 is a timing chart showing an example of a scaling process in the first image processing unit 3 in the second embodiment of the image processing apparatus of the present invention.
FIG. 9 is a block configuration diagram showing a third embodiment of the image processing apparatus of the present invention.
FIG. 10 is an explanatory diagram illustrating an example of the data width of image data between units when performing scaling processing in the main scanning direction in the third embodiment of the image processing apparatus of the present invention;
FIG. 11 is a timing chart showing an example of writing and reading timings to the first line memory 1 and the second line memory 2 in the third embodiment of the image processing apparatus of the present invention.
FIG. 12 is a block diagram showing a first application example of the image processing apparatus of the present invention.
FIG. 13 is a configuration diagram illustrating an example of a scanning device.
FIG. 14 is a block diagram illustrating a second application example of the image processing apparatus of the present invention.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 1 ... 1st line memory, 2, 5 ... 2nd line memory, 3 ... 1st image processing part, 4 ... Control part, 6 ... 2nd image processing part, 11 ... Input image data, 12 ... 1st clock signal, 13 ... second clock signal, 14 ... output image data, 21 ... scanning device, 22 ... image processing device, 31 ... platen, 32 ... original, 33 ... original reading unit, 41 ... scanning device, 42 ... image processing device, 43 Image recognition unit 44 Image output control unit 45 Read image data 46 Recognition result 47 Image data output

Claims (12)

  In an image processing apparatus that performs a scaling process of image data, the image data is written with a first clock and at least N times the first clock determined based on a frequency and a maximum magnification of the first clock. A first line memory that reads and outputs with a second clock of a frequency, and a second line memory that writes and outputs the output of the first line memory with the second clock and outputs with the second clock Then, based on one output from the first line memory and (N-1) outputs from the second line memory, the change in the sub-scanning direction by an arbitrary variable magnification up to the maximum magnification. An image processing apparatus comprising processing means for performing double processing.   The image processing apparatus according to claim 1, wherein the processing unit performs the scaling process by alternately referring to an output of the first line memory and an output of the second line memory.   A plurality of the second line memories are provided, and the output of the first line memory is written in one of them, and the output of another second line memory is written in each of the other second line memories. 2. The power supply circuit according to claim 1, wherein the processing means performs the scaling process based on an output of the first line memory and outputs of the plurality of second line memories. The image processing apparatus described.   The processing means performs the scaling process by selectively referring to one or a plurality of outputs from an output of the first line memory or a plurality of outputs of the second line memory. The image processing apparatus according to claim 3.   5. The image processing apparatus according to claim 1, wherein the frequency of the second clock is an integer multiple of the frequency of the first clock. 6.   The processing means generates a new scaling factor based on a set scaling factor and a frequency ratio of the first clock and the second clock, and performs the scaling process according to the scaling factor. The image processing apparatus according to claim 1, wherein:   And a pre-stage processing unit that performs a scaling process in the main scanning direction based on one or a plurality of pixel data in the image data, and an output of the pre-stage image processing unit is input to the first line memory. The image processing apparatus according to any one of claims 1 to 6.   The image processing apparatus according to claim 7, wherein the pre-stage image processing unit performs a scaling process in the main scanning direction only at the time of reduction.   The image processing apparatus according to claim 7, wherein the frequency of the second clock is determined in consideration of a scaling factor in a main scanning direction.   10. The image according to claim 1, wherein the image data is read by a document reading device that reads an image by relatively moving the document reading unit and the document. 11. Processing equipment.   Further, the image recognition means includes image recognition means for receiving the image data after scaling and recognizing a predetermined image, and the processing means performs scaling processing so that an optimum image is obtained in the image recognition means. The image processing apparatus according to claim 1.   An image reading unit that reads an image, an image processing device according to any one of claims 1 to 9 that performs a scaling process on image data read by the image reading unit, and the image processing device. Image recognition means for recognizing a predetermined image from the image data subjected to the scaling process, and image output control means for controlling the output of the image data read by the image reading means in accordance with the recognition result by the image recognition means. A characteristic image input device.

Images