JP2007156585A - Image processing program and image processor - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image processing program and an image processor for executing a pixel averaging method for acquiring a high image quality reduction image at a higher speed processing speed than a conventional manner. <P>SOLUTION: In this image processing program and an image processor, a mean pixel value in one pixel of a reduction image is calculated by using divided blocks generated by dividing an original image into nDev pieces as units, so that it is possible to acquire mean pixel values by an integer arithmetic operation without calculating decimal numbers to be generated according to a reduction rate by a floating-point arithmetic operation. As a result, it is possible to more sharply improve the processing speed of a pixel averaging method than a conventional manner, and to obtain a high image quality reduction image at a high speed. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to an image processing program and an image processing apparatus for reducing an image, and more particularly to an image processing program and an image processing apparatus having both high image quality and high processing speed.

  Conventionally, as an algorithm for obtaining an output image having a size different from that of the input image, a near-less traverser method (nearest neighbor method), a bilinear method (linear interpolation method), a bicubic method (third-order interpolation method), an average pixel method, etc. There are algorithms.

  Here, the average pixel method is a reduction-dedicated algorithm that performs reduction by averaging pixel values in consideration of the area occupied by the original image in the reduced image (output image). In many cases, a reduced image obtained using this average pixel method has higher image quality than reduced images obtained using other algorithms.

  One method for performing the average pixel method is a method in which the image size of the original image is temporarily enlarged to the size of the least common multiple of the image size of the reduced image and then the average value of the pixel values is obtained (for example, JP, 2003-256827, A (patent documents 1)).

  As another method of the average pixel method, there is a method of obtaining an average value of pixel values for an original image portion corresponding to a reduced image without enlarging the image size of the original image. FIG. 12 is a schematic diagram for explaining a method for obtaining the average value of the pixel values for the original image portion corresponding to the reduced image as described above. Note that FIG. 12 illustrates a case where only the main scanning direction (x direction) is reduced without considering reduction in the sub scanning direction (y direction) for easy understanding.

  FIG. 12 shows an original image P ′ that is an image composed of five pixels P′0 to P′4 arranged in the main scanning direction, and Q ′ obtained by reducing the original image P ′. A reduced image Q ′, which is an image composed of three pixels 0 to Q′2, is illustrated. As in the example shown in FIG. 12, the number of pixels of the original image P ′ corresponding to one pixel of the reduced image Q ′ may not be an integer. For example, as shown in FIG. 12, the pixel Q′0 in the reduced image Q ′ corresponds to the pixel P′0 and a part of the pixel P′1 in the original image P ′.

  When a method of obtaining an average value of pixel values for the original image portion corresponding to the reduced image without enlarging the image size of the original image, reduction processing by the average pixel method is performed by the processing shown in FIG. FIG. 13 is a flowchart showing conventional average pixel processing executed by the central processing unit when a method of obtaining an average value of pixel values for an original image portion corresponding to a reduced image is used.

  As shown in FIG. 13, this conventional average pixel processing is performed by first, the image size (ix ′ × iy ′) of the original image P ′ and the image size (ox ′ × oy) of the reduced image Q ′ designated by the user. '), The reduction rate is calculated (S1301), and the range of the original image P ′ corresponding to one pixel of the reduced image Q ′ is determined according to the calculated reduction rate (S1302).

Then, the coordinate j _'out' of the main scanning direction of the pixel at the coordinate i in the (x-direction) 'reduced image Q _out and the sub-scanning direction (y-direction) is set to a beginning coordinate (0,0) (S1303 ), The value of the variable Plane is set to 0 (S1304). This variable Plane is a variable for designating a color plane. Here, when the image data is composed of RGB, the value of the variable Plene can take three values from 0 to 2. According to the value of this variable Plane, one of the color planes of red (R) plane (Plane = 0), green (G) plane (Plane = 1), and blue (B) plane (Plane = 2) is designated. Can do. In the example shown in FIG. 13, the image data is assumed to be composed of RGB.

After the processing of S1304, the average value (average pixel) of the pixel values of the color plane specified by the Plane value in the area of the original image P ′ corresponding to the pixel at the coordinates (i ′ _out , j ′ _out ) in the reduced image Q ′ Value) is calculated (S1305), and the calculated average pixel value is output to the buffer (S1306).

  After the processing of S1306, it is confirmed whether the value of the variable Plane is 2 (S1307). If the value of the variable Plane is 0 or 1 (S1307: No), 1 is added to the value of the variable Plane (S1310), The process proceeds to S1305. That is, the processing of S1305 to S1306 is executed for the color plane corresponding to the next plane value.

On the other hand, if the value of the variable Plane is 2 as a result of the confirmation in S1307 (S1307: Yes), the pixel values of the pixels at the coordinates (i ′ _out , j ′ _out ) in the reduced image Q ′ are all Since it is obtained for the color plane (RGB), next, i ′ _out = ox ′, that is, the pixel at the coordinates (i ′ _out , j ′ _out ) in the output image Q is main scanned. It is confirmed whether the pixel is at the end of the direction (S1308).

If i ′ _out = ox ′ is not the result of the confirmation in S1308 (S1308: No), 1 is added to the value of i ′ _out , and the pixel of the reduced image Q ′ is advanced by 1 in the main scanning direction (S1311). , The process proceeds to S1304, and the processes of S1304 to S1307 and S1310 are executed for the pixels of the output image Q ′ advanced by 1 in the main scanning direction in S1311.

On the other hand, if i ′ _out = ox ′ as a result of confirmation in the processing of S1308 (S1308: Yes), the average pixel value is calculated for all the pixels located at the coordinate j ′ _out in the same sub-scanning direction. Next, j ′ _out = oy ′, that is, the pixel at the coordinates (i ′ _out , j ′ _out ) in the reduced image Q ′ is the end pixel in the sub-scanning direction. (S1309).

Judgment of the step S1309, else _{j 'out = oy' (S1309} : No), ' as well as 1 is added to the value of _out (S1312), i' j the value of _out is set to 0 (S1313) . As a result of the processing in S1312 and S1313, the position of the pixel selected from the pixels constituting the reduced image Q ′ is advanced by 1 in the sub-scanning direction and returned to the top coordinate in the main scanning direction. After the processing of S1313, the process proceeds to S1304, and the processing of S1304 to S1307 and S1310 is executed for the pixels of the output image Q ′ advanced by 1 in the sub-scanning direction in S1313.

On the other hand, if j ′ _out = oy is confirmed as a result of the processing in S1309 (S1309: Yes), the average pixel value of all ((ox ′ × oy ′)) pixels in the reduced image Q ′ is determined. Since this indicates that the calculation has been performed, this conventional average pixel processing is terminated. As an execution result of this conventional average pixel processing, a reduced image Q ′ having an output image size (ox ′ × oy ′) reduced by the average pixel method from the original image P ′ having the original image size (ix ′ × iy ′) is obtained. can get.

JP 2003-256827 A

  However, while the average pixel method has higher image quality than other algorithms, the processing speed is generally slow. Particularly, when the average pixel method is executed by the above method using the least common multiple, the reduction ratio is small. Depending on the value, there is a problem that the least common multiple (image size at which the original image is enlarged) becomes too large, and processing may become impossible.

  Further, when the average pixel method is executed by a method of obtaining an average value of pixel values for the original image portion corresponding to the reduced image without enlarging the image size of the original image (processing corresponding to the above-described flowchart of FIG. 13). ), The processing speed is improved compared to the case of enlarging the size of the original image, but it is slower than the processing speed at the time of reduction using other algorithms such as the near-less tray method and the bilinear method, and further processing is performed. An improvement in speed is desired.

  The reason is considered to be that floating point arithmetic is performed in S1305 in FIG. Specifically, using the example shown in FIG. 12, the pixel P′1 corresponding to the image Q′0 is divided into (1-t) by dividing the entire pixel by a ratio of (1-t): t. If it is a corresponding part, in S1305 in FIG. 13, an operation of {1 × P′0 + (1−t) × P′1} / {1+ (1−t)} is performed, and the pixel Q′0 Pixel values (average pixel values) are calculated. Here, since (1-t) is a decimal number, floating-point arithmetic is performed, and as a result, it is considered that there is a limit to the speed improvement of arithmetic processing.

  The present invention has been made to solve the above-described problems, and an image processing program and an image processing capable of executing an average pixel method capable of obtaining a high-quality reduced image at a higher processing speed than conventional methods. The object is to provide a device.

  In order to achieve this object, the image processing program according to claim 1 reduces the substantially rectangular original image having the number of pixels (ix × iy) at a predetermined reduction ratio and includes the number of pixels (ox × oy). A program executed by an arithmetic unit to obtain a substantially rectangular output image, wherein a reduction rate acquisition step for acquiring the reduction rate, and each pixel constituting the original image is (nDev × nDev) (nDev is 2 or more) A division number acquisition step of acquiring a division number nDev for dividing into a plurality of division blocks, and, for each pixel of the output image, an integer number of the division blocks corresponding to one pixel of the output image Is calculated by the processing block number calculating step for each pixel of the output image, and calculating the number of processing blocks based on the reduction ratio acquired by the reduction rate acquiring step. And a mean pixel value calculation step of calculating an average pixel value of the region of the original image corresponding to the number of divided blocks.

  The image processing program according to claim 2 is the image processing program according to claim 1, wherein in the processing block number calculation step, the number of divided blocks in the main scanning direction or the sub scanning direction based on the reduction ratio is expressed by a decimal number. In this case, it is treated as an integer number of the divided blocks by rounding down the decimal part.

  The image processing program according to claim 3 is the image processing program according to claim 1 or 2, wherein the processing block number calculation step calculates the number of the divided blocks corresponding to one pixel in the output image. In addition, the number of the divided blocks corresponding to one pixel of the output image is calculated from the divided block adjacent to the divided block already used in the average pixel value calculating step.

  The image processing program according to claim 4 is the image processing program according to any one of claims 1 to 3, wherein each pixel of the output image has its main scanning direction before the execution of the average pixel value calculating step. (1a) an area consisting of x1 (x1 is an integer less than or equal to 0 and less than nDev) corresponding to the unprocessed portion in the pixel that leaves the unprocessed portion in the original image; and (1b) the In the original image, an area adjacent to the area (1a) and in which N complete pixels are arranged (N is an integer of 1 or more), and (1c) in the original image, adjacent to the area (1b), 1 A main scanning direction dividing step of dividing into x2 (x2 is an integer less than or equal to 0 and less than nDev) remaining portion that is less than a pixel, and the average pixel value calculating step includes the main scanning direction For each region divided by the minute step, a sum of pixel values is obtained, and the sum of the pixel values is divided by the number of divided blocks corresponding to one pixel of the output image obtained by the processing block number calculating step. Ask by.

  An image processing program according to a fifth aspect is the image processing program according to the fourth aspect, further comprising a main scanning direction pixel position calculating step of calculating a pixel position in the main scanning direction of the pixel including the region (1a).

The image processing program according to claim 6 is the image processing program according to any one of claims 1 to 5, wherein each pixel of the output image is subjected to a sub-scanning direction before the execution of the average pixel value calculating step. (2a) an area composed of y1 (y1 is an integer greater than or equal to 0 and less than nDev) corresponding to the unprocessed portion in a pixel that leaves an unprocessed portion in the original image, and (2b) the In the original image, adjacent to the region (2a) and arranged with M complete pixels (M is an integer of 1 or more), and (2c) in the original image, adjacent to the region (2b), 1 A sub-scanning direction partitioning step for partitioning into a region composed of remaining portions of y2 (y2 is an integer less than or equal to 0 and less than nDev) less than a pixel;
The average pixel value calculating step is performed for each area divided by the sub-scanning direction dividing step, or for each area generated as a result of the division by the sub-scanning direction dividing step and the main scanning direction dividing step. The sum of the values is obtained, and the sum of the pixel values is obtained by dividing the sum of the pixel values by the number of the divided blocks corresponding to one pixel of the output image obtained in the processing block number calculating step.

  An image processing program according to a seventh aspect is the image processing program according to the sixth aspect, further comprising a sub-scanning direction pixel position calculating step of calculating a pixel position in the sub-scanning direction of the pixel including the region (2a).

  An image processing program according to claim 8 is the image processing program according to any one of claims 1 to 7, wherein an input image size acquisition step of acquiring a value of the number of input pixels (ix × iy); An output image size input step for inputting a value of the number of pixels of the output image (ox × oy), wherein the reduction rate acquisition step includes the number of pixels of the input image acquired by the input image size acquisition step (ix × The reduction ratio is acquired from the value of iy) and the value of the number of pixels (ox × oy) of the output image input in the output image size input step.

  An image processing program according to a ninth aspect is the image processing program according to any one of the first to eighth aspects, wherein the output image having the number of pixels (ix × iy) to the number of pixels (ox × oy) is changed. A reduction rate input step for inputting a reduction rate to obtain, wherein the reduction rate acquisition step acquires the reduction rate input in the reduction rate input step;

  The image processing apparatus according to claim 10, wherein an input unit that inputs a substantially rectangular original image having the number of pixels (ix × iy), and the original image input from the input unit is reduced at a predetermined reduction rate, and pixels A reduction unit that obtains a substantially rectangular output image having a number (ox × oy), wherein the reduction unit includes a reduction rate acquisition unit that acquires the reduction rate; and each pixel that forms the original image Division number acquisition means for acquiring a division number nDev for dividing (nDev × nDev) (nDev is an integer greater than or equal to 2) division blocks, and one pixel of the output image for each pixel of the output image Processing block number calculation means for calculating the number of integer blocks corresponding to the number of blocks based on the reduction ratio acquired by the reduction ratio acquisition means, and the processing block number calculation for each pixel of the output image Sought by means And a mean pixel value calculating means for calculating an average pixel value of the region of the original image corresponding to the number of divided blocks.

  The image processing device according to claim 11 is the image processing device according to claim 10, wherein the processing block number calculation means represents the number of divided blocks in the main scanning direction or the sub-scanning direction based on the reduction ratio as a decimal number. In this case, it is treated as an integer number of the divided blocks by rounding down the decimal part.

  The image processing device according to claim 12 is the image processing device according to claim 10 or 11, wherein the processing block number calculation means calculates the number of the divided blocks corresponding to one pixel in the output image. In addition, the number of the divided blocks corresponding to one pixel of the output image is calculated from the divided block adjacent to the divided block already used in the average pixel value calculating means.

  The image processing device according to claim 13 is the image processing device according to any one of claims 10 to 12, wherein the main scanning direction is set for each pixel of the output image before the average pixel value calculation means. (1a) x1 (x1 is an integer greater than or equal to 0 and less than nDev) corresponding to the unprocessed portion in the pixel that leaves the unprocessed portion in the original image, and (1b) the element In the image, an area adjacent to the area (1a) and in which N complete pixels are arranged (N is an integer of 1 or more), and (1c) in the original image, adjacent to the area (1b), 1 pixel Main scanning direction dividing means that divides the area into x2 (x2 is an integer less than or equal to 0 and less than nDev) remaining parts, and the average pixel value calculating means includes the main scanning direction dividing means Divided For each of the regions, we obtain the sum of the pixel values, the sum of the pixel values obtained by dividing the number of the divided blocks corresponding to one pixel of the output image obtained by the processing block number calculating means.

  The image processing device according to claim 14 is the image processing device according to any one of claims 10 to 13, wherein the sub-scanning direction is set for each pixel of the output image before the average pixel value calculation means. (2a) an area composed of y1 (y1 is an integer greater than or equal to 0 and less than nDev) corresponding to the unprocessed portion in a pixel that leaves an unprocessed portion in the original image; and (2b) the element In the image, an area adjacent to the area (2a) and in which N complete pixels are arranged (N is an integer of 1 or more), and (2c) in the original image, adjacent to the area (2b), 1 pixel Sub-scanning direction partitioning means for partitioning into regions consisting of the remaining portions of y2 (y2 is an integer less than or equal to 0 and less than nDev), and the average pixel value calculating means includes the sub-scanning direction partitioning means Divided The sum of pixel values is obtained for each area or for each area generated as a result of the division by the sub-scanning direction dividing means and the main scanning direction dividing means, and the sum of the pixel values is calculated as the number of processing blocks. It is obtained by dividing by the number of the divided blocks corresponding to one pixel of the output image obtained by the means.

  The image processing device according to claim 15 is the image processing device according to any one of claims 10 to 14, wherein an input image for obtaining a value of the number of input pixels (ix × iy) input from the input means. Size acquisition means, and output image size input means for inputting a value of the number of pixels (ox × oy) of the output image, wherein the reduction ratio acquisition means is the input image acquired by the input image size acquisition means. The reduction ratio is obtained from the value of the number of pixels (ix × iy) and the value of the number of pixels (ox × oy) of the output image input by the output image size input means.

  The image processing device according to claim 16 is the image processing device according to any one of claims 10 to 15, wherein the output image having the number of pixels (ix × iy) to the number of pixels (ox × oy) is output. 16. The reduction rate input means for inputting a reduction rate for obtaining, wherein the reduction rate acquisition means acquires the reduction rate input by the reduction rate input means. The image processing apparatus described.

  According to the image processing program of claim 1, the reduction ratio for converting the substantially rectangular original image having the number of pixels (ix × iy) into the substantially rectangular output image having the number of pixels (ox × oy) is reduced. The rate acquisition step is acquired by causing the arithmetic device to execute, while the division number acquisition step is executed by the arithmetic device, so that each pixel constituting the original image is (nDev × nDev) (nDev is 2 or more). The division number nDev for dividing into (integer) divided blocks is acquired.

  Next, by causing the arithmetic unit to execute the processing block number calculation step, the number of integer divided blocks corresponding to one pixel of the output image is determined based on the reduction rate acquired as the execution result of the reduction rate acquisition step. It is calculated for each pixel of the output image.

  Next, by causing the arithmetic unit to execute the average pixel value calculation step, the average pixel value of the original image area corresponding to the number of divided blocks obtained as the execution result of the processing block number calculation step is calculated for each pixel of the output image. Calculated every time.

  Therefore, since the average pixel value in one pixel of the reduced image (output image) is calculated in units of divided blocks generated by dividing the original image into nDev pieces, the decimal number generated according to the reduction rate is calculated by floating-point arithmetic. The average pixel value can be obtained by integer calculation instead of calculation. As a result, the processing speed of the average pixel method is remarkably improved as compared with the conventional method, and there is an effect that a high-quality reduced image can be obtained at a high speed. In addition, since the average pixel value can be calculated by integer arithmetic, there is an effect that it is suitable for mounting in an embedded system.

  In addition, since the reduced image can be obtained without increasing the image size of the original image to the least common multiple of the image size of the reduced image, the processing speed of the average pixel method is significantly improved compared to the conventional method. There is an effect that a high-quality reduced image can be obtained at a high speed.

  According to the image processing program of the second aspect, in addition to the effect produced by the image processing program of the first aspect, in the processing block number calculation step, the number of divided blocks in the main scanning direction or the sub scanning direction based on the reduction ratio is determined. When expressed as a decimal, it is handled as an integer number of divided blocks by truncating the decimal part.

  Accordingly, the average pixel value is calculated in units of one divided block by truncating the decimal part. As a result, the average pixel value can be calculated by integer arithmetic. As a result, the processing speed of the average pixel method is significantly improved compared to the conventional method, and a high-quality reduced image can be obtained at a high speed. There is an effect.

  According to the image processing program of claim 3, in addition to the effect produced by the image processing program of claim 1 or 2, in the processing block number calculation step, the number of divided blocks corresponding to one pixel in the output image In calculating the number of divided blocks corresponding to one pixel of the output image, the divided block adjacent to the divided block already used by the execution of the average pixel value calculating step is used as the starting point.

  Therefore, all the pixel information included in the original image is used to obtain a reduced image (output image). Therefore, it is possible to suppress image quality deterioration due to reduction and to obtain an output image with high image quality.

  According to the image processing program of the fourth aspect, in addition to the effect produced by the image processing program according to any one of the first to third aspects, the main scanning direction division step is calculated before the execution of the average pixel value calculating step. By causing the apparatus to execute, for each pixel of the output image, the main scanning direction of each pixel is (1a) x1 corresponding to the unprocessed portion in the pixel that leaves the unprocessed portion in the original image (x1 is An area composed of divided blocks of 0 or more and less than nDev), and (1b) an area adjacent to the area (1a) in the original image and arranged with N complete pixels (N is an integer of 1 or more); (1c) In the original image, the area is divided into an area that is adjacent to the area (1b) and is composed of x2 remaining parts (x2 is an integer not less than 0 and less than nDev) that is less than one pixel.

  In the average pixel value calculation step, the sum of the pixel values is obtained for each area divided as the execution result of the main scanning direction division step, and the sum of the pixel values is used as the execution result of the processing block number calculation step. Divide by the number of divided blocks corresponding to one pixel of the obtained output image.

  Therefore, in the main scanning direction of one pixel of the output image, the region (1b) composed of N complete original image pixels and the original image region (1a, 1c less than one pixel adjacent to both sides) ), It is possible to easily calculate the number of divided blocks included in the area (1a, 1c) of the original image that is less than one pixel, and as a result, it is possible to easily calculate the average pixel value.

  According to the image processing program of the fifth aspect, in addition to the effect produced by the image processing program according to the fourth aspect, the pixel including the region (1a) is obtained by causing the arithmetic unit to execute the main scanning direction pixel position calculating step. The pixel position in the main scanning direction is calculated.

  Therefore, as a result of calculating the pixel position in the main scanning direction of the pixel including the region (1a), the pixel position of the first pixel in the region (1b) adjacent in the main scanning direction of the pixel including the region (1a) The pixel position in the main scanning direction of the pixel including the region (1c) can be easily obtained from the number of pixels included in the region (1b). As a result, even when one pixel of the output image includes the area (1a) and the area (1c) of the original image that is less than one pixel, there is an effect that the average pixel value can be easily calculated.

  According to the image processing program of claim 6, in addition to the effect produced by the image processing program according to any one of claims 1 to 5, the sub-scanning direction division step is calculated before the execution of the average pixel value calculation step. By causing the apparatus to execute, for each pixel of the output image, the sub-scanning direction of each pixel is (2a) y1 corresponding to the unprocessed part in the pixel that leaves the unprocessed part in the original image (y1 is An area composed of divided blocks of 0 or more and less than nDev), and (2b) an area in which M pixels (M is an integer of 1 or more) are arranged adjacent to the area (2a) in the original image, (2c) In the original image, the area is divided into areas that are adjacent to the area (2b) and are composed of y2 remaining parts (y2 is an integer of 0 or more and less than nDev) that is less than one pixel.

  In the average pixel value calculation step, the pixel value is obtained for each area divided as the execution result of the sub-scanning direction division step or for each area divided as the execution result of the sub-scanning direction division step and the main scanning direction division step. Then, the sum of the pixel values is divided by the number of divided blocks corresponding to one pixel of the output image obtained as the execution result of the processing block number calculation step.

  Therefore, the sub-scanning direction of one pixel of the output image is divided into an area (2b) composed of M complete original image pixels and an original image area (2a, 2c less than one pixel adjacent to both sides). ), It is possible to easily calculate the number of divided blocks included in the area (2a, 2c) of the original image that is less than the pixels, and as a result, it is possible to easily calculate the average pixel value.

  According to the image processing program of the seventh aspect, in addition to the effect produced by the image processing program according to the sixth aspect, by causing the arithmetic unit to execute the sub-scanning direction pixel position calculating step, the pixel including the region (2a) The pixel position in the sub-scanning direction is calculated.

  Therefore, as a result of calculating the pixel position in the sub-scanning direction of the pixel including the region (2a), the pixel position of the first pixel in the region (2b) adjacent in the sub-scanning direction of the pixel including the region (2a) The pixel position in the sub-scanning direction of the pixel including the region (2c) can be easily obtained from the number of pixels included in the region (2b). As a result, even when one pixel of the output image includes the area (2a) and the area (2c) of the original image that is less than one pixel, the average pixel value can be easily calculated.

  According to the image processing program of the eighth aspect, in addition to the effect produced by the image processing program according to any one of the first to seventh aspects, the reduction ratio acquisition step is acquired as an execution result of the input image size acquisition step. The reduction ratio is acquired from the value of the number of pixels of the input image (ix × iy) and the value of the number of pixels of the output image (ox × oy) input as an execution result of the output image size input step.

  Therefore, there is an effect that the input image can be reduced at a reduction rate according to the input output image size (number of pixels of the output image). The “number of pixels of the output image input by the reduced image size input step” may be an output image size (number of pixels of the output image) designated by the user, or a condition (such as the size of the print medium) The output image size automatically determined according to () may be used.

  According to the image processing program of the ninth aspect, in addition to the effect produced by the image processing program according to any one of the first to eighth aspects, the reduction rate acquisition step is inputted as an execution result of the reduction rate input step. Get the reduction ratio. The “reduction ratio input as the execution result of the reduction ratio input step” is a reduction ratio for obtaining an output image having the number of pixels (ox × oy) from an input image having the number of pixels (ix × ii), that is, The value of ox / ix and / or oy / iy.

  Therefore, there is an effect that the input image can be reduced at the input reduction rate. The “reduction ratio input in the reduction ratio input step” may be a desired reduction ratio designated by the user, or is automatically determined according to conditions (such as the size of the print medium). It may be a reduction rate.

  The image processing apparatus according to claim 10, wherein the substantially rectangular original image having the number of pixels (ix × iy) input from the input unit is reduced by the reduction unit at a predetermined reduction rate, and the number of pixels (ox The output image is a substantially rectangular output image composed of (× oy).

  Here, in the reduction means, the reduction ratio for obtaining the original image composed of the number of pixels (ix × ii) as the output image composed of the number of pixels (ox × oy) is acquired by the reduction ratio acquisition means, The division number obtaining means obtains a division number nDev for dividing each pixel constituting the original image into (nDev × nDev) (nDev is an integer of 2 or more) divided blocks.

  In the reduction means, after the reduction ratio acquisition means and the division number acquisition means, the processing block number calculation means obtains an integer number of divided blocks corresponding to one pixel of the output image by the reduction ratio acquisition means. The average pixel value of the original image area corresponding to the number of divided blocks calculated by the average pixel value calculation means and further calculated by the average pixel value calculation means based on the reduction ratio. Is calculated for each pixel of the output image.

  Therefore, since the average pixel value in one pixel of the reduced image (output image) is calculated in units of divided blocks generated by dividing the original image into nDev pieces, the decimal number generated according to the reduction rate is calculated by floating-point arithmetic. The average pixel value can be obtained by integer calculation instead of calculation. As a result, the processing speed of the average pixel method is remarkably improved as compared with the conventional method, and there is an effect that a high-quality reduced image can be obtained at a high speed. In addition, since the average pixel value can be calculated by integer arithmetic, there is an effect that it is suitable for mounting in an embedded system.

  In addition, since the reduced image can be obtained without increasing the image size of the original image to the least common multiple of the image size of the reduced image, the processing speed of the average pixel method is significantly improved compared to the conventional method. There is an effect that a high-quality reduced image can be obtained at a high speed.

  According to the image processing apparatus of the eleventh aspect, in addition to the effect produced by the image processing apparatus of the tenth aspect, the processing block number calculating means determines the number of divided blocks in the main scanning direction or the sub scanning direction based on the reduction ratio. When expressed as a decimal, it is handled as an integer number of divided blocks by truncating the decimal part.

  Accordingly, the average pixel value is calculated in units of one divided block by truncating the decimal part. As a result, the average pixel value can be calculated by integer arithmetic. As a result, the processing speed of the average pixel method is significantly improved compared to the conventional method, and a high-quality reduced image can be obtained at a high speed. There is an effect.

  According to the image processing device of the twelfth aspect, in addition to the effect produced by the image processing device according to the tenth or eleventh aspect, the processing block number calculating means calculates the number of divided blocks corresponding to one pixel in the output image. When calculating the number of divided blocks corresponding to one pixel of the output image, a divided block adjacent to the divided block already used by the average pixel value calculating means is calculated.

  Therefore, all the pixel information included in the original image is used to obtain a reduced image (output image). Therefore, it is possible to suppress image quality deterioration due to reduction and to obtain an output image with high image quality.

  According to the image processing apparatus of the thirteenth aspect, in addition to the effect produced by the image processing apparatus according to any one of the tenth to twelfth aspects, the output is performed by the main scanning direction division means before the average pixel value calculation means. For each pixel of the image, the main scanning direction of each pixel is (1a) x1 corresponding to the unprocessed portion in the pixel that leaves the unprocessed portion in the original image (x1 is an integer greater than or equal to 0 and less than nDev) (1b) in the original image, an area that is adjacent to the area (1a) and in which N complete pixels are arranged (N is an integer of 1 or more), and (1c) in the original image It is divided into regions that are adjacent to (1b) and consist of x2 remaining portions (x2 is an integer not less than 0 and less than nDev) that is less than one pixel.

  The average pixel value calculating means calculates the sum of the pixel values for each area divided by the main scanning direction dividing means, and then calculates the sum of the pixel values of the output image obtained by the processing block number calculating means. Divide by the number of divided blocks corresponding to one pixel.

  Therefore, in the main scanning direction of one pixel of the output image, the region (1b) composed of N complete original image pixels and the original image region (1a, 1c less than one pixel adjacent to both sides) ), The number of divided blocks included in the region (1a) in the pixel to be processed next in the output image can be easily obtained based on the number of divided blocks included in the region (1c). There is an effect.

  According to the image processing apparatus of the fourteenth aspect, in addition to the effect produced by the image processing apparatus according to any one of the tenth to thirteenth aspects, the output is performed by the sub-scanning direction division means before the average pixel value calculation means. For each pixel of the image, the sub-scan direction of each pixel is (2a) y1 corresponding to the unprocessed portion in the pixel that leaves the unprocessed portion in the original image (y1 is an integer greater than or equal to 0 and less than nDev) (2b) in the original image, an area adjacent to the area (2a) and arranged with M complete pixels (M is an integer of 1 or more), and (2c) an area in the original image Adjacent to (2b), the area is divided into y2 (y2 is an integer greater than or equal to 0 and less than nDev) remaining area that is less than one pixel.

  The average pixel value calculation means obtains the sum of the pixel values for each area divided by the sub-scanning direction dividing means or for each area divided by the sub-scanning direction dividing means and the main scanning direction dividing means. Thus, the sum of the pixel values is divided by the number of divided blocks corresponding to one pixel of the output image obtained by the processing block number calculation means.

  Therefore, the sub-scanning direction of one pixel of the output image is divided into an area (2b) composed of M complete original image pixels and an original image area (2a, 2c less than one pixel adjacent to both sides). ), The number of divided blocks included in the region (2a) in the pixel to be processed next in the output image can be easily obtained based on the number of divided blocks included in the region (2c). There is an effect.

  According to the image processing device of the fifteenth aspect, in addition to the effect produced by the image processing device according to any one of the tenth to fourteenth aspects, the number of pixels of the input image (ix × The reduction ratio is acquired by the reduction ratio acquisition unit from the value of iy) and the value of the number of pixels (ox × oy) of the output image input by the output image size input unit.

  Therefore, there is an effect that the input image can be reduced at a reduction rate according to the input output image size (number of pixels of the output image). The “number of pixels of the output image input by the reduced image size input unit” may be an output image size (number of pixels of the output image) designated by the user, or a condition (such as the size of the print medium) The output image size automatically determined according to () may be used.

  According to the image processing apparatus of the sixteenth aspect, in addition to the effect produced by the image processing apparatus according to any one of the tenth to fifteenth aspects, the reduction ratio input by the reduction ratio input means is reduced by the reduction ratio acquisition means. To be acquired. The “reduction rate input by the reduction rate input means” is a reduction rate for obtaining an output image having the number of pixels (ox × oy) from the input image having the number of pixels (ix × iy), that is, ox / ix. And / or oy / iy.

  Therefore, there is an effect that the input image can be reduced at the input reduction rate. The “reduction ratio input by the reduction ratio input unit” may be a desired reduction ratio designated by the user, or automatically determined according to conditions (size of the print medium, etc.). It may be a reduction rate.

  Hereinafter, preferred embodiments of the present invention will be described with reference to the accompanying drawings. FIG. 1 is a block diagram showing an electrical configuration of a printer 1 according to an embodiment of the present invention.

  The printer 1 is configured so that image data input from a personal computer (hereinafter referred to as “PC”) 2, a digital camera 21, or an external medium 20 can be reduced and output at a predetermined reduction rate. ing.

  The printer 1 includes a CPU 11, a ROM 12, a RAM 13, a print head, and the like. The printer 1 performs printing (output) on a print medium (for example, a paper medium), and input values such as an output image size. And an operation panel 16 having a user operation unit (for example, a numeric keypad) that can be input by the user.

  The printer 1 includes an interface (hereinafter referred to as “I / F”) 17 that can be connected to the PC 2 via the cable 5, an I / F 18 that can be connected to the digital camera 21 via the cable 6, and an external medium 20. (For example, an SD memory card, a memory stick, etc.) and a removable external media slot 19 are provided.

  Therefore, the printer 1 can input the image data stored in the PC 20 via the cable 5 and the I / F 17 and also the image data photographed by the digital camera 21 via the cable 6 and the I / F 18. It is possible to input via. Further, it is possible to input image data stored in the external medium 21 from the external medium 20 mounted in the external media slot 19.

  The CPU 11 is an arithmetic processing device (arithmetic device) that controls the entire printer 1, and the ROM 12 stores various control programs executed by the CPU 11 and fixed value data that is referred to during the execution. Yes. Note that the ROM 12 has an area for storing an image reduction program 12a for executing the flowcharts shown in FIGS.

  The RAM 13 has a working area in which various register groups necessary for a control program executed by the CPU 11 are set, a temporary area for temporarily storing data being processed, and the like, which can be accessed at random. It is memory.

  The RAM 13 includes an original image memory 13a, an output image memory 13b, an output size memory 13c, and a reduction rate memory 13d. The original image memory 13 a is image data input from the PC 2, the digital camera 21, and the external medium 20 to the printer 1 via the I / F 17, I / F 18, and external media slot 19, respectively. Image data (hereinafter, referred to as “original image P” for convenience) is stored (hereinafter, the image data of the original image P is referred to as “original image data”).

  The output image memory 13b is an output image generated by reducing the original image data stored in the original image memory 13a in accordance with an image reduction program 12a (a program for executing flowcharts shown in FIGS. 6 to 10 described later). Image data (hereinafter referred to as “output image Q” for convenience) is stored (hereinafter, the image data of the output image Q is referred to as “output image data”). In the present embodiment, both the original image data and the output image data are composed of RGB.

  The output size memory 13c stores the size of the output image Q (that is, the output image size) input by the user operating the operation panel 16, and the reduction ratio memory 13d is stored in the original image memory 13a. The reduction ratio calculated from the size of the original image P and the output image size stored in the output size memory 13c is stored.

  Here, as will be described later, the image reduction program 12a generates the output image data by reducing the original image data stored in the original image memory 13a at the reduction rate stored in the reduction rate memory 13d. To do. Therefore, output image data generated by reduction according to the image reduction program 12a is stored in the output image memory 13b.

  Next, referring to FIG. 2, the reduction function by the printer 1 configured as described above will be described. FIG. 2 is a functional block diagram showing a reduction function by the printer 1. As shown in FIG. 2, the printer 1 uses the original image data of the image size (ix.ii) input from the input means (not shown) as the output image size (ox) input from the input means (not shown). . Oy), and a reduction unit 100 that is a portion that outputs as output image data of an output image size (ox.oy) from output means (not shown).

  More specifically, the reduction unit 100 includes an image size comparison unit 101, a reduction ratio calculation unit 102, a processing block number calculation unit 103, a processing block reduction unit 104, and a division number determination unit 105. .

  The image size comparison unit 101 is a part that compares the image size (ix.ii) of the original image data input from the input means (not shown) with the output image size (ox, oy), and the comparison result is obtained. Output to the reduction ratio calculation unit 102.

  The reduction rate calculation unit 102 is a part that calculates the reduction rate from the comparison result input from the image size comparison unit 101, and outputs the calculated reduction rate to the processing block number calculation unit. When the image size of the original image P is (ix, iy) and the output image size is (ox, oy), the reduction ratio is (ox / ix) in the main scanning direction, and the sub-scanning direction. (Oy / iy).

  The division number determination unit 105 is based on original image data input from an input unit (not shown), an output image size, a value indicating the processing capability of the printer 1, and the number of color data (palettes) that one pixel can take. This is a part for calculating the division number nDev for one pixel in the original image P, and outputs the calculated division number nDev to the processing block number calculation unit 103.

  Here, the division number nDev is a number indicating how many divisions are made in the main scanning direction (x direction) and the sub scanning direction (y direction) of each pixel of the original image P. Is divided by the division number nDev, and each pixel is divided into (nDev × nDev) fragments (hereinafter, each fragment obtained as a result of dividing one pixel into the division number nDev is referred to as a “division block”. Called). Since the division number nDev is an integer of 2 or more, one pixel is divided into an integer number ((nDev × nDev)) of divided blocks.

  However, since the value of the division number nDev must be a numerical value within the range that can be processed by the CPU 11 that is the arithmetic processing unit, the division number determining unit 105 sets the value of the division number nDev to the CPU 11 that is the arithmetic processing unit. Determine the value within the processable range. A configuration for determining the value of the division number nDev in the division number determination unit 105 to a numerical value within a range that can be processed by the CPU 11 that is the arithmetic processing unit will be described later with reference to FIG.

  The processing block number calculation unit 103 calculates the number of divided blocks corresponding to one pixel of the output image Q from the reduction rate input from the reduction rate calculation unit 102 and the division number nDev input from the division number determination unit 105. This is a calculation part, and the calculated number of divided blocks is output to the average pixel value calculation unit 104.

  The average pixel value calculation unit 104 is a part that calculates the average pixel value of the original image data area corresponding to the number of divided blocks input from the processing block number calculation unit 103. Although details will be described later with reference to FIGS. 3 and 4, when the average pixel value is calculated by the average pixel value calculation unit 104, the divided blocks included in one pixel of the output image Q are subdivided. The average pixel value is calculated above.

  Here, the calculation of the number of divided blocks by the processing block number calculation unit 103 and the calculation of the average pixel value by the average pixel value calculation unit 104 are repeated by the number of pixels of the output image Q. When the average pixel value is calculated by the average pixel value calculation unit 104 for all the pixels in the output image Q, the original image data input from the input unit (not shown) is input to the input unit (not shown). The output image data reduced to the output image size input from () is output from the output means (not shown).

  Next, with reference to FIG. 3, a more detailed configuration for the division number determination unit 105 will be described. FIG. 3 is a functional block diagram illustrating functions of the division number determining unit 105. As shown in FIG. 3, the division number determination unit 105 includes a maximum block number calculation unit 105a and a division number calculation unit 105b.

  The maximum block number calculation unit 105a is a part that calculates the maximum value of the number of divided blocks that one pixel of the output image Q can take.

When the image size (number of pixels) of the original image P is (ix × iy) and the image size (number of pixels) of the output image Q is (ox × oy), the coordinates (i _out , j _out ) in the output image Q The maximum values (nBlock_x) _max and (nBlock_y) _max that can be taken in the main scanning direction and the sub-scanning direction for the divided blocks included in the pixel are the values represented by the following equations (1a) and (1b).

  Here, nBlock_x (0) and nBlock_y (0) are the numbers of divided blocks in the main scanning direction and the sub-scanning direction included in the pixel included in the pixel of the first coordinate (0, 0) in the output image Q, respectively. . The operation symbol [] is a Gaussian symbol, and is a symbol representing the maximum integer that does not exceed the number.

  The maximum block number calculation unit 105a calculates the above formulas (1a) and (1b), and outputs the calculation result to the division number calculation unit 105b.

  The division number calculation unit 105b calculates the value of the division number nDev as a numerical value within a range that can be processed by the CPU 11 which is an arithmetic processing unit, using the maximum value of the maximum block number input from the maximum block number calculation unit 105a. To do.

  Here, in order for the high-speed average pixel processing (the processing of S602 in FIG. 6) in the present embodiment to be executed by the CPU 11 which is the arithmetic processing unit of the printer 1, the maximum value indicating the processing capability of the CPU 11 is set to ValueMax. When the maximum value of color data (palette) that can be taken by one pixel is PalletMax, the following inequality (2) must be satisfied.

  Note that PalletMax, which is the maximum value of color data (palette) that can be taken by one pixel, is the maximum value of the number of bits (for example, 8 bits or 16 bits) constituting one pixel. That is, for example, when the number of bits constituting one pixel is 8 bits, the value of PalletMax is 256.

  Here, when the remainder of ix / ox is rx, the remainder of iy / oy is ry, (ix−rx) / ox is represented as Dx, and (iy−ry) / oy is represented as Dy, the above formula Considering that nDev is an integer of 2 or more, nDev that satisfies (2) is an integer that satisfies the following inequality (3).

  If the division number nDev satisfies this inequality (3), overflow of the CPU 11 at the time of reduction is surely prevented.

  Here, since the values of ValueMax and PalletMax are values determined by the CPU 11, the division number calculation unit 105b uses these ValueMax and PalletMax and the maximum value of the maximum block number input from the maximum block number calculation unit 105a. Then, the division number nDev satisfying the inequality (3) is calculated, and the value of the calculated division number nDev is output to the processing block number calculation unit 103.

  As described above, the value of the division number nDev is determined in consideration of the reduction ratio (that is, ix / ox and iy / oy) with respect to the maximum value ValueMax indicating the processing capability of the CPU 11 (arithmetic processing unit). The Therefore, even when the original image P is reduced to various output image sizes, the value of the division number nDev can be appropriately used according to the reduction ratio as long as the inequality (3) is satisfied.

  In particular, the larger the value of the division number nDev, the smaller the size of each divided block (that is, the higher the resolution). Therefore, errors that can be caused by using divided block units are small, and the image quality of the output image Q is small. Will improve. Therefore, by using the maximum number of divisions nDev that satisfies the above formula (3) according to the reduction ratio, it is possible to obtain an output image Q having the highest possible image quality for any reduction ratio. It can be done.

  Further, the value of the division number nDev is determined in consideration of the maximum value (PalletMax) of color data (pallet) that one pixel can take with respect to the maximum value ValueMax indicating the processing capability of the CPU 11 (arithmetic processing unit). Therefore, the value of the division number nDev can be appropriately used according to various bit colors.

  Before describing the image reduction process (see FIG. 6) executed by the printer 1 having the above-described configuration, a high-speed average pixel process (in FIG. 6) as a part of the image reduction process will be described with reference to FIGS. The variables used in step S602) will be described. 4 and 5 are schematic diagrams for explaining variables used in the high-speed average pixel process (the process of S602 in FIG. 6) of the present embodiment.

  First, for the purpose of facilitating the understanding of the present invention, in FIG. 4, reduction in the sub-scanning direction (y direction) is not considered, and the high-speed average pixel processing (S602 in FIG. 6) described later in the main scanning direction (x direction). The variables that are required in the case of being used in the above process) will be described. As shown in FIG. 4, the original image P is an image composed of five pixels P0 to P4 arranged in the main scanning direction, and the output image Q is an image reduction process (see FIG. 6) described later on the original image P. It is the image obtained as a result of.

  In image reduction processing (see FIG. 6) described later, particularly high-speed average pixel processing (processing in S602 in FIG. 6), each pixel is handled in units of divided blocks. Here, “nDev” is the number of divisions for one pixel of the original image P, that is, how many each pixel in the original image P is, as described when explaining the division number determination unit 105 (see FIG. 2). It is a variable that indicates whether it has been divided. In the example shown in FIG. 4, since each pixel of the original image P is divided into four, the value of the division number nDev is 4 (see the thin dotted line in FIG. 4).

  Note that the value of the division number nDev must be a numerical value within a range that can be processed by the CPU 11 that is an arithmetic processing unit. Further, as the value of the division number nDev is larger, the size of each divided block becomes finer (that is, the resolution becomes higher), and as a result, errors that can be caused by using divided block units are reduced. This is preferable in terms of improving the image quality of the output image Q. Therefore, the value of the division number nDev is most preferably the maximum value among the numerical values that can be processed by the CPU 11.

  As shown in FIG. 4, the output image Q is reduced to an output image size of three pixels Q0 to Q2 as a result of the reduction. As shown in FIG. 4, the pixel Q0 is a pixel obtained by reducing the pixel P0 and the two divided blocks adjacent to the pixel P0 in the pixel P1. The pixel Q1 includes the remaining two divided blocks that are not used for reducing the pixel Q0 in the pixel P1, the pixel P2 adjacent to the pixel P1, and one division adjacent to the pixel P2 in the pixel P3. This is a pixel obtained by reducing the block. The pixel Q2 is a pixel obtained by reducing the remaining three divided blocks not used for reducing the pixel Q1 in the pixel P3 and the pixel P4 adjacent to the pixel P3.

  In the present embodiment, since the area of the original image P corresponding to the pixel of the output image Q is acquired in units of divided blocks, an error occurs with respect to the actual area of the original image P corresponding to the reduction ratio. For example, in FIG. 4, the position of the starting point s ′ of the area of the original image P corresponding to the actual pixel Q1 according to the reduction ratio is the position of the starting point s of the area of the original image obtained in units of divided blocks. Different. However, the larger the value of the division number nDev, the smaller the size of each divided block (that is, the higher the resolution), so the error between the position of the start point s ′ and the position of the start point s is small. Become. As a result, it is possible to suppress deterioration in image quality due to an error (an error between the start point s and the start point s ′) caused by using the divided block unit. Therefore, as described above, the value of the division number nDev is preferably as large as possible, and is most preferably the maximum value among the numerical values that can be processed by the CPU 11.

  When the starting pixels of the original image P and the output image Q are the pixel P0 and the pixel Q0, respectively, in the high-speed average pixel processing (the processing of S602 in FIG. 6) described later, the high-speed average pixel processing in the order of the pixel Q0 → the pixel Q2. Are executed sequentially. FIG. 4 illustrates the case where the process for obtaining the pixel Q1 is performed.

  Here, “nBlock_x” is the number of divided blocks included in the area of the original image P corresponding to one pixel of the output image, and is a variable representing the number of divided blocks in the main scanning direction. As described above, the pixel Q1 includes the remaining two divided blocks that are not used for reducing the pixel Q0 in the pixel P1, the pixel P2 adjacent to the pixel P1, and the pixel P2 adjacent to the pixel P2 in the pixel P3. Since the pixel is obtained by reducing the number of divided blocks, the value of the variable nBlock_x in the example illustrated in FIG. 4 is 7 (= 2 + 4 + 1).

  In the high-speed average pixel processing (the processing in S602 in FIG. 6) described later, (nBlock_x) divided blocks that are reduced in the main scanning direction are processed in three sections. Specifically, the divided block being processed is an area of the original image that is less than one pixel left unprocessed in the previous high-speed average pixel process (hereinafter, this area is referred to as a “front residual area”). Included in the area of the original image (hereinafter referred to as the “central area”) in which N blocks (N is an integer greater than or equal to 1) are arranged adjacent to the front remaining area and adjacent one pixel is included. Of the original image that is not included in the front residual area or the central area and is less than one pixel (of the number of divided blocks adjacent to the central area and represented by the variable nBlock_x). Hereinafter, this area is referred to as “rear remaining area”) and is divided into divided blocks.

  Here, “nResidue_x0” is a variable that represents the number of divided blocks included in the front residual area in the main scanning direction, and “nMain_x” is a variable that represents the number of pixels included in the central area in the main scanning direction, “NResidue_x1” is a variable representing the number of divided blocks included in the rear remaining area in the main scanning direction.

  In the example illustrated in FIG. 4, the value of the variable nResidue_x0 is 2 that is the number of divided blocks in the main scanning direction with respect to the remaining portion that is not used for the reduction of the pixel Q0 in the pixel P1, and the value of the variable nMain_x is Since only the pixel P2 corresponds to the central region, it is 1. Further, the value of the variable nResidue_x1 is 1, which is the number of divided blocks in the main scanning direction that is less than one pixel adjacent to the pixel P2 in the pixel P3.

  In addition, “nPosRes_x0” is a variable representing coordinates in the main scanning direction with respect to pixels including the front residual area in the original image, and “nPosMain_x0” is the N pixels included in the central area in the original image. “NPosRes_x1” is a variable representing coordinates in the main scanning direction with respect to the pixel including the rear remaining area in the original image, with respect to the first pixel (pixel adjacent to the front remaining area). .

  Here, in the example shown in FIG. 4, the value of the variable nPosRes_x0 is the coordinate in the main scanning direction with respect to the pixel P1 including the front residual area, and the value of the variable nMainRes_x0 is the pixel P2 that is the only pixel included in the central area. The value of the variable nPosRes_x1 is the coordinate in the main scanning direction for the pixel P3 including the rear residual area.

  On the other hand, “nReadBlock_x” is a variable representing the number of divided blocks in the main scanning direction that have been processed (used) in the high-speed average pixel processing (the processing of S602 in FIG. 6) up to the previous time. In the example shown in FIG. 4, since the divided block corresponding to the image Q1 has already been processed, the value of the variable nReadBlock_x is the number of divided blocks in the main scanning direction included in the area of the original image P corresponding to the image Q1, that is, , 6 (= 4 + 2).

  As described above, in order to facilitate understanding, reduction in only the main scanning direction (x direction) is considered, and the variables necessary for the reduction have been described. Since reduction is performed, the reduction in the sub-scanning direction (y direction) is considered as shown in FIG. In FIG. 5, the pixels of the original image P and the pixels of the output image Q are overlapped, and in order to easily distinguish these pixels, only one pixel being processed is selected as the pixel of the output image Q. And one of the pixels of the original image P is hatched with diagonal lines.

  In FIG. 5, as shown in FIG. 5, since the original image P is divided by the division number 4 (= nDev) in both the main scanning direction and the sub-scanning direction, each pixel of the original image P is 16 (= It is divided into nDev × nDev) divided blocks (see thin dotted lines in FIG. 5).

  Here, “nBlock_y”, which is a variable required for reduction in the sub-scanning direction, is a variable corresponding to the above-described variable “nBlock_x” required for reduction in the main scanning direction. This is a variable representing the number of divided blocks included in the corresponding area of the original image P and indicating the number of divided blocks in the sub-scanning direction.

  The variable “nReadBlock_y” is a variable corresponding to the variable “nReadBlock_x” required for the above-described reduction in the main scanning direction, and the number of divided blocks in the sub-scanning direction that has already been processed in the high-speed average pixel processing up to the previous time. It is a variable that represents.

  Also in the sub-scanning direction, in the high-speed average pixel processing (the processing in S602 in FIG. 6) described later, similarly to the main scanning direction described above, the (nBlock_y) divided blocks are reduced to the front residual area. Processing is divided into three sections, a central area and a rear residual area.

  The variable “nResidue_y0” is a variable corresponding to the variable “nResidue_x0” required for the reduction in the main scanning direction, and is a variable representing the number of divided blocks included in the front remaining area in the sub-scanning direction. The variable “nMain_y” is a variable corresponding to the variable “nMain_x” required for the reduction in the main scanning direction, and is a variable representing the number of pixels included in the central region in the sub-scanning direction. The variable “nResidue_y1” is a variable corresponding to the variable “nResidue_x1” required for the reduction in the main scanning direction, and is a variable representing the number of divided blocks included in the rear remaining area in the sub-scanning direction. .

  As shown in FIG. 5, (nBlock_x) × (nBlock_y) divided blocks are divided into three sections (nResidue_x0, nMain_x, nResidue_x1) in the main scanning direction and three sections (nResidue_y0, nMain_e, nReid_e, nResid_e, nReid_e, nResid ) Is divided into nine areas A1 to A9.

  The variable “nPosRes_y0” is a variable corresponding to the variable “nPosRes_x0” required for the reduction in the main scanning direction, and represents the coordinates in the sub-scanning direction with respect to the pixels including the front remaining area in the original image. It is. The variable “nPosMain_y0” is a variable corresponding to the variable “nPosMain_x0” necessary for the reduction in the main scanning direction, and is included in the central region in the original image (M is an integer of 1 or more). This is a variable representing coordinates in the sub-scanning direction with respect to the first pixel (pixel adjacent to the front residual area). A variable “nPosRes_y1” is a variable corresponding to the variable “nPosRes_x1” required for the reduction in the main scanning direction, and represents a coordinate in the sub-scanning direction with respect to the pixel including the rear residual area in the original image. It is.

  The number of divided blocks included in the area A1 is (nResidue_x0 × nResidue_y0), the number of divided blocks included in the area A2 is (nMain_x × nDev × nResidue_y0), and the number of divided blocks included in the area A3. Are (nResidue_x1 × nResidue_y0).

Similarly, the number of divided blocks included in the region A4 is (nResidue_x0 × nMain_y × nDev), and the number of divided blocks included in the region A5 is (nMain_x × nMain_y × nDev ² ) and included in the region A6. The number of divided blocks is (nResidue_x1 × nMain_y × nDev).

  Further, the number of divided blocks included in the region A7 is (nResidue_x0 × nResidue_y1), the number of divided blocks included in the region A8 is (nMain_x × nDev × nResidue_y1), and the number of divided blocks included in the region A9. Is (nResidue_x1 × nResidue_y1).

  Although details will be described later, in the high-speed average pixel processing (processing of S602 in FIG. 6) in the present embodiment, each pixel of the original image P is thus divided into nDev divided blocks and processed in units of divided blocks. As a result, the pixel value of one pixel of the output image Q can be obtained without enlarging to the least common multiple and without using a floating point in the calculation, so that the reduction by the average pixel method is accelerated. It can be done.

  Next, an image reduction process executed by the printer 1 having the above-described configuration will be described with reference to the flowchart of FIG. FIG. 6 is a flowchart illustrating image reduction processing executed by the printer 1. In this image reduction process, when a substantially rectangular original image P (original image data) having an image size (number of pixels) of (ix × ii) is input to the printer 1 (for example, via the cable 6 and the I / F 18). And the original image P is input from the digital camera 21).

  As shown in FIG. 6, in this image reduction process, first, the original image data input to the printer 1 is stored in the original image memory 13a (S601), and the original image data stored in the original image memory 13a is stored in the original image data. Using the accelerated average pixel method, the image is reduced to the output image size (ox × oy) input by the user, and is substantially rectangular output image data (output) having an image size (number of pixels) of (ox × oy). High-speed average pixel processing for generating the image Q) is executed (S602). Details of the high-speed average pixel processing (S602) will be described later with reference to FIG.

  After execution of the high-speed average pixel processing (S602), edge sharpening processing is performed on the obtained output image data (S603). The edge sharpening processing performed in S603 is not a gist of the present invention and is a known technique, and thus a detailed description thereof will be omitted. For example, a sharpening filter configured by a 3 × 3 matrix ( For example, each pixel value of the output image data is filtered using a Laplacian filter.

  After the process of S603, the output image data stored in the output image memory 13b as a result of the high-speed average pixel process (S602) is output to the printing unit 15 while rotating (S604), and this image reduction process is terminated. As a result of S604, the output image Q obtained by reducing the original image P to the output image size designated by the user is printed on the print medium in the printing unit 15.

  Next, the high-speed average pixel process (S602) executed in the above-described image reduction process (see FIG. 6) will be described with reference to the flowchart in FIG. FIG. 7 is a flowchart showing high-speed average pixel processing (S602).

  As shown in FIG. 7, in the high-speed average pixel processing (S602), first, the image size (ix × iy) of the original image P stored in the original image memory 13a and the output stored in the output size memory 13c. The reduction ratio is calculated from the image size, and the calculated reduction ratio is stored in the reduction ratio memory (S701). Note that the output image size (ox × oy) stored in the output size memory 13c is a value separately input from the operation panel 16 by the user.

  After the process of S701, the value of the division number nDev is calculated (S702). In this embodiment, in S702, the division number nDev that maximizes the image quality of the output image Q is calculated. That is, in this embodiment, in S702, the division number nDev is obtained by the following equation (4).

  Here, Dx is (ix−rx) / ox, rx is the remainder of ix / ox, Dy is (iy−ry) / oy, and ry is the remainder of iy / oy, ValueMax is the maximum value indicating the processing capability of the arithmetic unit, and PalletMax is the maximum color data (palette) that can be taken by one pixel.

  Since the division number nDev obtained by the above equation (4) is the maximum division number nDev that satisfies the above inequality (3), it is a value that reliably prevents overflow of the CPU 11 during reduction, and is output by the CPU 11. This is a reduced image with the highest possible image quality.

After the process of S702, variables nReadBlock_x, nResidue_x0, nPosRes_x0, nReadBlock_y, nResidue_y0, and both the value of nPosRes_y0 set to 0 (S703), the main scanning direction of the coordinate _{i out} and the sub-scanning direction of the coordinates of the pixels in the output image Q j _out is set to (0, 0) which is the top coordinate (S704).

After the process of S704, the sub-scanning direction block process is executed (S705). Although details will be described later with reference to FIG. 8A, the execution of the sub-scanning direction block process (S705) causes the original image P corresponding to the pixel of the coordinates (i _out , j _out ) in the coordinate output image Q to be processed. The values of nBlock_y, nMain_y, nResidue_y1, nPosMain_y0, and nPosRes_y1 that are variables in the sub-scanning direction are all obtained as integer values.

After the execution of the sub-scanning direction block process (S705), the main scanning direction block process is executed (S706). Although details will be described later with reference to FIG. 8B, the main image P corresponding to the pixel of the coordinates (i _out , j _out ) in the coordinate output image Q is obtained by executing the main scanning direction block processing (S706). The values of nBlock_x, nMain_x, nResidue_x1, nPosMain_x0, and nPosRes_x1, which are variables in the main scanning direction, are all obtained as integer values.

  After execution of the main scanning direction block processing (S706), the value of the variable Plane is set to zero. In the present embodiment, since the image data is composed of RGB, the value of the variable Plane can take three values from 0 to 2, and the red (R) plane (Plane = 0) is determined by the value of the variable Plane. ), A green (G) plane (Plane = 1), and a blue (B) plane (Plane = 2).

After the process of S707, a reduced image output process (S708) for acquiring the pixel value of the color plane corresponding to the plane value set in S707 or S712 described later is executed. Although details will be described later with reference to FIG. 9, in this reduced image output process (S708), the Plane value in the region of the original image P corresponding to the pixel at the coordinates (i _out , j _out ) in the output image Q is used. Calculate the average pixel value of the specified color plane.

  After execution of the reduced image output process (S708), it is confirmed whether the value of the variable Plane is 2 (S709). If the value of the variable Plane is 0 or 1 (S709: No), the value of the variable Plane is set to 1. Addition is performed (S712), and the process proceeds to reduced image output processing (S708). That is, the reduced image output process (S708) is executed for the color plane corresponding to the next plane value.

On the other hand, if the value of the variable Plane is 2 (S709: Yes) as a result of the confirmation in the process of S709, the pixel values of the pixels at the coordinates (i _out , j _out ) in the output image Q are all color planes ( Next, i _out = ox, that is, the pixel at the coordinates (i _out , j _out ) in the output image Q is the terminal pixel in the main scanning direction. (S710).

If i _out = ox is not confirmed as a result of the processing in S710 (S710: No), 1 is added to the value of i _out , and the pixel of the output image Q is advanced by 1 in the main scanning direction (S713). Next time initial value acquisition processing is executed (S714). Although details will be described later with reference to FIG. 10A, in this next initial value acquisition process in the main scanning direction (S714), the values of the variables nReadBlock_x, nResidue_x0, and nPosRes_x0 are updated.

  After the execution of the next initial value acquisition process (S714) in the main scanning direction, the process proceeds to the main scanning direction block process (S706), and in steps S713, variables nBlock_x, nMain_x, n corresponding to the pixels of the output image Q advanced by 1 in the main scanning direction. The values of nResidue_x1, nPosMain_x0, and nPosRes_x1 are obtained.

On the other hand, if i _out = ox as a result of the confirmation in S710 (S710: Yes), the reduced image output process (S708) is executed for all the pixels located at the coordinate j _out in the same sub-scanning direction. Next, j _out = oy, that is, the pixel at the coordinates (i _out , j _out ) in the output image Q is the end pixel in the sub-scanning direction. (S711).

As a result of checking in the process of S711, if j _{out is} not oy (S711: No), 1 is added to the value of j _out (S715), and the value of i _out is set to 0 (S716). As a result of the processing in S715 and S716, the position of the pixel selected from the pixels constituting the output image Q is advanced by 1 in the sub-scanning direction and returned to the top coordinate in the main scanning direction.

After the process of S716, the next initial value acquisition process in the sub-scanning direction is executed (S717). Although details will be described later with reference to FIG. 10B, the values of the variables nReadBlock_y, nResidue_y0, and nPosRes_y0 are updated in the next initial value acquisition process (S717) in the sub-scanning direction.

  After executing the next initial value acquisition process in the sub-scanning direction (S717), the next initial value acquisition process in the main scanning direction is executed (S718). This next initial value acquisition process in the main scanning direction (S718) is also a process for obtaining values of the variables nReadBlock_x, nResidue_x0, and nPosRes_x0 in the same manner as the next initial value acquisition process in the main scanning direction (S714). Although details will be described later with reference to FIG. 10A, since the next initial value acquisition process (S718) in the main scanning direction is a process for the top coordinate in the main scanning direction of the output image Q, the variables nReadBlock_x, nResidue_x0, The values of nPosRes_x are all set to 0.

  After the execution of the next initial value acquisition process (S718) in the main scanning direction, the process proceeds to the sub scanning direction block process (S705), and variables nBlock_y, nMain_y corresponding to the pixels of the output image Q advanced by 1 in the sub scanning direction in S715. The values of nResidue_y1, nPosMain_y0, and nPosRes_y1 are obtained.

On the other hand, if j _out = oy (S711: Yes), the reduced image output process (S708) is executed on all ((ox × oy)) pixels in the output image Q to obtain pixel values. The high-speed average pixel processing (S602) is terminated. The output image data for the output image Q of the output image size (ox × oy) obtained by reducing the original image P of the original image size (ix × iy) by the high-speed average pixel processing (S602) is output to the output image memory. 13b is stored.

  Next, the sub-scanning direction block processing (S705) and main scanning direction block processing (S706) executed in the above-described high-speed average pixel processing (see FIG. 7) will be described with reference to the flowchart of FIG. FIG. 8A is a flowchart showing sub-scanning direction block processing (S705), and FIG. 8B is a flowchart showing main scanning direction block processing (S706).

As shown in FIG. 8A, in the sub-scanning direction block processing (S705), first, the value of the variable nBlock_y is obtained from the expression nBlock_y = j _out × [(iy / oy) × nDev] −nReadBlock_y (S801). .

  As a result of S801, the number of divided blocks in the sub-scanning direction (nBlock_y) in one pixel of the output image Q is the divided block adjacent to the divided block in the sub-scanning direction that has already been processed in the high-speed average pixel processing (S602). Will be extracted starting from.

  In addition, the end point of the divided block in the sub-scanning direction in one pixel of the output image Q is obtained by rounding off the decimal part with a Gaussian symbol regardless of the reduction rate (oy / ii) in the sub-scanning direction. The variable nBlock_y can be expressed by the number of divided blocks, that is, an integer.

  After the processing of S801, the value of the variable nMain_y is obtained from the expression nMain_y = [(nBlock_y−nResidue_y0) / nDev] using the value of the variable nBlock_y obtained in S801 (S802).

  After the processing of S802, using the value of the variable nMain_y obtained in S802, the variable nResidue_y1 is obtained from the expression nResidue_y1 = nBlock_y-nResidue_y0- (nMain_y × nDev) (S803).

  Next, the value of the variable nPosMain_y0 is obtained from the equation = nPosRes_y0 + 1 (S804). Using the obtained value of the variable nPosMain_y0, the value of the variable nPosRes_y1 is obtained from the equation nPosRes_y1 = nPosMain_y0 × + nMain_y + 1, this sub direction 80 (S80). The block process (S705) is terminated.

As shown in FIG. 8 (b), the main scanning direction block process (S706), firstly, the value of the variable NBlock_x, determined from the equation _{nBlock_x = j out × [(ix} / ox) × nDev] -nReadBlock_x (S811) .

  As a result of S811, the number of divided blocks in the main scanning direction (nBlock_x) in one pixel of the output image Q is equal to the divided block adjacent to the divided block in the main scanning direction that has already been processed in the high-speed average pixel processing (S602). Will be extracted starting from.

  Further, the end point of the divided block in the main scanning direction in one pixel of the output image Q is obtained by rounding off the decimal part with a Gaussian symbol regardless of the reduction ratio (ox / ix) in the main scanning direction. The variable nBlock_x can be expressed by the number of divided blocks, that is, an integer.

  After the processing of S811, the value of the variable nMain_x is obtained from the expression nMain_x = [(nBlock_x−nResidue_x0) / nDev] using the value of the variable nBlock_x obtained in S811 (S812).

  After the processing of S812, using the value of the variable nMain_x obtained in S812, the variable nResidue_x1 is obtained from the expression nResidue_x1 = nBlock_x-nResidue_x0- (nMain_xx × nDev) (S813).

  Next, the value of the variable nPosMain_x0 is obtained from the equation = nPosRes_x0 + 1 (S814). Using the obtained value of the variable nPosMain_x0, the value of the variable nPosRes_x1 is obtained from the equation nPosRes_x1 = nPosMain_x0 × + nMain_x15 (main scan direction). The block process (S706) is terminated.

As a result of these sub-scanning direction block processing (S705) and main scanning direction block processing (S706), the pixels of the coordinates (i _out , j _out ) in the output image Q are divided into areas A1 to A9 (see FIG. 5). Will be.

Thus, the pixel of the coordinates (i _out , j _out ) in the output image Q has a region (A5) composed of pixels of the complete original image P, and a region (A1 to A1) that is less than one pixel of the original image P. A4, A6 to A9), the number of divided blocks of an area (A1 to A4, A6 to A9) less than one pixel is also equal to (number of divided blocks in the main scanning direction × sub-scanning direction) in that area. The number of divided blocks can be easily calculated. As a result, it is easy to calculate partial pixel values (see S901) and final average pixel values (S903) of each region in the reduced pixel output processing (S708) described below with reference to FIG. It can be done.

  Further, since truncation is performed in the arithmetic expressions in S801 and S811, the variable nBlock_x and the variable are not dependent on the reduction ratio (ox / ix) in the main scanning direction and the reduction ratio (oy / ii) in the sub-scanning direction. The value of nBlock_y can be represented by an integer. As a result, the calculation of the average pixel value by the reduced pixel output process (S708) described next with reference to FIG. 9 is performed by integer arithmetic, so that the processing speed can be improved (processing time can be shortened). become.

  As a result of S801 and S811, the number of divided blocks in the main scanning direction (nBlock_x) and the number of divided blocks in the sub-scanning direction (nBlock_y) for one pixel of the output image Q are both high-speed average pixel processing (S602). ), The divided block adjacent to the already processed divided block is extracted as the starting point. As a result, the calculation of the average pixel value by the reduced pixel output process (S708) described next with reference to FIG. 9 uses all of the image information included in the original image P without omission. Image quality deterioration can be suppressed.

  Next, the reduced pixel output process (S708) executed in the above-described high-speed average pixel process (see FIG. 7) will be described with reference to the flowchart in FIG. FIG. 9 is a flowchart showing the reduced pixel output process (S708).

  As shown in FIG. 9, in the reduced image output processing (S708), first, the Plane value is set for each area divided as a result of the main scanning direction block processing (S706) and the sub scanning direction block processing (S705). The partial pixel value of the corresponding color plane is calculated (S901). Note that the “partial pixel value” in the present embodiment means a value obtained by the sum of (the pixel value of the color plane corresponding to the plane value × the number of divided blocks included in the pixel) in each region. .

  In S901, for example, the partial pixel value of the area A1 is (the pixel value of the color plane corresponding to the Plane value in the area A1 × the number of divided blocks included in the area A1), that is, the coordinates of the original image P (nResidue_x0, nResidue_y0). ) Of the color plane corresponding to the Plane value in the pixel of (× Residue_x0 × nResidue_y0). At this time, if the value of nResidue_x0 is 0, the partial pixel value of the area A1 is calculated as 0.

  After the processing of S901, the partial pixel values calculated for each region are summed (S902), and the obtained total value is divided by the total number of divided blocks (nBlock_x × nBlock_y) included in the pixel being processed. A pixel value is calculated (S903).

After the processing of S903, the calculated average pixel value is stored in the output image memory 13b as the pixel value of the pixel at the coordinates (i _out , j _out ) in the output image Q (S904), and this reduced pixel output processing (S708) is performed. finish.

That is, each time this reduced pixel output process (S708) is executed once, the pixel value of one pixel at the coordinates (i _out , j _out ) in the output image Q is stored in the output image memory 13b. At this time, the pixel value of one pixel of the output image Q stored in the output image memory 13b is an average pixel value of the area of the original image P corresponding to the pixel.

  Next, referring to the flowchart of FIG. 10, the next initial value acquisition process in the main scanning direction (S714, S717) and the next initial value acquisition process in the sub-scanning direction (see FIG. 7) performed in the above-described high-speed average pixel processing (see FIG. 7). S718) will be described. FIG. 10A is a flowchart showing the next initial value acquisition process in the main scanning direction (S714, S717), and FIG. 10B is a flowchart showing the next initial value acquisition process in the sub-scanning direction (S718).

As shown in FIG. 10A, in the next initial value acquisition process (S714, S717) in the main scanning direction, first, it is confirmed whether i _out = 0 (S1001), and if i _out = 0 (S1001: Yes). ), The values of the variables nReadBlock_x, nResidue_x0, and nPosRes_x0 are all set to 0 (S1002), and the next initial value acquisition process in the main scanning direction (S714, S717) ends.

On the other hand, if the value of i _out is not 0 (S1001: No) as a result of the confirmation in S1001, the value of the variable nResidue_x0 is obtained from the expression nResidue_x0 = nDev-nResidue_x1 (S1003).

  After the process of S1003, the value of the variable nPosRes_x0 is obtained from the expression nPosRes_x0 = nPosRes_x1 (S1004), and the value of the variable nReadBlock_x is obtained from the expression nReadBlock_x = nReadBlock_x + nBlock_x (S1005, the next main acquisition process S17). ) Ends.

  As shown in FIG. 10B, in the next initial value acquisition process in the sub-scanning direction (S718), first, the value of the variable nResidue_y0 is obtained from the expression nResidue_y0 = nDev-nResidue_y1 (S1011).

  After the processing of S1003, the value of the variable nPosRes_y0 is obtained from the expression nPosRes_y0 = nPosRes_y1 (S1012), and the value of the variable nReadBlock_y is obtained from the expression nReadBlock_y = nReadBlock_y + nBlock_y (S1013). finish.

  As a result of these next initial value acquisition processing in the main scanning direction (S714, S717) and next initial value acquisition processing in the sub scanning direction (S718), next time in the main scanning direction block processing (S706) and sub scanning direction block processing (S705). The values of the variables nReadBlock_x, nResidue_x0, nPosRes_x0, nReadBlock_y, nResidue_y0, and nPosRes_y0 that are necessary for executing the above are obtained.

  In particular, by obtaining the values of the variable nPosRes_x0 and the variable nPosRes_y0, it is possible to easily specify the pixel positions (pixel coordinates) of the original image P corresponding to the regions A1 to A9 in the pixels of the output image Q to be processed next time. . As a result, the average pixel can be easily calculated by the reduced pixel output process (S708, see FIG. 9) even for the area of the original image P including the area (A1 to A4, A6 to A9) that is less than one pixel. Can be done.

  Next, with reference to FIG. 11, the superiority in the case of using the image reduction processing (FIG. 5) of the present embodiment including the above-described high-speed average pixel processing (S502) will be demonstrated. In the following description, a reduction method using image reduction processing according to the present embodiment including high-speed average pixel processing (S502) is referred to as “high-speed average pixel method”. FIG. 11 is a comparison diagram between the processing time when the high-speed average pixel method of the present embodiment is used and the processing time when other reduction methods are used.

  In FIG. 11, the original image P (natural image) of 8 million pixels (2448 × 3264) is reduced to an output image Q of various image sizes using various reduction methods such as the high-speed average pixel method of the present embodiment. The results of the processing time are listed.

  Note that the specifications used when performing these reduction processes (including the image reduction process (FIG. 5) of the present embodiment) are Pentium 4 3.4GH (Pentium) as a CPU (corresponding to the CPU 11 of the present embodiment). Is a registered trademark), and a 2 GB RAM (corresponding to the RAM 13 of this embodiment) is used as a memory.

  Also, as shown in FIG. 11, the image size of the output pixel Q is 2 million pixels (1200 × 1600), 3 million pixels (1536 × 2048), 4 million pixels (1728 × 2304), 5 million pixels (1920 × 2560), 6 million pixels (2112 × 2816), and 7 million pixels (2304 × 3072).

  As shown in FIG. 11, when the high-speed average pixel method (number of divisions nDev = 128) of the present embodiment is used, 2 million pixels, 3 million pixels, 4 million pixels, 5 million pixels, 6 million pixels, and Processing times for obtaining an output image Q of 7 million pixels were 1391 ms, 1719 ms, 1938 ms, 2156 ms, 2375 ms, and 2672 ms, respectively.

  On the other hand, when the conventional average pixel method using the least common multiple (“conventional average pixel method (lowest common multiple)” in FIG. 11) is used, as shown in FIG. The output image Q was obtained with a processing time of 1046 ms, but the output image Q with 3 to 7 million pixels did not finish (cannot be processed).

  Note that the processing time for obtaining an output image Q of 2 million pixels in the conventional average pixel method (the least common multiple) is shorter than the processing time by the high-speed average pixel method of the present embodiment. This is because the image size (1224 × 1632) of the output image Q is exactly ½ of the image size (2448 × 3264) of the original image P, and the value of the least common multiple is the image size of the original image P itself.

  As described above, even if the conventional average pixel method (the least common multiple) is used, if the value of the least common multiple is determined to be relatively small, the processing may be faster than the high-speed average pixel method of the present embodiment. However, in many cases, since the value of the least common multiple becomes considerably large, the processing speed is considerably slow even if processing is impossible or processing is not possible.

  Therefore, the results shown in FIG. 11 show that the high-speed average pixel method of the present embodiment solves the slow processing speed that has been a problem of the conventional average pixel method (the least common multiple), and can reduce high-quality reduced images at high speed. It can be obtained with

  On the other hand, when the conventional average pixel method “conventional average pixel method (floating point number)” described with reference to FIG. 13 is used, 2 million pixels, 3 million pixels, 4 million pixels, 5 million pixels Processing times for obtaining an output image Q of pixels, 6 million pixels, and 7 million pixels were 1422 ms, 1765 ms, 2031 ms, 2390 ms, 2641 ms, and 2922 ms, respectively.

  Here, as is apparent from FIG. 11, the processing time (processing speed) in the case of using the conventional average pixel method (floating point number) and the above-described high-speed average pixel method of this embodiment is the near-less tray method. Slower than bilinear method. However, considering that the output image Q (reduced image) obtained by the average pixel method is a high-quality image that cannot be obtained by the near-less tray method or the bilinear method, the high-speed average pixel method of the present embodiment or It can be said that the processing time by the conventional average pixel method (floating point number) is comparable to the processing time by the near-less tray method or the bilinear method.

  As shown in FIG. 11, the value of the processing time according to the conventional average pixel method (floating point) is equal to the processing time according to the high-speed average pixel method of the present embodiment, regardless of the output image Q of any image size. It was bigger than the value of. This result shows that the processing speed is improved by using the high-speed average pixel method of this embodiment as compared with the case of using the “conventional average pixel method (floating point number)”.

  Such an improvement in processing speed increases as the image size of the output image Q increases. When obtaining output images Q of 2 million pixels and 3 million pixels, the processing speed is about 2-3%. When the output image of 4 million pixels is obtained, the processing speed is improved by about 5%, and when the output image Q of 5 million pixels, 6 million pixels, and 7 million pixels is obtained, about 10%. , Processing speed improved.

  Therefore, the results shown in FIG. 11 indicate that the reduction by the high-speed average pixel method of the present embodiment is executed at a significantly faster processing speed than the reduction by the conventional average pixel method (floating point number). .

  The processing time was also measured when the value of the division number nDev was increased from 128 to 256 and the fast average pixel method of this embodiment was used. The measurement was performed only when the output image Q of 2 million pixels and 4 million pixels was obtained, but the obtained result was about 1 to 2% compared to the case of the division number nDev = 128, but decreased.

  As described above, the larger the value of the division number nDev, the higher the quality of the output image Q can be obtained, and as a result, the high-speed average pixel method of this embodiment sets the value of the division number nDev from 128 to 256. This indicates that a higher-quality output image Q can be obtained at a higher processing speed.

  In this embodiment, the processing time is measured with a machine specification in which the CPU is Pentium 4 3.4 GH and the memory is a 2 GB RAM. However, even if the memory size of the RAM is the same, the cache size The processing time (processing speed) changes by using different CPUs.

  As described above, the printer 1 according to the present embodiment executes the image reduction program 12a, and outputs the output image Q (reduced image) in units of divided blocks generated by dividing the original image P into nDev. ) Is calculated, an average pixel value can be obtained by integer calculation. In other words, the floating-point arithmetic performed on the decimal number generated according to the reduction ratio can be omitted, so that the processing speed of the average pixel method is remarkably improved as compared with the conventional method, and a high-quality reduced image is obtained at a high speed. It can be done.

  Further, the image reduction program 12a executed by the printer 1 of the present embodiment calculates the average pixel value by integer arithmetic as described above, and is therefore suitable for implementation in an embedded system.

  In addition, the printer 1 of the present embodiment can obtain the output image Q by executing the image reduction program 12a without increasing the image size of the original image P to the least common multiple of the output image size. Therefore, the processing speed of the average pixel method is remarkably improved as compared with the prior art, and a high-quality reduced image can be obtained at a high speed.

  Furthermore, since the image reduction program 12a executed by the printer 1 of the present embodiment determines the value of the division number nDev to be a numerical value within the range that can be processed by the CPU 11 as the arithmetic processing unit, the CPU 11 overflows during reduction. Can be reliably prevented.

  Further, the image reduction program 12a executed by the printer 1 according to the present embodiment determines the value of the division number nDev as the maximum value within the range that can be processed by the CPU 11, so that the original image P is divided. The size of the divided block generated by dividing into nDev is the finest in the range that can be processed by the arithmetic processing unit. That is, in such a case, since the resolution of the size of each divided block is the highest, an error that can be caused by using the divided block unit is surely reduced. As a result, a reduced image (output image) with the highest image quality can be obtained within a range that can be processed by the CPU 11.

  Note that the processing of S701 corresponds to the reduction rate acquisition step according to claim 1, the processing of S702 corresponds to the division number acquisition step according to claim 1, and the processing block number calculation step according to claim 1 Are the processes of S801 and S811, and the average pixel value calculation step according to claim 1 is the process of S708.

  Further, the main scanning direction division step according to claim 4 corresponds to the setting of the value of the variable nResidue_x0 in S703, the setting of the value of the variable nResidue_x0 in S1002, and the processing of S812, S813, and S1003.

  Further, the main scanning direction pixel position calculation step described in claim 5 corresponds to the setting of the value of the variable nPosRes_x0 in S703, the setting of the value of nPosRes_x0 in S1002, and the processing of S1004.

  Further, the sub-scanning direction division step described in claim 6 corresponds to the setting of the value of the variable nResidue_y0 in S703 and the processing of S802, S803, and S1011.

  Further, the sub-scanning direction pixel position calculating step according to claim 7 corresponds to the setting of the value of the variable nPosRes_y0 in S703 and the processing in S1012.

  The input image size acquisition step according to claim 8 corresponds to acquiring the value of the image size (ix, iy) of the original image P stored in the original image memory 13a in S701. As the output image size input step, when the output image size (ox, oy) is input by the operation of the operation panel 16 by the user, the value is read and output (stored) in the output size memory 13c. Is applicable.

  The reduction means according to claim 10 corresponds to image reduction processing (FIG. 6), and the reduction rate acquisition means according to claim 10 corresponds to the processing of S701, and the division number acquisition according to claim 10 is performed. As the means, the process of S702 corresponds, the processing block number calculation means according to claim 10 corresponds to the processes of S801 and S811, and the average pixel value calculation means of claim 10 corresponds to the process of S708. Applicable.

  The main scanning direction classifying means according to claim 13 corresponds to the setting of the value of variable nResidue_x0 in S703, the setting of the value of variable nResidue_x0 in S1002, and the processing of S812, S813, and S1003.

  Further, the sub-scanning direction sorting means described in claim 14 corresponds to the setting of the value of the variable nResidue_y0 in S703 and the processing in S802, S803, and S1011.

  The input image size acquisition means described in claim 15 corresponds to acquiring the value of the image size (ix, iy) of the original image P stored in the original image memory 13a in S701. As the output image size input means, when the output image size (ox, oy) is input by the operation of the operation panel 16 by the user, the value is read and output (stored) in the output size memory 13c. Is applicable.

  The present invention has been described above based on the embodiments. However, the present invention is not limited to the above-described embodiments, and various modifications and changes can be easily made without departing from the spirit of the present invention. Can be inferred.

  For example, in the above embodiment, the image reduction program 12a is installed in the printer 1 and the original image P input to the printer 1 is reduced to obtain the output image Q. However, the image reduction program 12a is installed in the PC 2. The CPU 2 of the PC 2 may execute the image reduction program 12a to obtain an output image Q that is a reduced image of the original image P.

  In the above embodiment, the image reduction program 12a is used to perform software reduction. However, each processing performed by the image reduction program 12a is calculated by the hardware configuration of the printer 1 or the PC 2. May be.

  In the above embodiment, the reduction ratio is obtained from the size of the original image P and the output image size input by the user. However, the value input by the user is not the output image size but the reduction ratio. There may be.

  In the above embodiment, the reduction ratio is obtained from the output image size input by the user. However, a value automatically set by the printer 1 may be used. For example, a configuration in which a reduction ratio is appropriately set according to the size of the print medium loaded in the printer 1 may be used.

  In the above-described embodiment, the end points of the divided blocks in the main scanning direction and the sub-scanning direction in one pixel of the output image Q are obtained by rounding down the decimal part with a Gaussian symbol. There may be.

1 is a block diagram illustrating an electrical configuration of a printer according to an embodiment of the present invention. It is a functional block diagram which shows the reduction function by the printer in one Embodiment of this invention. It is a functional block diagram which shows the function of a division number determination part. It is a schematic diagram for demonstrating the variable used in a high-speed average pixel process. It is a schematic diagram for demonstrating the variable used in a high-speed average pixel process. 6 is a flowchart illustrating a reduction process executed by a printer. It is a flowchart which shows the high-speed average pixel process performed in the reduction process of FIG. (A) is a flowchart showing a sub-scanning direction block process executed in the high-speed average pixel process of FIG. 7, and (b) is a main scanning direction executed in the high-speed average pixel process of FIG. It is a flowchart which shows a block process. It is a flowchart which shows the reduction pixel output process performed in the high-speed average pixel process of FIG. (A) is a flowchart showing the next initial value acquisition process in the main scanning direction executed in the high-speed average pixel process of FIG. 7, and (b) is executed in the high-speed average pixel process of FIG. It is a flowchart which shows a sub-scanning direction next initial value acquisition process. FIG. 11 is a comparison diagram between the processing time when the high-speed average pixel method of the present embodiment is used and the processing time when other reduction methods are used. It is a schematic diagram for demonstrating the conventional average pixel method. It is a flowchart which shows the conventional average pixel process.

Explanation of symbols

1 Printer (image processing device)
11 CPU (arithmetic unit)
12a Image reduction program (image processing program)
102 Reduction rate calculation unit (reduction rate acquisition step, reduction rate acquisition means)
103 processing block calculating means (processing block number calculating step, processing block number calculating means)
104 Average pixel value calculating means (average pixel value calculating step, main scanning direction dividing step, sub-scanning direction dividing step, average pixel value calculating means, main scanning direction dividing means, sub-scanning direction dividing means)
105 Division number determination means (division number determination step, division number determination means)
105a Maximum block number calculation unit (division number determination step, division number determination means)
105b Division number calculation unit (division number determination step, division number determination means)

Claims (16)

In an image processing program executed by an arithmetic unit to reduce a substantially rectangular original image having a number of pixels (ix × iy) at a predetermined reduction ratio and obtain an output image having a substantially rectangular shape having a number of pixels (ox × oy). ,
A reduction rate acquisition step of acquiring the reduction rate;
A division number acquisition step of acquiring a division number nDev for dividing each pixel constituting the original image into (nDev × nDev) (nDev is an integer of 2 or more) divided blocks;
A processing block number calculating step for calculating, for each pixel of the output image, an integer number of the divided blocks corresponding to one pixel of the output image based on the reduction rate acquired by the reduction rate acquisition step; ,
An average pixel value calculating step for calculating an average pixel value of the area of the original image corresponding to the number of divided blocks obtained by the processing block number calculating step for each pixel of the output image. A characteristic image processing program.   The processing block number calculation step treats the number of divided blocks in the main scanning direction or sub-scanning direction based on the reduction ratio as an integer number of divided blocks by rounding down the decimal part. The image processing program according to claim 1, wherein:   In the processing block number calculating step, when calculating the number of the divided blocks corresponding to one pixel in the output image, the divided blocks adjacent to the divided blocks already used in the average pixel value calculating step are calculated. The image processing program according to claim 1, wherein the number of the divided blocks corresponding to one pixel of the output image is calculated as a starting point. Before executing the average pixel value calculating step, for each pixel of the output image, the main scanning direction is defined as (1a) x1 corresponding to the unprocessed portion in the pixel that leaves the unprocessed portion in the original image ( x1 is an area composed of the divided blocks of 0 or more and less than nDev), and (1b) N pixels (N is an integer of 1 or more) adjacent to the area (1a) in the original image ) A region arranged side by side, and (1c) in the original image, a region that is adjacent to the region (1b) and includes a remaining portion of x2 (x2 is an integer of 0 or more and less than nDev) that is less than one pixel. A main scanning direction dividing step for dividing,
The average pixel value calculating step calculates a sum of pixel values for each area divided by the main scanning direction dividing step, and calculates the sum of the pixel values of the output image obtained by the processing block number calculating step. 4. The image processing program according to claim 1, wherein the image processing program is obtained by dividing by the number of the divided blocks corresponding to one pixel.   The image processing program according to claim 4, further comprising a main scanning direction pixel position calculating step of calculating a pixel position in the main scanning direction of a pixel including the region (1 a). Before the execution of the average pixel value calculating step, for each pixel of the output image, the sub-scanning direction is expressed as (2a) y1 (corresponding to the unprocessed portion in the pixel that leaves the unprocessed portion in the original image) y1 is an area composed of the divided blocks of 0 or more and less than nDev), and (2b) M pixels (M is an integer of 1 or more) adjacent to the area (2a) in the original image. ) A region arranged side by side, and (2c) in the original image, a region that is adjacent to the region (2b) and is composed of y2 remaining portions (y2 is an integer not less than 0 and less than nDev) that is less than one pixel. A sub-scanning direction dividing step for dividing,
The average pixel value calculating step is performed for each region divided by the sub-scanning direction dividing step or for each region generated as a result of the division by the sub-scanning direction dividing step and the main scanning direction dividing step. The sum of the values is obtained, and the sum of the pixel values is obtained by dividing by the number of the divided blocks corresponding to one pixel of the output image obtained by the processing block number calculation step. 6. The image processing program according to any one of 5.   The image processing program according to claim 6, further comprising a sub-scanning direction pixel position calculating step of calculating a pixel position in a sub-scanning direction of a pixel including the region (2a). An input image size acquisition step of acquiring a value of the number of pixels of the input pixels (ix × iy);
An output image size input step of inputting a value of the number of pixels of the output image (ox × oy),
The reduction ratio acquisition step includes the value of the number of pixels of the input image (ix × ii) acquired by the input image size acquisition step and the number of pixels of the output image input by the output image size input step (ox × The image processing program according to any one of claims 1 to 7, wherein a reduction ratio is acquired from a value of (oy). A reduction rate input step of inputting a reduction rate for obtaining the output image having the number of pixels (ox × oy) from the number of pixels of the input pixels (ix × ii);
9. The image processing program according to claim 1, wherein the reduction rate acquisition step acquires the reduction rate input in the reduction rate input step. An input means for inputting a substantially rectangular original image having the number of pixels (ix × iy), and a substantially rectangular shape having the number of pixels (ox × oy) by reducing the original image input from the input means at a predetermined reduction ratio. In an image processing apparatus provided with a reduction means for obtaining an output image of
The reduction means includes
Reduction rate acquisition means for acquiring the reduction rate;
Division number acquisition means for acquiring a division number nDev for dividing each pixel constituting the original image into (nDev × nDev) (nDev is an integer of 2 or more) divided blocks;
Processing block number calculating means for calculating, for each pixel of the output image, an integer number of the divided blocks corresponding to one pixel of the output image based on the reduction rate acquired by the reduction rate acquisition means; ,
Average pixel value calculating means for calculating an average pixel value of the area of the original image corresponding to the number of divided blocks obtained by the processing block number calculating means for each pixel of the output image. An image processing apparatus.   When the number of divided blocks in the main scanning direction or the sub-scanning direction based on the reduction ratio is represented by a decimal number, the processing block number calculation means treats the number of divided blocks as an integer number by rounding down the decimal part. The image processing apparatus according to claim 10.   The processing block number calculating means calculates the divided blocks adjacent to the divided blocks already used in the average pixel value calculating means when calculating the number of the divided blocks corresponding to one pixel in the output image. 12. The image processing apparatus according to claim 10, wherein the number of the divided blocks corresponding to one pixel of the output image is calculated as a starting point. Before the average pixel value calculation means, for each pixel of the output image, the main scanning direction is (1a) x1 (x1) corresponding to the unprocessed portion in the pixel that leaves the unprocessed portion in the original image. Is an area composed of the divided blocks of 0 or more and less than nDev), and (1b) N complete pixels adjacent to the area (1a) in the original image (N is an integer of 1 or more) A region that is arranged side by side and (1c) in the original image, is divided into a region that is adjacent to the region (1b) and that includes x2 remaining portions that are less than one pixel (x2 is an integer that is greater than or equal to 0 and less than nDev). Main scanning direction sectioning means
The average pixel value calculation means obtains a sum of pixel values for each area divided by the main scanning direction division means, and the sum of the pixel values is calculated for the output image obtained by the processing block number calculation means. The image processing apparatus according to claim 10, wherein the image processing apparatus is obtained by dividing by the number of the divided blocks corresponding to one pixel. Before the average pixel value calculation means, for each pixel of the output image, (2a) y1 (y1) corresponding to the unprocessed portion in the pixel that leaves the unprocessed portion in the original image Is an area composed of the divided blocks of 0 or more and less than nDev), and (2b) N complete pixels adjacent to the area (2a) in the original image (N is an integer of 1 or more) A region arranged side by side and (2c) in the original image, a region adjacent to the region (2b) and composed of y2 remaining portions that are less than one pixel (y2 is an integer not less than 0 and less than nDev). Sub-scanning direction dividing means for
The average pixel value calculation means includes a pixel for each area divided by the sub-scanning direction dividing means or for each area generated as a result of the division by the sub-scanning direction dividing means and the main scanning direction dividing means. The total sum of the values is obtained, and the sum of the pixel values is obtained by dividing by the number of the divided blocks corresponding to one pixel of the output image obtained by the processing block number calculating means. The image processing apparatus according to any one of claims 13 to 13. Input image size acquisition means for acquiring the value of the number of input pixels (ix × ii) input from the input means;
Output image size input means for inputting a value of the number of pixels of the output image (ox × oy),
The reduction ratio acquisition unit includes a value of the number of pixels of the input image (ix × ii) acquired by the input image size acquisition unit and a number of pixels of the output image input by the output image size input unit (ox × The image processing apparatus according to claim 10, wherein a reduction ratio is acquired from a value of (oy). Reduction rate input means for inputting a reduction rate for obtaining the output image having the number of pixels (ox × oy) from the number of pixels of the input pixels (ix × ii);
The image processing apparatus according to claim 10, wherein the reduction rate acquisition unit acquires the reduction rate input by the reduction rate input unit.

Images