JP4504861B2 - Image processing method and image processing apparatus - Google Patents

Description

  The present invention relates to an image processing method and an image processing apparatus that perform image data rotation processing.

  When performing rotation processing on image data, the original image data arranged in the raster scan order is read obliquely and stored in a storage device as continuous data, thereby generating post-rotation image data There is. Further, the original image data arranged in the raster scan order may be read out in the raster scan order and stored so as to be in an oblique direction when stored in the storage device, thereby generating post-rotation image data.

  Furthermore, rotated image data may be generated by performing coordinate transformation (for example, affine transformation) on all the pixels in the image area to obtain the rotated image position.

  In the above-described conventional procedure, the address for reading out the image data to be rotated and the address for writing in the memory to be written is obtained by calculation, and the address is directly accessed according to these addresses. A special circuit is required. Furthermore, the above-described conventional procedure requires complicated operations such as affine transformation.

Note that the image interpolating apparatus disclosed in Patent Document 1 includes a means for calculating a pair of interpolation coefficients and a page memory for determining a new image data position and value two-dimensionally in rotation image processing. Interpolation is performed with a means for retrieving image data, a new image data value is calculated, stored in a new page memory, and output. The image processing apparatus disclosed in Patent Document 2 includes a unit that encodes each block by dividing m-line image data read in line units into blocks for every n pixels in the main scanning direction, and (m pixels) × The image data is rotated by means for rotating the pixel data of (n pixels) in units of 90 °.
Japanese Patent Laid-Open No. 7-220061 JP-A-10-93805

  An object of the present invention is to perform image data rotation processing only by using a line data holding memory and interpolation processing without performing complicated calculations.

The present invention has been made to achieve the above object. The image processing method according to claim 1 according to the present invention includes:
In image processing that captures line-scanned image data and rotates the image,
The position where the line connecting the sampling points before rotation in the sub-scanning direction intersects the line in the main scanning direction after rotation is obtained from the pixel data in the sub-scanning direction before rotation by the first interpolation operation, and sub-scanning Place the interpolated image data,
The position where the line in the main scanning direction after rotation and the line connecting the rotated sampling point in the sub scanning direction intersect is determined by the second interpolation calculation from the sub scanning interpolation image data in the main scanning direction after rotation. Find and place the rotated image data,
Calculating a plurality of intersecting positions in the first interpolation calculation by adding a constant difference in the sub-scanning direction before rotation with reference to the main scanning direction axis before rotation;
An image processing method characterized in that a plurality of intersecting positions in the second interpolation calculation are calculated by adding a constant difference in the main scanning direction before rotation with reference to the sub-scanning direction axis before rotation. .

According to the second aspect of the present invention, there is provided an image processing method.
In image processing that captures line-scanned image data and rotates the image,
The position where the line connecting the sampling points before rotation in the sub-scanning direction intersects the line in the main scanning direction after rotation is obtained from the pixel data in the sub-scanning direction before rotation by the first interpolation operation, and sub-scanning Place the interpolated image data,
The position where the line in the main scanning direction after rotation and the line connecting the rotated sampling point in the sub scanning direction intersect is determined by the second interpolation calculation from the sub scanning interpolation image data in the main scanning direction after rotation. Find and place the rotated image data,
(1) Hold all the line data necessary for processing one line data in the memory and perform the rotation process,
(2) Next, by inputting only additionally necessary line data and additionally holding it in the memory, all the line data necessary for processing the next one line data is held in the memory,
(3) Rotate the next one line data in (2) above,
(4) By repeating the above (2) and (3) and rotating the entire image data,
In this image processing method, only line data used in the first interpolation calculation and the second interpolation calculation is used .

An image processing apparatus according to claim 3 of the present invention is provided.
An image processing apparatus using the image processing method according to claim 1,
A processor having a plurality of arithmetic registers and performing parallel processing;
A plurality of addition processes in the first interpolation calculation or the second calculation process are simultaneously performed by a processor that performs the parallel processing .

  By using the present invention, image data rotation processing can be performed only by using the line data holding memory and interpolation processing without performing complicated operations such as affine transformation. Further, the rotation processing at a minute angle can be performed with a small amount of memory without using a large-capacity memory equivalent to a page memory. In addition, the processing time related to the time for reading the image data into the memory can be improved. Furthermore, rotation processing can be performed using a parallel processor.

  DESCRIPTION OF EXEMPLARY EMBODIMENTS Hereinafter, preferred embodiments of the invention will be described with reference to the drawings.

[First Embodiment]
As shown in FIG. 7, the image processing apparatus (30) according to the present invention receives image data (42), temporarily holds necessary data in a holding memory (36), and is an internal arithmetic processing apparatus. It is a device that performs rotation processing and outputs rotated image data (44).

  1 to 4 are schematic views of partial image data for explaining the image processing method according to the first embodiment performed by the image processing apparatus (30) according to the present invention. First, FIG. 1 is a schematic diagram showing a part of input data of image data before rotation processing.

  In the data shown in FIG. 1, the main scanning direction (2) is taken in the horizontal direction (from left to right), and the sub-scanning direction (4) is taken in the vertical direction (from top to bottom). It is assumed that the image data (6) which is a pixel is arranged in a grid pattern. In the figure, the x-coordinate represents the position in the main scanning direction, and the y-coordinate represents the position in the sub-scanning direction.

  That is, in the image data arranged in a grid pattern, the position at the upper left corner of the figure is represented as (x, y), and the position at the right side is (x + 1, y), (x + 2, y), (x + 3, y). ), (X + 4, y)... Similarly, the position of the lower part of the image data at the upper left corner of the figure is represented as (x, y + 1), (x, y + 2), (x, y + 3).

  When inputting image data (6), line scanning is performed in the order of S (x, y), S (x + 1, y), S (x + 2, y)... In the main scanning direction (2) to form one line. . Subsequently, the input is performed in the sub-scanning direction (4) for each line.

  After data input to the image processing device (30), for the image data (6) shown in FIG. 1, a line (10) connecting sampling points before rotation in the sub-scanning direction and a main scanning direction (8) after rotation Is obtained, and sub-scan interpolation image data (12) T (x, y) is arranged. FIG. 2 shows an example of the arrangement of the sub-scan interpolation image data (12). Here, the arrangement of the sub-scanning interpolated image data (12) can be obtained by performing interpolation calculation by a known bicubic method using the four image data in the vicinity of the sub-scanning image data of the target image data.

  The above interpolation can also be performed by a well-known bilinear method using two neighboring image data. The bicubic method may be used when accuracy is required in image processing, and the bilinear method may be used when it is desired to reduce the number of input pixels of the original image data.

  After calculating the sub-scanning interpolated image data (12) T (x, y) after rotation, the main scanning direction (8) after rotation intersects the line (16) connecting the sampling points after rotation in the sub-scanning direction. The position of the main scanning image data (18) R (x, y) after rotation is arranged. FIG. 3 shows an example of the arrangement of the main scan image data (18) after rotation. Here, the arrangement of the main-scanning image data (18) R (x, y) after rotation is well-known using the four image data in the main-scanning direction (8) after the rotation of the sub-scanning interpolation image data (12). It can be obtained by the enlargement process of the bicubic method. This enlargement process can also be performed by the well-known bilinear enlargement process using the neighboring two-pixel data.

  After calculating the rotated main scanning image data (23) R (x, y) and outputting it as line data in accordance with the rotated main scanning direction (8), the rotated image data can be formed. In the image processing method according to the first embodiment, the interpolation in the sub-scanning direction after the rotation is performed, and then the interpolation in the main scanning direction after the rotation is performed. Output can be realized.

In the image processing method according to the first embodiment, when rotation processing is performed on image data in one line in the main scanning direction, the number of lines in the sub-scanning direction necessary for output is expressed by the following equation (Expression 1). (See FIG. 4).
[Equation 1]
(Number of main scanning images) × sin (rotation angle)
For example, if the rotation angle is 1 degree and the number of pixels in the main scanning direction is 1024,
(1024) × sin ((1/180) π) = 18
Therefore, if a memory for data for 18 lines in the sub-scanning direction is secured, the rotation processing by the above method can be realized.

[Second Embodiment]
As shown in FIG. 7, the image processing apparatus (30) according to the second embodiment of the present invention also receives the image data (42), temporarily holds the necessary data in the holding memory (36), This is a device that performs rotation processing with an internal processing unit and outputs the rotated image data (44). The image processing apparatus and the image processing method according to the second embodiment are substantially the same as the image processing apparatus and the image processing method according to the first embodiment. Therefore, the same parts are denoted by the same reference numerals and description thereof is omitted.

  5 and 6 are schematic views of partial image data for explaining an image processing method according to the second embodiment performed by the image processing apparatus (30) according to the present invention. In the image processing method according to the second embodiment, as in the image processing method according to the first embodiment, after data is input to the image processing device (30), the image data (6) (shown in FIG. 1) is related. The position where the line (10) connecting the sampling points before rotation in the sub-scanning direction intersects the main scanning direction (8) after rotation is obtained, and sub-scanning interpolated image data (12) T (x, y) is obtained. Arrange (see FIG. 5).

At this time, the position where the line (10) connecting the sampling points before the rotation in the sub-scanning direction and the main scanning direction (8) after the rotation intersects from the rotation angle and the image data position.
(Image data position in the main scanning direction) × sin (rotation angle)
Can be calculated as However, if the process of calculating the image data position in the main scanning direction is performed for each image data before rotation, the entire processing time becomes enormous.

  Therefore, paying attention to the fact that the relative positional relationship between the sub-scanning interpolated image data (12) adjacent in the post-rotation main scanning direction (8) is all the same, the image processing method according to the second embodiment uses the post-rotation main scanning direction. The positions of two adjacent pixels in the scanning direction (8) are obtained, and then the difference (image data position difference (22)) of the two pixels in the pre-rotation sub-scanning direction (4) is obtained. Further, as shown in FIG. 5, the arrangement of the remaining sub-scanning interpolated image data (12) is calculated by sequentially adding the image data position difference (22) to the two pixels serving as a reference. That is, a plurality of intersecting positions in the interpolation calculation are calculated by adding a certain difference in the sub-scanning direction before rotation with reference to the main scanning direction axis before rotation.

  As in FIG. 5, in FIG. 6 as well, the main scan direction after rotation is arranged at a position where the post-rotation main scanning direction (8) and the line (16) connecting the sampling points after rotation in the sub-scanning direction intersect. Regarding the image data (18), the relative positional relationship between two adjacent pixels is the same. Accordingly, the positions of the two adjacent pixels are obtained, the difference (image data position difference (26)) in the pre-rotation main scanning direction (2) of the two pixels is obtained, and the image data for the two pixels serving as the reference is obtained. The position difference (26) is added in order, and the arrangement of the entire rotated main-scan image data (18) is calculated. That is, a plurality of intersecting positions in the interpolation calculation are calculated by adding a certain difference in the main scanning direction before rotation with reference to the sub-scanning direction axis before rotation.

  In the image processing method according to the second embodiment, a different calculation is not performed for each image data. That is, rotation processing can be realized by sequentially performing addition processing of a certain value. Therefore, the image processing method according to the second embodiment can be performed by a processor that includes a plurality of arithmetic registers and performs parallel processing. If a parallel processor is used, the processing speed of the entire rotation process can be improved.

[Third Embodiment]
FIG. 7 is a block diagram focusing on the flow of data for explaining the image processing method according to the third embodiment performed by the image processing apparatus (30) according to the present invention.

  According to the image processing method according to the first embodiment, for example, when processing the first line of the image data by the bicubic interpolation process, data for four pixels is necessary, and therefore, from the first line of the input data. It is necessary to load up to the fourth line into the line data holding memory (36).

  Accordingly, as shown in FIG. 7, when the line data 2 is processed and output from the output device (34), the line data 2 of the input data, the line data 3, the line data 4, It is necessary to load line data 5.

Subsequently, in order to process the line data 3 in the image processing apparatus (30) shown in FIG. 7, the line data 3, the line data 4, the line data 5, and the line data of the input data are stored in the line data holding memory (36). 6 needs to be loaded. At this time, since line data 3, line data 4, and line data 5 are already held in the line data holding memory (36), only the line data 6 of the pre-processing image data (42) is received from the input device (32). Is loaded into the line data holding memory (36). As a result, the processing speed can be improved by the time for loading the image data for three lines.

Therefore, an image processing apparatus that performs the image processing method according to the third embodiment
(1) All the line data necessary to process one line data is first held in the memory and rotated.
(2) Next, by inputting only additionally necessary line data and additionally holding it in the memory, all the line data necessary for processing the next one line data is held in the memory,
(3) Rotate one line data following (2) above,
(4) The entire image data is rotated by repeating the steps (2) and (3).

It is the schematic (1) of the partial image data explaining the image processing method which concerns on 1st Embodiment performed with the image processing apparatus which concerns on this invention. It is the schematic (2) of the partial image data explaining the image processing method which concerns on 1st Embodiment performed with the image processing apparatus which concerns on this invention. It is the schematic (3) of the partial image data explaining the image processing method which concerns on 1st Embodiment performed with the image processing apparatus which concerns on this invention. FIG. 6 is a schematic diagram showing the number of lines in the sub-scanning direction necessary for output when rotating image data in one main scanning direction by the image processing method according to the first embodiment. It is the schematic (1) of the partial image data explaining the image processing method which concerns on 2nd Embodiment performed with the image processing apparatus which concerns on this invention. It is the schematic (2) of the partial image data explaining the image processing method which concerns on 2nd Embodiment performed with the image processing apparatus which concerns on this invention. 1 is an overall block diagram of an image processing apparatus according to the present invention. In addition, it is a block diagram focusing on the flow of data for explaining the image processing method according to the third embodiment of the present invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 30 ... Image processing device, 32 ... Input device, 34 ... Output device, 36 ... Line data holding memory, 42 ... Image data before processing, 44 ... Image data after processing .

Claims (3)

In image processing that captures line-scanned image data and rotates the image,
The position where the line connecting the sampling points before rotation in the sub-scanning direction intersects the line in the main scanning direction after rotation is obtained from the pixel data in the sub-scanning direction before rotation by the first interpolation operation, and sub-scanning Place the interpolated image data,
The position where the line in the main scanning direction after rotation and the line connecting the rotated sampling point in the sub scanning direction intersect is determined by the second interpolation calculation from the sub scanning interpolation image data in the main scanning direction after rotation. Find and place the rotated image data,
Calculating a plurality of intersecting positions in the first interpolation calculation by adding a constant difference in the sub-scanning direction before rotation with reference to the main scanning direction axis before rotation;
An image processing method comprising: calculating a plurality of intersecting positions in the second interpolation calculation by adding a constant difference in the main scanning direction before rotation with reference to the sub-scanning direction axis before rotation . In image processing that captures line-scanned image data and rotates the image,
The position where the line connecting the sampling points before rotation in the sub-scanning direction intersects the line in the main scanning direction after rotation is obtained from the pixel data in the sub-scanning direction before rotation by the first interpolation operation, and sub-scanning Place the interpolated image data,
The position where the line in the main scanning direction after rotation and the line connecting the rotated sampling point in the sub scanning direction intersect is determined by the second interpolation calculation from the sub scanning interpolation image data in the main scanning direction after rotation. Find and place the rotated image data,
(1) Hold all the line data necessary for processing one line data in the memory and perform the rotation process,
(2) Next, by inputting only additionally necessary line data and additionally holding it in the memory, all the line data necessary for processing the next one line data is held in the memory,
(3) Rotate the next one line data in (2) above,
(4) By repeating the above (2) and (3) and rotating the entire image data,
An image processing method using only line data used in the first interpolation calculation and the second interpolation calculation . An image processing apparatus using the image processing method according to claim 1,
A processor having a plurality of arithmetic registers and performing parallel processing;
An image processing apparatus, wherein a plurality of addition processes in the first interpolation calculation or the second calculation process are simultaneously performed by a processor that performs the parallel processing.

Images