JP2646478B2 - Compression storage device and composite output device for divided reading of character and graphics - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an apparatus for dividing, reading, compressing, storing, and reproducing a large original picture including a character and a graphic by using an apparatus having a small memory capacity. Record of figures,
It is an object of the present invention to provide an apparatus that can perform reproduction with a memory having a small capacity, has excellent recording quality, and has excellent reproduction characters and graphics.

[0002]

2. Description of the Related Art There are already many devices for reading characters and figures by an image scanner, digitizing the characters and figures, and storing them in a memory. The screen of the image scanner is divided into vertical and horizontal pixels. Assuming that M pixels are arranged horizontally and N pixels vertically, MN pixels exist in total.
Pixels are sometimes simply called dots. Assuming that there are n bits of data for one pixel, the total amount of data is n
MN bits. The higher the division number MN of the screen is, the higher the image clarity is, of course. However, when the MN is large, a huge amount of data can be obtained only by the data amount nMN on one screen.
It becomes the amount of data. If this is stored as it is, a very large memory is required. In fact, a printing machine currently used uses an enormous amount of memory for storing image data.

On the other hand, if the number of divisions MN is reduced, the screen becomes coarse and the quality of the reproduced image deteriorates. An appropriate number of divisions is determined based on the memory capacity and the demand for image quality. Actual character figures are often continuous figures surrounded by continuous outlines. Therefore, the contour is often a set of straight lines, a set of straight lines and smooth curves, or a set of perfect circles. A figure whose gradient changes every dot is not handled. Therefore, it is attempted to read the original image with an image scanner, digitize the image, and compress the image instead of immediately storing it in a memory. However, at present, no other compression method has been found other than that of the present inventor, which can sufficiently compress data and has a high quality reproduced image.

The inventor's method of compressing character data and graphic data has been proposed in Japanese Patent Application Nos. 4-259137 and 4-269646. This is to read the original image, digitize it, extract the contour, calculate the curvature at each point of the contour, use the point with the large curvature as the joining point, arrange the joining point, and correct the position of the joining point Later, the parameters of the line connecting the adjacent junctions are determined. At this time, if the line connecting the joining points is a curve, the full-energy function created by the inventor is used. This is a major feature.
The coordinates of the joining point thus determined and the parameters of the line connecting the joining points are stored in the memory. This eliminates the need for data on pixels other than the contour line. Further, even on the contour line, data at points other than the joining point becomes unnecessary. Therefore, the amount of data to be stored is greatly reduced. The memory capacity may be small. It is a truly excellent invention.

[0005]

With a large computer having a large memory capacity, an original picture of A4 or B4 size can be read and compressed as it is and stored in the memory. In the case of an apparatus having a large memory capacity, the method of the present invention can be applied as it is. We would like to implement these inventions using an inexpensive personal computer. However, with a small memory capacity of a personal computer, only a small original image can be processed. All data of the input image must be stored even temporarily, because the required memory capacity is considerably large.

For example, when a personal computer is used, an apparatus for reading an image in a small square area of 256 dots.times.256 dots can be manufactured according to the above-mentioned invention. When the number of dots is limited, the size of the original image is also limited. For example, if an image scanner of 320 dpi is used, the size of 256 dots × 256 dots is approximately 2c.
The size is m square. This makes it possible to effectively use small originals of 2 cm square. There is also a benefit in storing and reproducing such a small original image. However, the readable area is extremely small, which limits its use. It is urgently desired to be able to process a larger area of the original image.

The use of a coarser dot image scanner increases the size of the reading area, but on the contrary, the reading data becomes coarser, which may be undesirable. If the original image of A4 size (30 cm × 21 cm) is read and processed with the same accuracy, 15 × 11 × 165
Requires twice the memory capacity. It is impossible for ordinary personal computers to handle such a huge amount of data as it is. SUMMARY OF THE INVENTION It is a first object of the present invention to provide an apparatus which solves such a difficulty and which can apply the above-mentioned invention to an original having a larger area while using a computer having a small memory capacity. It is a second object of the present invention to provide a device capable of simply describing a contour by appropriate function approximation and improving the quality of a reproduced image.

[0008]

According to the method of the present invention, a large original image is read by an image scanner, and the read image is divided into small image processing basic processing areas overlapping by one dot. Is treated as one figure, contour line extraction and joint point extraction are performed, and the line between the joint points is approximated by a function. The same processing is performed for all the divided images. The processing to be performed on the divided image is as described in the above-mentioned Japanese Patent Application No. 4-2.
This is the same as the method proposed in No. 59137. The joint points for all the divided images and the parameters of the line connecting the joint points are obtained. Then, these are stored in the memory. Reproduction is also performed for each divided image. Since the dots overlap by one dot, a continuous large image is reproduced.

[0009] Since the original image is divided vertically and horizontally by the basic processing area, an area where the original image does not exist in a part of the divided image occurs at the upper, lower, left and right edges. In this part, the pixel data is set to 0. In other words, zeros are embedded in the remaining portions that protrude from the original image. Thus, all the divided images have the size of the basic processing area. Some measures are taken so that curves and straight lines are smoothly connected between adjacent divided images. One dot overlapping between adjacent divided images is referred to as a boundary portion. The pixels at the border are particularly important. This portion has the same dot data in adjacent divided images. It undergoes different processing on different divided images.
The dot data at the boundary must not disappear during the approximation of the junction extraction function. Therefore, at the time of extraction of the junction, the black pixels at both ends are always used as the junction at the boundary.

Next, the outline of the method of the present invention will be described in advance. An arbitrary size character figure or the like is read by optical means. Data read by optical means by the first image storage device is stored. The image is divided into basic processing areas capable of image processing, and data is cut out for each divided image. At this time, one dot is overlapped along the boundary line. The data of the divided image is stored in the second image storage device. Contour extraction is performed on each divided image. This is the second
This is done by reading out data for each divided image from the image storage device. The outline is given as an outline point sequence. An outline point sequence is stored for each divided image. Extract boundary junctions. This is a joint point near the boundary between the two divided images. This is one of the features of the present invention. In order to smoothly connect the divided data when they are joined at the time of reproduction, the joining points at the boundary portions in contact with the surroundings are forcibly extracted. Since the extracted joint point must exist as a joint point in any case, it will be treated as a special joint point in the subsequent processing and will not be removed by a joint point removing mechanism or the like. Data approximation is performed on one divided image to compress the data. The compressed data is stored. Data compression is performed in the same manner as in the aforementioned Japanese Patent Application Nos. 4-259137 and 4-269646. A joint point is extracted from the contour line, and its coordinates are obtained. Remove unnecessary junction points.
It is determined whether the line connecting the remaining adjacent joint points is a straight line, a circular arc, or a free curve, and is stored together with parameters defining these lines. A full-energy function is used when connecting adjacent junctions with a free curve. The number of data is remarkably reduced, and the straight line becomes clear and the curve becomes smooth. This is called data compression because the number of data is reduced. This is a data compression operation performed for each divided image. When the compressed data of one divided image is stored in the storage device, the same is performed for the next divided image. Data compression is sequentially performed on all the divided images. In this way, all the divided images are represented by lines connecting the joining points. The reproduction of the image is performed by repeating the reproduction for each divided image. The reproduction of an image from compressed data is also described in Japanese Patent Application Nos. 4-25913 and 4-26964.
Perform in the same manner as in No. 6. At this time, a layout editor is used to lay the divided images together. The divided images are reproduced so as to overlap by one dot at the boundary.

[0011]

The method of the present invention will be described with reference to the drawings. FIG. 1 shows an operation of cutting out a divided image from the first image storage device, compressing the data, and storing the data in the second image storage device. The original figure is, for example, the size of a normal document having the size of A4 or A3. This is read optically by an image scanner. This is the first image storage device 1
Is stored. Since the original is large, the amount of data is also large. Since the amount of data is too large, the data cannot be compressed as it is with the capacity of a commercially available personal computer. Therefore, this is divided into basic processing areas where data can be compressed. This is, for example, a small area of 256 × 256. The image in the basic processing area is referred to as a divided image. In accordance with the procedure described below, the divided images are data-compressed one by one, and then stored in the second image storage device 1 '.

Since the divided image is sufficiently small, the data can be easily compressed by the method of the present invention using a commercially available personal computer. Data compression includes contour point sequence extraction, curvature calculation, curvature storage, perfect circle extraction, junction extraction, optimal junction extraction, junction removal, junction storage, derivation of lines connecting junctions, This is performed by storing parameters.
The entire original image is stored in the first image storage device 1.
It divides into the unit size from the upper left to the right. This is compressed one by one. Cut out the first divided image,
The data obtained by compressing the data is stored in a second image storage device 1 '.
(Compressed data storage device). This is shown in FIG. Subsequently, a second basic processing area is cut out to be a second divided image. The same data compression operation is performed on the second divided image. In other words, contour point sequence extraction, curvature calculation, curvature storage, perfect circle extraction, junction extraction, optimal junction extraction, junction removal, junction storage, derivation of the line connecting the junctions, parameters of this line The storage is performed on the second divided image, and the result is stored in the second image storage device 1 '(compressed data storage device). Figure 1
(2).

When the third divided image is cut out, it is entirely made of white pixels. There are no black pixels. That is, it is a blank portion. This eliminates the need for processing such as contour point sequence extraction. Immediately, the processing can be shifted to the next divided image clipping and compression processing. Hereinafter, data compression for each divided image is repeated in the same manner, and data compression is performed for all divided images. If the original image is divided into M 'horizontal and N' vertical divided images, there are a total of M'N 'divided images. In this case, the same data compression operation is repeated for the M'N 'divided images.

In the present invention, when a large original image is divided, the original image is divided such that one pixel at the boundary overlaps. FIG. 2 is a diagram showing that boundaries are overlapped at the time of division. The divided image is bordered on other sides by four sides, three sides or two sides. The region is divided so that one pixel, that is, one dot overlaps the vertical and horizontal boundaries. This is because the reproduced image correctly reproduces the original image. For example, the size of the basic area, that is, the size of the divided image is, for example, 256 dots × 256 dots. The size of the divided image is not limited to this, and may be set to a square such as 128, 512 dots, which is a power of two. However, it does not have to be a power of two. It is sufficient if the data can be easily compressed by an appropriate size computer. The reason why one dot overlap is required will be described. It is assumed that there is a pattern as shown in FIG. In the left divided image P, three black pixels are vertically arranged on the boundary. The same three black pixels are arranged in the right divided image Q. This is because the left and right divided images share one dot at the boundary line. The data is compressed and stored in the second image storage device. When this is read out and reproduced again, a lower reproduced image P + Q is obtained. Since the three black pixels on the boundary line are shared, the two divided images can be smoothly connected at the boundary.

If the boundary line is not shared, a gap or deformation occurs in the reproduced image. This will be described with reference to FIG. In FIG. 3, the original image is divided into unit sizes without overlapping one pixel. It is assumed that there is a pattern similar to the above in the divided images P and Q. In the left divided image, there are three black pixels vertically arranged on the boundary. The right divided image does not have this, and two black pixels are arranged. This is data-compressed and stored in the second image storage device 1 '. When this is reproduced, a gap is formed between the divided images P and Q as shown on the lower left. This is because there is no overlap and no connection is made. Alternatively, a step is formed as shown on the lower right. Either way, you will be away from the original. The original picture cannot be faithfully reproduced. For this reason, the divided images are overlapped with each other by one dot at the boundary as shown in FIG.

The total number of dots of the original picture is determined by the number of dots of the image scanner in addition to the geometric size of the original picture. The vertical and horizontal sizes vary, and are not integral multiples of the basic size (for example, 256 × 256 dots). When the original image is cut out from the divided images (overlapping one dot at a time), a remainder is left in the basic unit area at the right end as shown in FIG. The same applies to the lower end.
If there is a surplus, a white pixel (that is, 0) is made to correspond to the surplus portion. Zeros are embedded in the remaining portions of the right and lower basic unit areas. When dividing an image, the end points of a straight line along the boundary are left as joint points. As will be described later, a junction point is originally defined as a point having a large curvature in a part of a curve. This is important as data for data compression. The compression data is a joint point and a straight line or a curve connecting these points. Even if it initially appears as a junction, it may disappear by the removal operation. The end point of the straight line on the boundary is left as a junction even if it should be erased by the removal operation. This is for the purpose of smoothly connecting the two divided images at the boundary line.

When the boundary line and the pattern line form a large angle (here, a right angle) as shown in FIG. 2, the lowermost black pixel of the right divided image Q does not disappear, so that it is smooth. Connected. However, if the pattern line and the boundary line form a small angle, the joints required to reproduce the pattern may disappear. In order to prevent this, it is decided not to remove the junction which forms the end point of the straight line on the boundary. This will be described with reference to FIG. As shown in (1), it is assumed that there is an oblique line that forms a small angle with respect to the dividing line in the original drawing. This oblique line and the dividing line intersect. In this case, the right divided image is as shown in (2). RS is a straight line along the boundary. Black pixels are lined up. When the junction is extracted, the result is as shown in (3). R, S, and T are extracted as junction points. When a normal joint removing operation is performed, the intermediate joint S having a low curvature disappears. If this disappears, as shown in (4), a data connecting T and R is obtained.
It becomes a data. Since the point S disappears, the reproduced image is (5)
As shown in the figure, the point S and the point T are not connected, and the points R and T above this point are connected. This is different from the original picture of (1). Why did this happen?
This is because the intermediate junction S has disappeared. In order to faithfully reproduce the original image, it is necessary to leave an intermediate joining point S. For this reason, the junction point, which is the end point of the straight line on the boundary, is to be retained in a later operation.

FIG. 6 is a flowchart showing such an operation. This is a flow chart for preserving the junction at the left boundary (x = 0) of the divided image. The coordinates take the X axis to the right and the Y axis to the bottom with the upper left as the origin. First, one closed contour ｛(x
_i , y _i )｝ _{i = 0} ^{Read out the data of leng-1} . here,
(X _i , y _i ) represents the coordinates of the contour point sequence. i is the number of the contour point sequence. i takes a value from 0 to length-1. “Leng” is the number of contour points constituting the closed contour. Since the target is a junction at the left boundary line (x = 0) of the divided image, x _i = 0 and x _{i + 1} = 0 are conditions.
In this case (x _i, y _i) sets a junction flag linear started. The i-th and subsequent coordinate points may be on this straight line or may deviate from this straight line. The last coordinate point on this straight line is left as a junction point, and a flag is set so that the curve starts from this point.

Let k be a number from i + 1 to length-1. If x _k = 0, it is a point on the boundary line. First, x _k ≠ 0
If any, the coordinates (x _k-1 , y
_k-1 ) is the starting point of the curve. The first end point of the straight line along the boundary was (x _i , y _i ), but the last end point was (x _i , y _i ).
_k-1 , y _k-1 ). So first x _k ≠
A flag to start the curve is set at the coordinates (x _k−1 , y _k−1 ) immediately before the point at which the value becomes zero. Another condition is y _k
Even if −y _k−1 <0, the immediately preceding point (x _k−1 , y _k−1 )
From the start of the curve. That is, this point is left as a joining point. The meaning of this will be described with reference to FIG. Although the above description is along the right boundary of the divided image, the junction along the left boundary is also set to the junction point at the end point of the straight line. x _i = 255, x _{i + 1} = 2
The condition of 55 is written at the bottom of the flowchart, which means that the same operation is performed for the linear component on the right boundary of the divided image. The same applies to a straight line along the upper line of the divided image. The condition for this is y _i
= 0, y _{i + 1} = 0. The same is true for a straight line along the lower boundary of the divided image. This can be represented by y _i = 255 and y _{i + 1} = 255.

FIG. 7 explains the meaning of setting a curve start flag at (x _k−1 , y _k−1 ) even when y _k −y _k−1 <0. FIG. 7 shows the X-axis downward and the Y-axis rightward in order to represent the order of processing vertically. Consider a junction at x = 0 (a boundary line on the left side) of the divided image. Consider a figure like (1). It is assumed that the contour point sequence is numbered in the order of the arrows. For convenience of description, the pixels are denoted by Irohanihohettilinul. In b, since x = 0 for the first time, a straight line flag indicating that a straight line along the boundary line starts is set here. If there is only the condition x _{k + 1} ≠ 0, then
As shown in (2), a curve start flag is set at F. In this case, the reproduced image is as shown in (3), and two points at the ends, i. Therefore, if the flag of the start of the curve is set even when y _k −y _k−1 <0, then as shown in (4), y _k −y _k-1 <
0 stands. Therefore, a curve start flag is set at the point giving (x _k−1 , y _k−1 ). In this case,
As shown in (5), all the straight lines along the boundary can be reproduced.

According to the present invention, a large image is simply divided and subjected to data compression, which is stored, and conversely, the divided images are combined and reproduced as an image. Since the contour is extracted and the image is stored as a number of joints and straight lines and curves connecting the joints, the contour is first when reproduced.
Will be reproduced. Therefore, the present invention can be used not only for simply reproducing an image as it is, but also for cutting a sheet along a contour line using a cutting plotter.
There is no problem in reproducing the image as it is. However, a further problem remains when trying to cut out the contour. Outputting only the outline is herein referred to as outline output.

The difficulty of outline output will be described with reference to FIG. It is assumed that the input image is a figure surrounded by a closed curve extending over four divided images as indicated by ヲ. This is divided into four small images of a unit size like a wa. The divided images are individually approximated by a function. Then, a figure surrounded by the boundary and the originally existing outline is obtained. In this simplified example, data is constituted by three joining points in a divided image, two straight lines connecting them, and one quadrant arc. This data is stored in the image storage device. In the case of playback, there are two methods. In the case of filling, the data of four partial divided images are superimposed as in the case of (ta), and the filling is performed in order from the upper left closed area (ta). Next, the upper right closed area is also filled. It looks like Les. If you paint this on the lower left as well, it will look like Seo. Furthermore, if the lower right is also painted, it will look like a lip. This is the end of the reproduced image.

This is a problem because the boundary line is buried in black since the place that is originally black becomes black.
However, when trying to cut the outline with a cutting plotter or the like, the outline will appear as if the four divided images are simply attached. This is the problem. This will be described more specifically. FIG. 9 is an original drawing drawn by a free curve. A line is a collection of curves, and there are both thick and thin lines. However, since the size is restricted by the conditions of the drawings of the patent application, the size is reduced and reduced.

FIG. 10 shows a reproduced image based on a contour line. This corresponds to a case where the sheet is cut by a cutting plotter. Such a diagram can be obtained by replacing the blade of the plotter with a pen. The outline appears so as to surround the area that should become a black pixel. A boundary line (division line) of the divided image also appears inside a region to be a black pixel.
The size of the basic divided image is about 20 mm x 20 mm
Since the size is about 30 mm × 30 mm, boundary dividing lines appear vertically and horizontally at the pitch of the divided image. Those having many black pixels clearly have boundary dividing lines. Such a dividing line does not originally exist but appears because the original image is divided into small divided images. Therefore, when cutting the contour line, it is necessary to prevent such a boundary dividing line from appearing.

FIG. 11 shows a reproduced image obtained by filling.
The output device is a 400 dpi laser printer. This is because the part surrounded by the closed contour line as shown in FIG.
Fill it black. The same figure as the original is obtained. Since the above-mentioned boundary dividing line exists only in the portion of the black pixel, such a boundary dividing line disappears if it is painted. It is reduced because it is a drawing of a patent application and its size is limited. Therefore, there is no difference in size from the original image of FIG. However, since the data is compressed and stored, enlargement and reduction can be performed in a short time by a simple calculation. Here, the size of the reproduction screen is different from that of the original image, but the size cannot be shown because of the restriction of the application drawing. It can be seen that the reproduction screen of FIG. 11 is a faithful reproduction of the original picture of FIG.

FIG. 12 is another original image. This is also an illustration of an animal mainly composed of curves. Includes fine lines, thick lines, and solid-colored parts. This is divided into small basic processing screens and data is compressed. FIG. 13 shows a reproduced image based on the outline. This also shows a boundary dividing line. In particular, a dividing line appears in a picture of a panda having many black portions. FIG. 14 shows a reproduced image by filling. This also used Rico-Imagio MF530 (400 dpi) as an output device. Although the size is different from the original, it is almost the same size because it is a patent drawing. From this, it can be clearly understood that the divided data compressed, stored and reproduced by the method of the present invention faithfully reproduces the original image.

The problem of outline output will now be described. Referring to FIG. 15, the features of the contour on the division boundary will be described. The contour on the division boundary has the following features. (1) It is a straight line. (2) Both joint points on both ends are on the boundary line, and the x coordinates or the y coordinates are equal. That is, the two ends of the coordinates of the straight line (X _a, Y _a), when a _{(X b, Y b),} X a
_{= X b = 0, X a} = X b = 255, Y a = Y b = 0,
Y _a = Y _b = 255, holds one of the. For example,
In FIG. 15, both ends of the curved line are on the left side of the divided image (that is, X _a = X _b = 0). Both ends of the curved line are on the upper side of the divided image (Y _a = Y _b = 0). Both ends of the curved line are also on the upper side of the divided image. Both ends of the curve beam is on the right side of the divided image _{(X a = X b = 255} ), one end of the curve c is in the lower side of the divided image (Y _a = 255), other end the right side of the divided image (X _b = 255). The line segment connecting these two points appears due to the division, and in the case of the outline output, it is necessary to devise a method of not outputting this.

FIG. 16 shows an example of an algorithm which does not output the contour on the dividing boundary line at the time of output. FIG.
, Let the coordinates of a point on the boundary be (X _a , Y _a ) and the coordinates of E be (X _b , Y _b ). From the starting point ヰ, draw a figure on paper or release paper with a pen or blade.
From ヰ to ノ, trace on the paper as it is, but when reaching the point ノ on the boundary line, raise the pen or blade, move the pen or blade along the boundary line while raising it, and The pen and the blade are lowered, and the contour line in the divided image is drawn from here. As described above, by moving the pen and the blade up and moving between the joining points on the boundary line, the dividing line can be prevented from appearing.

The novel aspect of the present invention has been described above. Next, the data output of the divided image will be described.
Regarding the handling of divided images, refer to the aforementioned Japanese Patent Application No. 4-25913.
7 and Japanese Patent Application No. 4-269646. A split image is an image of unit size,
For example, the image is 256 dots × 256 dots. This is data compressed and stored. FIG. 17 shows a configuration for that purpose.

A. Image storage device 1 B. Contour point sequence extraction device C. Outline point sequence storage device D. Data approximation mechanism AE Curvature calculation mechanism E '. Approximate curvature storage device F. Perfect circle extraction mechanism G. Perfect circular storage device H. Junction point position extraction mechanism I. Junction point position storage device Optimal junction extraction mechanism K. Optimal junction storage device L. Junction point removal mechanism Final junction storage device Data approximation mechanism B.O. Compressed data output mechanism Compressed data storage device Contour reproduction mechanism Image reproduction mechanism Reproduction data output mechanism U. Image storage device 2

First, an original image is read by an image scanner having an appropriate resolution. The whole is stored in the first image storage device A. This is the basic size (for example, 256 × 25
(6 dots), and the following compression processing is performed one by one. In the following, the target image is a divided image. The divided image is a binary image of white and black, and a contour line that is a boundary between the white region and the black region is extracted. Since there are 256 × 256 dots of pixels, the outline becomes an outline point sequence. There are several sequences of closed contour points. A number is assigned to each group of contour point sequences. The points in the outline point sequence are also numbered sequentially. A point can be represented as a point on the X, Y coordinates. The coordinates can be represented by numbering the sequence of points.

Further, using the parameter display, the X and Y coordinates are defined as functions of t. Since the contour point sequence is a point sequence continuously arranged along a closed curve, it can be displayed as a parameter.
Each contour point sequence group is approximated by a piecewise polynomial throughout. Since a piecewise polynomial becomes a continuous function, the curvature is obtained by performing second order differentiation on each contour point sequence. A contour point sequence having a constant curvature is a perfect circle. This separates as a perfect circle. Regarding the remaining contour point sequence group, a point having a large curvature is set as a joining point. The joining point is a point representing the outline point sequence. This is a point where the contour can be expressed as a joint point and a line connecting the joint point. At first, a portion having a large curvature is simply set as a junction, but this is a temporary junction. Junctions may be added later or removed. Candidate joining points are taken around the temporary joining point, and a suitable joining point is left as an optimal joining point.

When the joining points are determined, a straight line is drawn between the joining points,
Approximate by arc and free curve. The approximations are straight lines, arcs,
Perform in the order of free curves. In the case of a straight line, the time required for approximation is short. The amount of data is also small. The approximation of the arc is also easy. What is necessary is just to give the starting point of the arc, the radius, the central angle, the coefficient of the function, and the like. If it cannot be approximated by a straight line or a circular arc, a free curve is approximated. In this case, the space between the joining points is divided into M subsections and approximated by a piecewise polynomial. If the number of subdivisions is increased, the approximation can be increased, so that an approximation with desired accuracy can be made.

In this way, the parameters of the joint point, the straight line, the arc, and the free curve are obtained, and these are stored as data representing the divided image. For example, in the case of 256 × 256 dots, when the data compression of the present invention is carried out, it is approximately 300 to 256 dots.
Only about 500 bytes of data is required. If the image is a monochrome image, the amount of data is equal to the number of all pixels constituting the screen, so that the data amount is 8 kilobytes. According to the present invention, data can be greatly compressed. In the present invention, such data compression and storage are performed for each divided image. These data can be read out in reverse to reproduce straight lines, arcs, and free curves based on the junction point. It can be reproduced to any size and any position by calculation.

[A. Image storage device 1] This reads the entire original image by optical means and stores it as information decomposed for each pixel. Actually, commercially available images
Use a scanner. The input data is stored as, for example, a black and white binary image.

[B. Contour Point String Extraction Device] The contour point sequence extraction device extracts a contour point sequence of a divided image. The outline of the set of black pixels is called an outline. The divided image has several closed contour lines. Some contours have both ends at the boundary. This also becomes a closed contour by regarding the boundary dividing line as a contour. Since the image is a group of dots, the outline can be regarded as a group of points. A contour line is a set of points, and a sequence of successive points is called a contour point sequence. Let U be the total number of contour point sequences. The U contour point sequences are numbered from 0 to U-1. The total number of contour points in the u-th contour point sequence is represented by N (u). A number k is assigned to consecutive points in one outline point sequence. k is an integer of 0 to N (u) -1.
The u-th k-th coordinate of the contour points of the contour point sequence (x _k ^u,
y _k ^u) be represented by. All contour points are

[0037] ^{_{{(x k u, y k}} u)} k = 0 N u -1 u = 0 U-1 (1)

Is represented by because _{k = 0} ^{N u -1} is that can take values from the point sequence number k is 0 to N (u) -1. N (u) -1 contains parentheses and is 1
When it becomes a variable suffix because it cannot be a square, the parentheses are removed and written as Nu-1. Nu
-1 = N (u) -1. Since it is a suffix, it should be written on the top and bottom, but since this is not possible, it is divided into lower left and upper right. _{u = 0} ^U-1 means that the number u of the outline point sequence group is _{0 to} ^U -1
Is to take the value of The number u of the contour point sequence should be indicated by adding parentheses to the right shoulder of the variable.
Omit the brackets because they cannot be square.

[C. Contour Point Sequence Storage Device] The contour point sequence storage device is a device for storing the contour point sequence obtained in the preceding stage. (X
_k ^u, and stores this in the form of ^{_{y k u) k = 0 N}} u -1 u = 0 U-1. As described above, U is the number of all point sequences, and u is the number assigned to the point sequence. The number of points in the point sequence u is N (u)
And k is the point number assigned to this. Such things expressed by ^{_{k = 0 N u -1 u =}} 0 U-1. (X _k ^u, y _k
^u ) is the x, y coordinates of the kth point in the uth point sequence.

[D. Data approximation mechanism A] There are two data approximation mechanisms. This is the first one, but is appended with A to distinguish it. This is necessary for temporarily finding a place where the curvature of the outline point sequence is large and for finding a joint point. In order to obtain the curvature, some of the discrete points can be adopted as data and can be obtained from now on.
The present invention can be practiced even with such discrete curvatures. However, here, the entire contour point sequence group is approximated by a continuous function consisting of a piecewise polynomial, and then this is second-order differentiated to obtain a curvature. The approximation for this is the data approximation here. The independent variable is t, the dependent variables are x and y, and the x and y coordinates of the contour point sequence for each continuous group are approximated by a quadratic piecewise polynomial where t is an independent variable and x and y are dependent variables. The least squares approximation is repeated until the approximation accuracy falls within a predetermined range to obtain an approximate polynomial for each group of contour point sequences.
The range approximated by the piecewise polynomial is the entire contour point sequence group. This is not for obtaining final data but for obtaining a temporary junction.

The group contour point sequence in ^{_{u (x k u, y k}} u) of x
The coordinates are approximated by S _x (t), and the y coordinates are approximated by S _y (t). t is a parameter. S _x (t) and S _y
(T) is given as a linear combination based on a quadratic full-energy function system { _m }. S _x (t) is developed based on an aperiodic m-th order full-energy function ψ _k as a base.

[0042] _{S x (t) = Σ k} = -m M + m C k x ψ k (t) (2)

The full-energy function is a function name named by the present inventor. The degree m corresponds to the degree of the polynomial. M is the number of dimensions. In general, the m-th order full-energy function has a domain of [0, T], a parameter of k,
This parameter is represented with a suffix. C _k ^x
Is the coefficient of linear linear combination. ψ _k itself is a polynomial having a value near k.

Ψ_k (T) = 3 (T / M)^-mΣ_{q = 0} ^{m + 1}(-1)^q {T- (k + q) (T / M)} ^m ₊ / ｛Q! (M + 1-q)! ｝ (3) where k = −m, −m + 1,..., 0, 1, 2,.
・, M + M

Here, the plus added below the m-th power is 0 when the value in the parenthesis is negative, and is m-th when the value is positive, and is defined as follows.

(T−a) ^m ₊ = (t−a) ^m t> a, (4) 0 t ≦ a (5)

The basis function ψ _k is a chevron-shaped function having finite values from section numbers _{k to k} + m + 1 and having 0 on both sides. This is 0 such as {t- (k + q) (T / M)} ^m ₊
The coordinates of the m-th order function rising from are shifted one by one (q is increased one by one), and these are superimposed. Must be equal to 0 when t> (k + m + 1) (T / M). Under this condition, the superimposition coefficient is (−1) ^q / ｛q! (M + 1-q)! It is determined as｝. The size T of the area is the number N of points in the outline point sequence group.
Although it is easy to make it equal to (u), it may be defined as being proportional. As described above, the contour point sequence is approximated by using the full-energy function. However, since there are many division points having a T / M interval, the contour point sequence can be approximated without a joint point. To increase the degree of approximation, the order m of the full-energy function may be increased. However, when the degree is high, the number of calculations increases, so that processing time is required.

The inventor's claim is that many functions representing the variation of physical quantities in the natural world are represented as linear combinations of first-order and second-order full-energy functions. The m-th order full-energy function is a piecewise polynomial consisting of an m-th order polynomial having one peak over m + 1 regions. m = 0 is a simple box-shaped function, m = 1 is a triangular function, and m = 2 is a quadratic polynomial function spanning three regions. Here, only m = 2 is adopted. As a result, the contour lines can be expressed without excess or shortage. Of course, the present invention can also be configured using a full-energy function of m = 3 or more. m = 2
Therefore, the basis function has a value over three sections,
In a section having a maximum at the center, a quadratic polynomial is obtained. The above equation is for m = 2,

[0049] _{S x (t) = Σ k} = -2 M + 2 C k x ψ k (t) (6) ψ k (t) = 3 (T / M) -2 Σ q = 0 3 (-1 ) ^Q {t- (k + q) (T / M)} ² ₊ / ｛q! (3-q)! ｝ (7)

Is as follows. The basis function is Δt− (k + q) (T
/ M)} is a superposition of the quadratic function rising from four 0-shifted laterally by T / M is the ^two _+. The number of subdivisions is a function with values from k to k + 3. k + 4
As described above, since the constant is 0, the superimposition coefficient becomes (−
1) ^q / ｛q! (3-q)! It becomes｝. The number of basis functions is M + 5. M is the number of divisions of all sections, and this is called an approximate dimension number. This should not be confused with the order m of the full-energy function. Similarly, S _y (t) approximating the y-coordinate can be expanded by a full-energy function using C _k ^x as a coefficient.

[0051] The degree of approximation can be judged by this is how the original outline point sequence _{^{(x k u, y k u}} ) of how close to. This is evaluated by the least squares method, which is

[0052] _{Q = Σ {S x (t} k u) -x k u} 2 + {S y (t k u) -y k u} 2 (8)

Is to be minimized. The range of integration is all points of the outline point sequence group u. From this condition, it is possible to obtain coefficients, C _k ^x, the C _k ^y. by this,
Approximate functions S _x (t) and S _y (t) are obtained. When the calculation can be performed for the contour point sequence group u = 0, the following is performed for the group of u = 1. Similarly, for all contour point sequence groups, (u = U−
The calculation of 1 group) is performed.

[E. Curvature operation mechanism] Since the approximate function has been obtained for all the contour point sequence groups, this is second-order differentiated to obtain the curvature at each contour point sequence group and each point. K-th point of the contour point sequence group ^{_{u (x k u, y k}} u) curvature at K (t _k ^u
)

K (t_k ^u ) = ｛S_x '(T_k ^u ) S_y '' (T_k ^u ) -S_x '' (T_k ^u ) S _y '(T_k ^u )｝ / ｛S_x '(T_k ^u )^Two + S_y '(T_k ^u )^Two ｝^3/2 (9)

Can be calculated. The curvatures are obtained for all points in all contour point sequences.

[E '. Approximate curvature memory] curvature K obtained in the previous stage of (t _k ^u) point (edge point sequence group u, the point number k) is a device for storing for each.

[F. Perfect circle extraction mechanism] This is to determine whether a certain contour point sequence is a perfect circle or not based on the approximate curvature and extract a perfect circle. The part of the perfect circle is another straight line,
If they are in cross contact with the curve, they are not extracted as perfect circles. The pattern extracted by the perfect circle extraction is thereafter removed from the approximation calculation. This is denoted by Circle (u). u is the number of the contour point sequence group. Extracting a perfect circle has the following advantages. In the first case, even if a circle that is originally a circle is slightly distorted due to noise, the data is converted into a true circle, so that the noise drops and the shape can be determined more accurately. Since a circle can be specified only by the coordinates of the radius and the center, it is extremely effective in data compression. A perfect circle means that the curvatures at each point in the outline point sequence are all equal. Since the curvatures are obtained at all points in the contour point sequence, the curvature at each point can be checked to extract a perfect circle.

[G. Perfect circle storage device] Stores the center coordinates and radius r of the perfect circle obtained in the previous stage. Thus, data of the group u which is a perfect circle can be described by three values. In the case of a single perfect circle, this is an isolated circle point because the entire inside is a circle filled with black pixels. In the case of a double perfect circle, the space between the double circles corresponds to a circle filled with black pixels.

[H. Joining Point Position Extraction Mechanism] The joining points include a joint between a straight line and a straight line, a joint between a curve and a curve, a joint between a straight line and a curve, and the like. Since the lines having different slopes and different curvatures come into contact, this is called a junction. The junction points play a very important role when approximating the sequence of contour points as a function.
The junction determined through the curvature is a temporary junction, and the junction is additionally removed. According to the present invention, the amount of data can be minimized while maintaining a high quality reproduced image by accurately and appropriately determining the joining point.
FIG. 18 shows the outline of the junction extraction procedure. A divided image is input, and a contour point sequence forming an outer periphery is extracted. A joint point is extracted for each group of contour points obtained as described above. First, obvious joint points are extracted. This can be determined as a point having a large curvature. This is a temporary junction and not optimal. Also, this alone cannot find all the junctions. Conversely, those unnecessary as junctions are also included.

Therefore, it is necessary to correct the designation of the joining point. After that, a junction at the right angle portion is extracted. This is to make the connection at the intersection smooth when two straight lines intersect. The joining points of the straight lines are the starting point and the ending point of the thin line, and adding a joining point that is a thin line and does not appear under the condition that the curvature is large. Conversely, there are unnecessary junctions, which need to be removed. One is a junction in a straight line. When the joining point is in the middle of the straight line and the distance from the straight line is small, the straight line may be maintained even if this is removed. Since this is a junction point caused by noise, it is removed. The other is the junction of the arcs. What is supposed to be a single arc may appear as two arcs separated by two junctions due to the large number of joints. This also removes intermediate junctions and combines the arcs into a single arc. FIG. 19 shows an example of the junction. (A) shows the joint point of the outline of "a" by X, and (b) shows the joint point of Po by X. A point at which a straight line intersects, a cusp, or a point at which the curvature changes is a junction point. Although this is a case of a character, a joint point of a contour line can be obtained for an arbitrary figure.

When the joining point is determined, each section is approximated by a piecewise polynomial. The approximation of this piecewise polynomial is the same as that described above for the data approximation mechanism A, but in the previous one the approximation interval spanned the entire contour point sequence group. This time, instead, a piecewise polynomial approximation is performed for each junction. The accuracy of the polynomial approximation is achieved by performing a least-squares approximation to minimize the sum of the squares of the differences from the original coordinate values, as described previously.

[I. Joining point position storage device] This is the number of the temporary joining point and the coordinates ｛d _i (x _i ^u , y
_i ^u )｝.

[J. Optimal Joint Point Extraction Mechanism] What has been obtained so far is a temporary joint point determined only by the magnitude of the curvature. These may not be actual junctions. So we need to find the best joint. For example, when two lines intersect, four corner points are formed. In this case, the corners are crushed and easily enlarged. Therefore, an optimum joint point is obtained from the fact that adjacent corner points are on the contour of those lines. In this way, the corners at the intersections of the lines do not grow grossly. The optimal junction should be near the temporary junction determined by the method described above. Therefore, some of the vicinity of the temporary connection point are set as connection point candidates, and the optimum point is determined as the optimum connection point from the connection point candidates.

[K. Optimum junction storage device] This is a device that stores all the optimal junctions obtained by the above operation. Stores the optimum junction as _{^{(x i u, y i u}} ). u is the number of the contour point sequence. i is the number assigned to the junction in group u.

[L. Junction point removal mechanism] Here, unnecessary junction points are removed from the junction points extracted so far.
Unnecessary joints are removed from straight or circular joints.
The joint points determined so far are obtained by approximating the contour line with a piecewise polynomial, determining the curvature, determining the temporary joint point, and determining the optimum joint point from the joint point candidates. Therefore, there should be almost no unnecessary junctions, but there is still a possibility that there are unnecessary junctions due to the influence of image noise. It is the junction of a straight line and the junction of an arc. There may be unnecessary junctions in the free curve, but in the case of the free curve, it cannot be determined whether or not this is unnecessary. In the case of a straight line or an arc, it is a definitive figure, and one that deviates from this and one that does not deviate can be clearly distinguished. For this purpose, unnecessary joining points of straight lines and arcs are removed.

[Removal of Joint Points of Straight Line] For example, there are cases where three joint points U, V, and W are arranged substantially on a straight line, but slightly deviate from the straight line. In this case, it is assumed that the distance between the straight line connecting the end points U and W and the intermediate point V is within a certain distance. V is considered to be generated by noise. In this case, the junction point V in the middle of the straight line is removed. Excluding this, one coordinate point is reduced and one straight line is reduced. The data is reduced by two. This is more important in reducing the amount of data than in reproducing straight lines.

[Removal of Arc Joining Point] A plurality of joining points may be continuous among arc joining points extracted up to the previous stage. In this case, if the data quality is maintained even if the intermediate junction is removed, the junction is removed. In other words, if the center and radius of the arc connecting the intermediate connection point and the next connection point are close to the center and radius of the arc connecting the first connection point and the final connection point, the intermediate connection point is removed. Eliminating the junction reduces the data. However, the removal of the arc junction is not for data reduction but for reproducing a beautiful figure. What was actually a single arc was blurred due to noise, and the line looked like a set of multiple arcs. So this also means shaping figures. One of the reasons why the quality of the reproduced image of the present invention is good is that a straight line or an arc is correctly drawn by removing the joint points.

[M. Final junction storage device] The final junction obtained by removing unnecessary junctions is stored in the final junction storage. The final junction is at coordinates (x _i ^u , y _i ^u )
_{Indicated by i = 0} ^I-1 . Where i is the number of the junction, u
Is the number of the contour point sequence group, and I is the total number of final joining points of the contour point sequence group u.

[N. Data approximation mechanism B] This is a mechanism for approximating data from data such as a contour point sequence, a final joint point, and a perfect circle obtained so far. It is a central part of the present invention. What is input from the respective storage, contour point string storage device ... contour point sequence {(x _k ^u, y _k
^u )｝ _{k = 0} ^{N u -1} final junction storage device .. final junction ｛(x _i ^u , y _i
^u )｝ _{i = 0} ^I-1 perfect circle storage device... Circle Circle (u). The suffix u of the outline point sequence is the number of the outline point sequence group. k is a contour point sequence number in the contour point sequence group u. N (u) (represented by N u) is the number of contour point sequences in the group u. i is the final junction in group u. Although the final joint point is obtained, a straight line, a circular arc, and a free curve approximation are made between two adjacent joint points (between the joint points). Using the parameter t in the approximation, the x component is given by s _x (t):
The y component is represented by s _y (t).

First, the entire contour point sequence is defined as a parameter t.
This is the same method as expressed in. But this time, since the region is between the junctions, the range of t and t and s _x
The correspondence between (t) and s _y (t) is different from the previous one. Whether the section starting from a certain junction is a section of a straight line, a section of an arc, or a section of a free curve is known when the curvature is obtained at each point. If it is a straight line, it is only necessary to connect adjacent points. In the case of an arc, the coordinates and radius of the center may be obtained. In the case of a free curve, it is necessary to determine the coefficient of the full-energy function.

[Approximation of Straight Line Section] A section starting from a joint point of a straight line is determined to be a straight line at the stage of extracting the joint point. The coordinates of the start and end points are known. Therefore, there is no need to determine the gradient of the straight line. You only need to flag it as a straight line. [Approximation of Circular Arc Section] Approximation of a section starting from a junction of arcs will be described. This section is determined to be a circular arc at the point of joint point extraction. The approximate curves s _x (t) and s _y (t) representing the arc are expressed by the following linear combination of trigonometric functions. Let the observation interval be t∈ [0, T]
Then, s _x (t) and s _y (t) are

S_x (T) = A_xcos (2πt / (T / n_arc )) + B_x sin (2πT / (T / n _arc )) + C_x (10) s_y (T) = A_ycos (2πt / (T / n_arc )) + B_y sin (2πT / (T / n _arc )) + C_y (11)

Is represented by n _arc is the ratio of the arc to the total circle. That is, it is a value obtained by dividing the central angle of the arc by 360 degrees. 2πn _arc is the central angle of this arc.
The variable 2πt / (T / n _arc ) is a parameter from the starting point of the arc.
It is the central angle up to the point corresponding to point t. (C _x , C _y )
Is the coordinates of the center of the arc. At this time,

[0075] _{^{A x 2 + B x 2 =}} A y 2 + B y 2 (12) B y / A y = B x / A x (13)

If the above holds, the approximation function becomes an arc. In this case, parameter defines an arc - data each coefficient A _x of the function, B _x, is _{C x, A y, B y} , C y, n arc. If it is known from the beginning that this section is a circular arc, such parameters can be uniquely determined from the coordinates of the start and end points, the curvature and the coordinates of one point in the middle.
[Approximation of Free Curve] A section starting from a joint that is neither a joint of a straight line nor a joint of an arc is approximated by a free curve. Although represented by the parameter t, the outline point sequence is (x _i3 ^u , y _i3 ^u
), And corresponding to t, (t _i3 ^u , x
_i3 ^u ) and (t _i3 ^u , y _i3 ^u ). Until now, the suffix of the contour point sequence was k. However, since k is used as the section number of the section, i3 is used as the contour point sequence number instead of k. Then, the total number of contour point sequences is set to n3.

Then, s _x (t) and s _y (t) are developed with the quadratic full-energy function ψ _k3 as the base. This is 3
A function that has a value in only one subdivision. The interval is [0,
T], the quadratic full-energy function ψ _k3 is an M-dimensional functional system

Ψ _k3 (t) = 3 (T / M) ^-2 Σ _{q = 0} ³ (−1) ^q (t−ξ _{k + q} ) ² ₊ / {(q! (3-q)!)} (14)

K = −2, −1, 0, 1, 2,..., M + 2. Using this as the base, s _x (t _i3 ) and s _y (t _i3 )
Is calculated using coefficients c _k ^x and c _k ^y .

S _x (t _i3 ) = Σ _{k = −2} ^{M + 2} c _k ^x ψ _k3 (t _i3 ) (15) _sy (t _i3 ) = Σ _{k = −2} ^{M + 2} c _k ^y ψ _k3 (T _i3 ) (16) Here, when t> ξ _{k + q} (t−ξ _{k + q} ) ² ₊ = (t−ξ _{k + q} ) ² (17) When t ≦ ξ _{k + q} (t−ξ _{k + q} ) ² ₊ = 0 (18)

Is defined as ξ _{k + q} is _defined as
It is a subdivision when divided equally. ξ _{k + q} = (k + q) T / M (19)

The coefficients c _k ^x and c _k ^y are defined as the values (x _i3
^u , y _i3 ^u ) and the values of s _x (t _i3 ) and s _y (t _i3 ) are determined to be similar. Determine the value of the coefficient by the least squares method. The square error Q is

Q = Σ _{i3 = 1} ⁿ³ | x _i3 ^u −s _x (t _i3 ) | ² −Σ _{i3 = 1} ⁿ³ | y _i3 ^u −s _y (t _i3 ) | ² (20)

Is defined by By minimizing this, the coefficients c _k ^x and c _k ^y can be obtained.

[O. Compressed Data Output Mechanism] The outline of a figure is separated into straight lines (line segments), perfect circles, circular arcs, and free curves by the above procedure. These have a starting point and an ending point, and have parameters such as inclination, center, and radius.
The data to be stored differs depending on the type. In the case of straight line data, a flag indicating a straight line,
The coordinates of the starting point of the straight line are stored as data. Since the end point coordinates are given as the start point of the next section, there is no need to store them here. The inclination of the straight line is a parameter that defines the straight line. However, since the straight line can be formed by connecting the coordinates of the starting point and the coordinates of the next junction, no inclination is required. This is not stored either.

In the case of the perfect circle data, the data can be obtained directly from the perfect circle storage device G. This has already been selected by the first data approximation mechanism A. In the case of a perfect circle, the flag indicating the perfect circle, the center coordinates of the circle, and the radius of the circle are provided.
Stored as data. As the arc data, a flag indicating an arc, the coordinates of the starting point of the arc, the arc division length (arc length / perimeter), the number of contour points, and the coefficient of the function are stored. De of free curve - The data, dimensionality of the function, the contour points, the middle point (μ _x, μ _y) of the variation of the contour point sequence and the coefficient of the function c ^x, stores c ^y.

[P. Compressed data storage device] Stores data such as straight lines, perfect circles, circular arcs, and free curves output from the compressed data output mechanism. This is output at an appropriate time after being stored. Up to this point, the data is compressed and generated and stored. A large original image is input, stored in the first image storage device, cut out for each divided image, subjected to the above data compression operation, and stored in the compressed data storage device P. This is what was expressed as a twelfth image storage device 1 '. Similarly, the processing of the next divided image is performed.
Data compression processing of the divided images is performed one after another, and this is compressed data.
Data in the data storage device. Table 1 shows the data structure stored in the compressed data storage device P.

[0088]

[Table 1]

The size of the data will be described. If a straight line exists between the joining points, one byte is used for the flag indicating the straight line, and two bytes are used for indicating the start point of the line segment (x coordinate and y coordinate).
Requires a total of 3 bytes. When the arc between the joining points is an arc, the flag indicating the arc is 1 byte, and the start point of the arc is 2 bytes.
4 bytes to represent the arc center angle, 1 byte to represent the number of contour points, 1 to represent the approximate function coefficients (there are 6)
A total of 20 bytes are required for 2 bytes. In the case of a free curve between the joining points, 1 byte is used to represent the dimension number M of the function, 1 byte is used as the number of contour points, 2 bytes is used to represent the center of variation of the contour points, and 2M is used to represent the coefficient of the approximate function. Bytes, 4 in total
+2 Mbytes.

[R. Contour Reproducing Mechanism] This is a mechanism for reproducing a contour line to be a skeleton of an image from stored compressed data. The contour may be a straight line, a perfect circle, an arc, or a free curve. [Reproduction of Straight Line] The reproduction of a straight line is performed by connecting a straight line from the coordinates of the starting point to the coordinates of the joining point in the next section. No data regarding the inclination of the straight line is required. [Reproduction of Perfect Circle] Reproduction of a perfect circle is performed by drawing a circle having a given radius around the center coordinates from data of the coordinates and the radius of the center. [Reproduction of circular arc] Reproduction of circular arc is performed by storing the stored data (A _x , B _x ,...).
・) Is substituted into the following equation.

S _x (t) = A _x cos ｛2πt / (T / n _arc )｝ + B _x sin ｛2πt / (T / n _arc )｝ + C _x (21) S _y (t) = A _y cos {2πt / (T / n _arc )} + B _y sin {2πt / (T / n _arc )} + C _y (22)

The x and y coordinates are obtained from S _x (t) and S _y (t) by varying the parameter t in the interval [0 to T]. [Reproduction of Free Curve] The value of the basis ψ _K3 of the approximation function at each sample point t _i is such that the sample point t _i is in the interval [(L−1) (T /
M), L (T / M)] (1 ≦ L ≦ M), p =
By using the _{L-t i × M / T} ,

Ψ _k3 (t _i ) = 0.5p ² k = L (23) ψ _k3 (t _i ) = p (1-p) +0.5 k = L + 1 (24) ψ _k3 (t _i ) = 1 −ψ _L3 (t _i ) −ψ _{L + 13} (t _i ) k = L + 2 (25) ψ _k3 (t _i ) = 0 k ≦ L−1, L + 3 ≦ k (26)

Is represented by Here, L is a natural number of dimensions M or less. Using such a basis ψ _K3 , the approximate function value S (t _i ) at each sample point is

[0095] _{S (t i) = Σ k} = L L + 2 C k ψ k3 (t i) (27)

Is obtained by If the output is a cutting plotter or the like and a sheet member is cut out, the sheet may be cut along such a contour line.

[S. Image reproducing mechanism] Since the outline is obtained, a portion surrounded by the outline is set as a black pixel, and a black-and-white binary image is reproduced in a logo / illustration shape. Or, conversely, the portion surrounded by the outline can be white pixels, and the rest can be black pixels. Further, a portion surrounded by the outline may be set to a certain color, and another portion may be set to another color.

[T. Reproduction data output mechanism] In order to reproduce an image from the coefficient of the function, for example, the size of the reproduced image is set to 1
Use a layout editor that can specify from mm square to A3 size large. The reproduced image is printed by using a laser printer having an accuracy of 300 DPI or a laser printer having an accuracy of 600 DPI. Further, it can be output by a 3000 DPI electroplanter (printing device), a 400 DPI postscript compatible printer, a laser cutter, or the like. The original figure is stored in the form of function coefficients, and can be scaled to any magnification. Also, the center can be arbitrarily specified for the coordinates. Therefore, it is possible to output a character, a figure, or the like having an arbitrary shape at an arbitrary position and an arbitrary size.

[0099]

DESCRIPTION OF THE PREFERRED EMBODIMENTS According to the present invention, an original image is read (input) by an image scanner, divided into divided images, compressed, and stored in a storage device. Then, the data is read from the storage device and synthesized, and a reproduction diagram is drawn (output) by a laser printer, a cutting plotter or the like. In the present invention, the size of the divided image can be set to an arbitrary size. For example, 400 × 40
It can be 0 dots or 1000 × 1000 dots. However, the size of the divided image is 2 n (6
4, 256, 512, 1024) is more convenient. In the embodiment described here, the size of the divided image is 256
× 256 dots.

Since the divided image is defined by the number of dots,
The actual size of the divided image changes depending on the dpi (dots per inch) of the image scanner for reading the image. In the case of an image scanner with a low reading accuracy (dpi is small), the divided image becomes large.
In this case, when handling original images of the same size, the number of divisions of the divided image (here, the number of boxes) is reduced. On the other hand, in the case of an image scanner that performs fine reading (dpi is large), the divided image becomes small. The number of divisions (the number of boxes) increases and the processing time needs to be long. The reading accuracy (dpi) of the image scanner varies from 90 to 400 dpi.

[0101]

[Table 2]

Table 2 shows the dpi of the image scanner, the dimensions of the divided images, and the number of divided images (the number of boxes) that can cover the A4 screen when it is input. It also shows the amount of data read at that time. Since the dpi of the image scanner is the number of dots per inch, 256
One side size S of the divided image of the dot is 25.4 × 256 /
dpi (mm). For 90 dpi,
When S = 72.2 mm and 400 dpi, S = 16.
2 mm. In the case of A4 original picture, 210 mm x 296
mm, R = [210 / S] +1 is the number of divided images in the horizontal direction. Here, [...] means an integer part of .... In the vertical direction, Q = [296 / S] +1. The number of boxes at the time of full-size reading of A4 is the product PQ of these. At 200 dpi, the number of boxes is 70. The number of boxes is 266 at 400 dpi. The data amount at the time of reading is 256 × 256 × (the number of boxes). For example, the following can be used as the input / output device.

[Input Device] System Quality IS
-500 (400 dpi) This is an image scanner having a resolution of 400 DPI and 256 gradations. In the case of the original picture of A4 size, 40
It is about 8 seconds at 0 dpi and about 5.5 seconds at 240 dpi.

[Output Device] Laser printer manufactured by Ricoh Co., Ltd. (trade name: Ricoh S)
P8) (240 dpi, pseudo 600 dpi) This is a sharp laser printer output. Although depending on the complexity of the original image, the output speed is about 1 minute or less when the original image is A4 size. It is an extremely fast output machine. Output dimensions are up to A4.

Laser printer manufactured by Ricoh Co., Ltd. (trade name: Ricoh Imagio MF530) (400 dpi) A high-resolution beautiful laser printer output can be obtained. By using the composition as it is, it can be used as a DTP printer. Output speed is slow. When outputting an image stored by the method of the present invention, the size of A4 is 15
It takes between minutes and 30 minutes.

Cutting plotter For example, FC2100-50, FC21 manufactured by Graphtec
00-90A or the like can be used. This is width 50
A film having a width of 0 mm to 50 mm can be cut. The cutting blade is a super steel blade, a ceramic blade, a sapphire blade or the like. This cuts the film consisting of release paper and backing paper. In this case, the cut film can be attached to some kind of base material and used for signboards, information boards, building decorations, road signs, and the like. When using thermal transfer film,
It can be used to draw letters and patterns on t-shirts, outdoor goods, T-shirts, etc. By changing the blade of the cutting plotter to a pen, you can draw drawings on paper.

Thermal plotter (400 dpi) This draws a drawing on thermal paper. For example, TM1000-OPF of Graphtec can be used. It can output drawings and characters on paper of A0, A1, A2, A3, etc. Suitable for single character output such as banners, venue guides, funeral name tags, etc. The ability to play large screens makes it possible to produce oversized banners. When outputting a playback image using a cutting plotter, use a blade or pen to outline (outline output)
Will be drawn. As described above, if the outline is output without outputting the division boundary line, the sheet member or the like can be cut along the outline. Signs, information boards,
It can be used for a wide range of applications such as display boards, name tags, banners, and iron prints. The position and size can be freely set by calculation. FIG. 20 shows the mark (the original picture is the actual size shown in FIG. 28).
Is read out by the method of the present invention, divided, data compressed, stored, synthesized, and output as an outline. The boundary dividing line has disappeared. FIG. 21 shows the word verd (the original picture is the actual size of FIG. 29) read and divided by the method of the present invention, compressed and stored, further synthesized, and output as an outline. Also in this case, the dividing boundary line has disappeared. FIG.
Similarly, the present invention is applied to the word “Nyo-sai” (the original picture is the actual size of FIG. 30) and output in outline. As described above, since the present invention can cut a sheet accurately along a contour line, it can be used for various applications. FIG. 27 shows a cutting sheet in which only the upper layer is cut by a cutting plotter. No division boundaries appear. It has a long side of 1120 mm and a short side of 46
This is a large sheet that is 1.5 mm. The height of the letter K is 122 mm. Such a wide sheet can be easily cut out.

Next, a reproduced image output by a normal printer will be described. FIG. 23 shows a certain original picture. Characters such as autumn visiting clothes and autumn kimono market are written. The size has been reduced to meet the restrictions for patent applications. This is read by the image scanner, divided according to the method of the present invention, and compressed and stored. For data compression and the like, a personal computer (trade name: PC-9801BA) manufactured by NEC Corporation was used. FIG.
Is a reproduced image by a laser printer (trade name: Ricoh Imagio MF530) manufactured by Ricoh Co., Ltd. This is Figure 2
Looks exactly like 3. Although the size is of course different, the size is similar because the drawing is for a patent application. It can be seen that the original is reproduced beautifully. FIG.
Is a reproduced image of a portion of the arrival at the autumn visit by a laser printer (trade name: Ricoh SP8) manufactured by Ricoh Co., Ltd. The size can be variously changed by calculation. FIG.
Is an image reproduced by a laser printer (trade name: Ricoh SP8) manufactured by Ricoh Co., Ltd. in the area of Kimono City in autumn. It can be seen that this is also a faithful reproduction of the original picture.

[0109]

[Table 3]

Table 3 shows the time required for such processing. It takes 13 minutes to read the portion of the autumn kimono city by the image scanner, and it takes 1 minute and 40 seconds to cut out the divided image using a personal computer (trade name: PC-9801BA) manufactured by NEC Corporation, compress the data, and store the data. When outputting this, a laser printer (trade name Ricoh SP) (trade name: Ricoh Imagio MF530) manufactured by Ricoh Co., Ltd.
8) It took 1 minute and 19 seconds. In the latter case, the total time from data reading to output is only 3 minutes and 12 seconds.

[0111]

[Table 4]

Table 4 shows the time of reading, data compression, and output of the portion of the arrival at autumn. 21 seconds for reading with an image scanner and 1 minute 57 for processing with a personal computer (trade name: PC-9801BA) manufactured by NEC Corporation.
It took seconds. Output of the reproduced image took 8 minutes and 44 seconds with a laser printer manufactured by Ricoh Co., Ltd. (trade name: Ricoh Imagio MF530) and 1 minute, 4 seconds by laser printer manufactured by Ricoh Co., Ltd. (trade name: Ricoh SP8). In any case, it can be seen that in a very short time, a large image can be divided, data approximated by function approximation, stored, and the stored data can be combined into an image. Conventionally, when the original image is optically read, noise is generated and the reproduced image becomes dirty.However, in the present invention, since there is also an operation of actively correcting a linear portion and an arc portion, etc. Images can be obtained.

[0113]

According to the present invention, a large image is divided into images having a small unit size, and the resulting image is subjected to contour point sequence extraction, joint point extraction, function approximation, etc. as a black and white image, and the compressed data is obtained.
Data in the storage device. This is repeated many times to compress the entire image. Conversely, the data of each divided image is read out and synthesized and reproduced. Since outline point sequence extraction, junction point extraction, and function approximation are performed for a small range, a normal personal computer can be used sufficiently. Since junctions are extracted and connected by a straight line, a circular arc, or a free curve, it is possible to eliminate the influence of noise caused by optical factors that occur during reading. Conversely, a smooth curved portion can be beautifully reproduced, and a high-quality reproduced image can be obtained. Data to be memorized because of function approximation
The amount of data can be reduced and smooth approximation can be achieved. Therefore, the output time is shortened, and the quality of the reproduced image can be improved.

[Brief description of the drawings]

FIG. 1 is a diagram showing an original image which is read by an image scanner, stored in a first image storage device, divided into smaller basic divided images and subjected to data compression processing in order, and transferred to a second image storage device 1 '. The figure which shows the procedure of this invention which writes compressed data.

FIG. 2 is a diagram showing that one pixel is overlapped vertically and horizontally when a divided image is cut out from an original image.

FIG. 3 is a view for explaining what happens to a reproduced image when one pixel is not overlapped when cutting out a divided image from an original image.

FIG. 4 is a diagram showing that when the original image is divided into divided images of a unit size, there is a surplus at the right end and the lower end, but 0 is embedded in this.

FIG. 5 is a diagram for explaining that an end point of a straight line along a boundary needs to be left as a joining point;

FIG. 6 is a diagram illustrating an algorithm for leaving an end point of a straight line at a boundary as a joint point.

FIG. 7 shows that y _k −y
The figure for explaining that _k-1 <0 is required.

FIG. 8 is a view showing how a pattern extending over four divided images is divided and reproduced. There is no problem in the case of filling, but in the case of drawing an outline, it will be described that a dividing line due to a boundary appears in a black image portion.

FIG. 9 is an original view showing the effect of the present invention.

FIG. 10 is a diagram showing the original image of FIG. 9 compressed and stored in accordance with the method of the present invention and reproducing the contour image by synthesizing the divided data. A dividing boundary line appears at the black pixel portion.

FIG. 11 is a diagram showing the original drawing of FIG. 9 divided and compressed by the method of the present invention, stored, and the divided drawing is synthesized to reproduce the original drawing.

FIG. 12 is another original view showing the effect of the present invention.

FIG. 13 divides the original drawing of FIG. 12 by the method of the present invention, and
FIG. 9 is a diagram showing a data obtained by compressing data, storing the data, and synthesizing the divided data to reproduce an outline image. Reproduced (outline output) image with contour lines. A dividing boundary line appears at the black pixel portion.

FIG. 14 is a diagram showing the original image of FIG. 12 reproduced by dividing and compressing the data by the method of the present invention, storing the divided data, synthesizing the divided data, and filling the closed contour lines.

FIG. 15 is a view showing characteristics of a contour on a division boundary.

FIG. 16 is a view for explaining one algorithm for not outputting an outline on a division boundary at the time of output.

FIG. 17 is a configuration diagram for explaining a mechanism for compressing data of a divided image.

FIG. 18 is a flowchart for explaining a procedure of a joint extraction method adopted in the data compression method of the present invention.
G.

FIG. 19 is a diagram exemplifying a junction extraction result of “a” and “po”;

FIG. 20 is a diagram in which characters are input, divided, data compressed, and contour lines are output (outline outputs) according to the method of the present invention. The dividing border has disappeared.

FIG. 21 is a diagram in which characters are input, divided, data compressed, and contour lines are output (outline outputs) according to the method of the present invention. The dividing border has disappeared.

FIG. 22 is a diagram in which the character “Nyo-sai” is input, divided, data-compressed, and output as an outline (outline output) by the method of the present invention. The dividing border has disappeared.

FIG. 23: “Autumn kimono city,
Original drawings that include words such as "Welcome to the autumn, shades, kimono festival".

FIG. 24 is a reproduction view of a portion of the arrival at autumn in FIG. 23 by a laser printer manufactured by Ricoh Co., Ltd. (trade name: Ricoh SP8).

FIG. 25 is a reproduction diagram of a portion of the autumn clothing market in FIG. 23 using a laser printer (trade name: Ricoh SP8) manufactured by Ricoh Co., Ltd.

FIG. 26 is a reproduction view of FIG. 23 reproduced by a laser printer (trade name: Ricoh Imagio MF530) manufactured by Ricoh Co., Ltd.

FIG. 27 is a diagram showing a cutting sheet in which only the upper layer is cut by a cutting plotter.

FIG. 28 is a full-scale view of a character that is the original image of FIG. 20;

FIG. 29 is a full-scale view of a character that is the original image of FIG. 21;

FIG. 30 is a full-scale drawing of the character “cool” which is the original picture of FIG. 22;

Continuation of front page (56) References JP-A-60-49483 (JP, A) JP-A-57-159363 (JP, A) JP-A-62-274372 (JP, A) JP-A-61-221979 (JP) , A) IEICE Transactions D VOL. J70-D NO. 6P. 1164-1172 (June 1987) IEICE Transactions D-II VOL. J73-D-II NO. 9 P. 1448-1457 (September 1990)

Claims (10)

(57) [Claims] 1. A first method of storing data obtained by optically reading a character / graphic figure in correspondence with a finite number of pixels arranged vertically and horizontally.
(B) a data compression device that divides the stored image data into basic small regions and sequentially compresses the divided images by function approximation, and (c) stores the data compressed. A second image storage device; and (d) a reproduction device that synthesizes data relating to the previous divided image stored in the second image storage device and reproduces the original image. b): (b1) a contour extraction mechanism for extracting a contour of a character figure from a divided image; and (b2) storing two-dimensional coordinates (X, Y) of the extracted contour for each continuous group. (B3) approximating the X and Y coordinates of the contour point sequence for each group with a quadratic piecewise polynomial in which t is an independent variable and x and y are dependent variables; Least square approximation until the specified range A data approximation mechanism A for calculating an approximate polynomial for each group of the repeated contour point sequence; and (b4) a curvature calculation mechanism for calculating a curvature at each point of the point sequence for each group in the x, y space from the approximation result; (B5) a perfect circle extraction mechanism for extracting a perfect circle from the curvature data for each group; and (b6) an infinity from the curvature data of the contour point sequence excluding the perfect circle.
Or a temporary joint position extraction mechanism for extracting a point having a curvature greater than or equal to a certain value as a temporary joint point. (B7) Joint points corresponding to adjacent corner points form corner points.
The corner points are temporarily set according to the condition that they
An optimal joint extraction mechanism for finding an optimal joint from candidate joints near the joint ; and (b8) an optimal joint that is a starting point of two or more continuous straight lines or two or more continuous arcs. and unnecessary junction removing mechanism also remove it found unnecessary junctions such approximation accuracy is maintained by, between the (b9) adjacent the junction of the sequence of points the same group linear,
When approximated by an arc of the order which at a given approximation accuracy is not obtained
When t is an independent variable and x and y are dependent variables, it is approximated by a second-order piecewise polynomial and the second order is used until the approximation accuracy falls within a predetermined value.
Wherein between the piecewise polynomial junctions while increasing the number of dimensions adjacent repeated a least squares approximation straight line, an arc, a data approximation mechanism B approximated by piecewise polynomials, every group of (b10) point sequence compressed data storage mechanism for storing the parameters of the function approximating the one coordinate and the adjacent <br/> junction of the junction and be more of, stored in a second image store compressed data of the (c) And a compression / synthesis / combination / output device for reading / dividing a character / figure. 2. (a) Data obtained by optically reading a character / graphic
Data that is stored in correspondence with a finite number of pixels
And (b) storing the stored image data.One pixel at the border
So that they overlapDivided into basic small areas and divided
Data compression for images sequentially by function approximationdata
compression(C) a second image storage for storing the compressed data;
Storage device; and (d) relating to the previous divided image stored in the second image storage device.
And a playback device that plays back the original image.
Only,  The data compression device (b) for the divided image extracts (b1) a contour of a character figure from the divided image.
(B2) two-dimensional coordinates (X, Y) of the extracted contour line
(B3) a contour point sequence storage mechanism for storing the contour point sequence for each groupX,YIndependent coordinates t
Variables, a quadratic piecewise polynomial with x and y as dependent variables
Similar, and repeat the least squares approximation until the approximation accuracy is within the specified range.
Data approximation to find an approximate polynomial for each group of repeated contour points
(B4) a point sequence for each group in the x, y space from the approximation result
And (b5) a perfect circle extraction for extracting a perfect circle from the curvature data for each group.
(B6)Outline point sequence excluding perfect circleFrom the curvature dataInfinity
Or has a curvature greater than a certain valueTemporarily connect points
A temporary junction position extraction mechanism for extracting as a joint point; (b7)Joint points corresponding to adjacent corner points form corner points
The corner points are temporarily set according to the condition that they
From candidate junctions near the junctionOptimum for finding the optimal junction
(B8) two or more continuous straight lines or two or more continuous circles
Approximate from the best joint point which is the starting point of the arc without it
Unnecessary joints that maintain accuracy are found and eliminated.
(B9) Same as the unnecessary joint removal mechanism to be removedGroup point sequenceNeighbors withinDoJunctionofA straight line between
Approximate in the order of circular arcs and this does not provide the required approximation accuracyWhen
ComeIs a quadratic classification where t is an independent variable and x and y are dependent variables.
Approximate with a polynomial and quadratic until approximation accuracy falls within a specified value
ofLeast-squares approximation while increasing the number of dimensions of piecewise polynomial
Repeat adjacentDoJunctionofStraight line, circular arc, piecewise
A data approximation mechanism B approximating with a polynomial;Do
JunctionofPressure that stores the parameters of the function that approximates
Compressed data is stored in the second image storage device (c).
Character / graphic division reading characterized by being read
Compression storage synthesis output device. 3. A first method of storing data obtained by optically reading a character / graphic figure in correspondence with a finite number of pixels arranged vertically and horizontally.
And (b) the stored image data is separated by one pixel at the boundary.
Data for data compression by sequential function approximation for the divided image divided divided into basic sub-area manner overlapping One
A compression device; (c) a second image storage device for storing the compressed data; and (d) data relating to the pre-divided image stored in the second image storage device, and reproducing the original image. A data compression device (b) for the divided image, wherein: (b1) a contour extraction mechanism for extracting a contour of a character graphic from the divided image; A contour point sequence storage mechanism for storing dimensional coordinates (X, Y) for each continuous group; and (b3) X and Y coordinates of the contour point sequence for each of the groups are t as independent variables, and x and y as dependent variables. A data approximation mechanism A for approximating with a quadratic piecewise polynomial and repeating the least squares approximation until the approximation accuracy falls within a predetermined range to obtain an approximate polynomial for each group of contour point sequences; and (b4) song at each point in the point sequence for each group in x, y space A curvature calculation device for obtaining the, (b5) and circularity extraction mechanism for extracting a perfect circle from the data of the curvature of each group, (b6) infinity from the data of the curvature of the contour point sequence except circularity
Or a temporary joint position extraction mechanism for extracting a point having a curvature greater than or equal to a certain value as a temporary joint point. (B7) Joint points corresponding to adjacent corner points form corner points.
The corner points are temporarily set according to the condition that they
The optimum junction extraction mechanism for determining the optimum junction from the junction point candidate in the vicinity of the junction, (b8) is a two or more arc start point to the straight line or a continuous two or more consecutive, not on the division boundary line optimal junction also the required junction removing mechanism to remove it found unnecessary junctions such approximation accuracy is maintained without it out of, between the (b9) adjacent the junction of the sequence of points the same group Straight line,
When approximated by an arc of the order which at a given approximation accuracy is not obtained
When t is an independent variable and x and y are dependent variables, it is approximated by a second-order piecewise polynomial and the second order is used until the approximation accuracy falls within a predetermined value.
Wherein between the piecewise polynomial junctions while increasing the number of dimensions adjacent repeated a least squares approximation straight line, an arc, a data approximation mechanism B approximated by piecewise polynomials, every group of (b10) point sequence compressed data storage mechanism for storing the parameters of the function approximating the one coordinate and the adjacent <br/> junction of the junction and be more of, stored in a second image store compressed data of the (c) And a compression / synthesis / combination / output device for reading / dividing a character / figure. 4. A first method for storing data obtained by optically reading a character / graphic figure in correspondence with a finite number of pixels arranged vertically and horizontally.
And (b) the stored image data is separated by one pixel at the boundary.
Data for data compression by sequential function approximation for the divided image divided divided into basic sub-area manner overlapping One
Includes a compression device, and a second image memory for storing the data compressed (c) data, the data compression device for said divided image (b), for (b1) dividing the image, the graphic character outlines (B2) a contour point sequence storage mechanism that stores the two-dimensional coordinates (X, Y) of the extracted contour line for each continuous group; and (b3) a contour line for each group. The X and Y coordinates of the point sequence are approximated by a quadratic piecewise polynomial with t as an independent variable and x and y as dependent variables, and least square approximation is repeated until the approximation accuracy falls within a predetermined range. A data approximation mechanism A for obtaining an approximate polynomial of (b4); a curvature calculation mechanism for obtaining a curvature at each point of a point sequence for each group in the x, y space from the approximation result; and (b5) a curvature calculation mechanism for each group. A perfect circle extraction mechanism for extracting perfect circles from data; (b6) perfect circles Infinity from curvature data of contour point sequence excluding
Or a temporary joint position extraction mechanism for extracting a point having a curvature greater than or equal to a certain value as a temporary joint point. (B7) Joint points corresponding to adjacent corner points form corner points.
The corner points are temporarily set according to the condition that they
The optimum junction extraction mechanism for determining the optimum junction from the junction point candidate in the vicinity of the junction, (b8) is a two or more arc start point to the straight line or a continuous two or more consecutive, not on the division boundary line optimal junction also the required junction removing mechanism to remove it found unnecessary junctions such approximation accuracy is maintained without it out of, between the (b9) adjacent the junction of the sequence of points the same group Straight line,
When approximated by an arc of the order which at a given approximation accuracy is not obtained
When t is an independent variable and x and y are dependent variables, it is approximated by a second-order piecewise polynomial and the second order is used until the approximation accuracy falls within a predetermined value.
Wherein between the piecewise polynomial junctions while increasing the number of dimensions adjacent repeated a least squares approximation straight line, an arc, a data approximation mechanism B approximated by piecewise polynomials, every group of (b10) point sequence compressed data storage mechanism for storing the parameters of the function approximating the one coordinate and the adjacent <br/> junction of the junction and be more of, stored in a second image store compressed data of the (c) A character / graphic divisional read / compression storage device characterized in that: 5. A method of optically reading character / graphic data.
Data that is stored in correspondence with a finite number of pixels
And (b) storing the stored image data.One pixel at the border
So that they overlapDivided into basic small areas and divided
Data compression for images sequentially by function approximationdata
compression(C) a second image storage for storing the compressed data;
Storage device; and (d) relating to the previous divided image stored in the second image storage device.
Playback that combines the data to be played and only the outline of the original image with a pen or blade
Equipment and including,  The data compression device (b) for the divided image extracts (b1) a contour of a character figure from the divided image.
(B2) two-dimensional coordinates (X, Y) of the extracted contour line
(B3) a contour point sequence storage mechanism for storing the contour point sequence for each groupX,YIndependent coordinates t
Variables, a quadratic piecewise polynomial with x and y as dependent variables
Similar, and repeat the least squares approximation until the approximation accuracy is within the specified range.
Data approximation to find an approximate polynomial for each group of repeated contour points
(B4) a point sequence for each group in the x, y space from the approximation result
And (b5) a perfect circle extraction for extracting a perfect circle from the curvature data for each group.
(B6)Outline point sequence excluding perfect circleFrom the curvature dataInfinity
Or has a curvature greater than a certain valueTemporarily connect points
A temporary junction position extraction mechanism for extracting as a joint point; (b7)Joint points corresponding to adjacent corner points form corner points
The corner points are temporarily set according to the condition that they
From candidate junctions near the junctionOptimum for finding the optimal junction
(B8) two or more continuous straight lines or two or more continuous circles
Approximate from the best joint point which is the starting point of the arc without it
Unnecessary joints that maintain accuracy are found and eliminated.
(B9) Same as the unnecessary joint removal mechanism to be removedGroup point sequenceNeighbors withinDoJunctionofA straight line between
Approximate in the order of circular arcs and this does not provide the required approximation accuracyWhen
ComeIs a quadratic classification where t is an independent variable and x and y are dependent variables.
Approximate with a polynomial and quadratic until approximation accuracy falls within a specified value
ofLeast-squares approximation while increasing the number of dimensions of piecewise polynomial
Repeat adjacentDoJunctionofStraight line, circular arc, piecewise
A data approximation mechanism B approximating with a polynomial;Do
JunctionofPressure that stores the parameters of the function that approximates
Compressed data is stored in the second image storage device (c).
Output contoursWhenWhen tracing the division boundary,
Or raise the blade so that the dividing border does not appear
Characterized figure / character division reading compression storage / synthesis output device
Place. 6. A first method of storing data obtained by optically reading a character / graphic figure in correspondence with a finite number of pixels arranged vertically and horizontally.
(B) a data compression device that divides the stored image data into basic small regions and sequentially compresses the divided images by function approximation, and (c) stores the data compressed. A data compression device (b) for the divided image, wherein: (b1) a contour line extraction mechanism for extracting a contour line of a character graphic from the divided image; and (b2) an extracted line. A contour point sequence storage mechanism for storing the two-dimensional coordinates (X, Y) of the contour line for each continuous group; and (b3) the X and Y coordinates of the contour point sequence for each group are represented by t as independent variables, x and a data approximation mechanism A for approximating by a quadratic piecewise polynomial with y as a dependent variable to obtain an approximate polynomial for each group of contour point sequences; and (b4) a point for each group in x, y space from the above approximation results. Curvature calculation mechanism for finding curvature at each point in a sequence , (B5) and circularity extraction mechanism for extracting a perfect circle from the data of the curvature of each group, (b6) infinity from the data of the curvature of the contour point sequence except circularity
Lines and junction position extraction mechanism, between the (b9) adjacent the junction of the sequence of points the same group for extracting a point having a curvature of constant value greater than or some is as junction,
When approximated by an arc of the order which at a given approximation accuracy is not obtained
Can independently a t variable, x, a straight line between the junction points adjacent approximated by piecewise polynomials as the dependent variable y, arcs, and the data approximation mechanism B approximated by piecewise polynomial, (b10) point sequence more becomes, the second image store compressed data of the said junction points of the coordinates for each group and the compressed data storage mechanism for storing the parameters of the function that approximates between adjacent <br/> junction (C) a character / figure divided read / compressed / stored / synthesized output device; 7. (a) Data obtained by optically reading a character or graphic
Data that is stored in correspondence with a finite number of pixels
(B) dividing and dividing the stored image data into basic small areas
Data compression of the divided images sequentially by function approximation
DoData compression(C) a second image storage for storing the compressed data;
Storage device; and (d) relating to the previous divided image stored in the second image storage device.
And a playback device that plays back the original image.
The data compression apparatus (b) for the divided image includes: (b1) extracting a contour line of a character / graphic figure from the divided image;
(B2) two-dimensional coordinates (X, Y) of the extracted contour line
(B3) a contour point sequence storage mechanism for storing the contour point sequence for each groupX,YIndependent coordinates t
Approximate by a piecewise polynomial with variables x and y as dependent variables
A Data Approximation Mechanism for Calculating Approximate Polynomials for Groups of Gut PointsA
And (b4) a point sequence for each group in the x, y space from the above approximation result
And (b5) a perfect circle extraction for extracting a perfect circle from the curvature data for each group.
(B6)Outline point sequence excluding perfect circleFrom the curvature dataInfinity
Or has a curvature greater than a certain valueJoin points
(B9) Same as junction point extraction mechanism that extracts as pointsGroup point sequenceNeighbors withinDoJunctionofA straight line between
Approximate in the order of circular arcs and this does not provide the required approximation accuracyWhen
ComeIs a piecewise polynomial where t is the independent variable and x and y are the dependent variables
Approximate by formula and adjacentDoJunctionofStraight line, circular arc, piecewise
Data approximation mechanism approximated by polynomialBAnd (b10) the coordinates of the junction point and the neighbors for each group of point sequencesDo
JunctionofPressure that stores the parameters of the function that approximates
Compressed data is stored in the second image storage device (c).
Character / graphic division reading characterized by being read
Compression storage synthesis output device. 8. (a) Data obtained by optically reading a character or graphic
Data that is stored in correspondence with a finite number of pixels
And (b) storing the stored image data.One pixel at the border
So that they overlapDivided into basic small areas and divided
Data compression for images sequentially by function approximationdata
compression(C) a second image storage for storing the compressed data;
Storage device; and (d) relating to the previous divided image stored in the second image storage device.
And a playback device that plays back the original image.
The data compression apparatus (b) for the divided image includes: (b1) extracting a contour line of a character / graphic figure from the divided image;
(B2) two-dimensional coordinates (X, Y) of the extracted contour line
(B3) a contour point sequence storage mechanism for storing the contour point sequence for each groupX,YIndependent coordinates t
Approximate by a piecewise polynomial with variables x and y as dependent variables
A Data Approximation Mechanism for Calculating Approximate Polynomials for Groups of Gut PointsA
And (b4) a point sequence for each group in the x, y space from the above approximation result
And (b5) a perfect circle extraction for extracting a perfect circle from the curvature data for each group.
(B6)Outline point sequence excluding perfect circleFrom the curvature dataInfinity
Or has a curvature greater than a certain valueJoin points
(B9) Same as junction point extraction mechanism that extracts as pointsGroup point sequenceNeighbors withinDoJunctionofA straight line between
Approximate in the order of circular arcs and this does not provide the required approximation accuracyWhen
ComeIs a piecewise polynomial where t is the independent variable and x and y are the dependent variables
Approximate by formula and adjacentDoJunctionofStraight line, circular arc, piecewise
Data approximation mechanism approximated by polynomialBAnd (b10) the coordinates of the junction point and the neighbors for each group of point sequencesDo
JunctionofPressure that stores the parameters of the function that approximates
Compressed data is stored in the second image storage device (c).
Character / graphic division reading characterized by being read
Compression storage synthesis output device. 9. (a) Data obtained by optically reading a character or graphic
Data that is stored in correspondence with a finite number of pixels
And (b) storing the stored image data.One pixel at the border
So that they overlapDivided into basic small areas and divided
Data compression for images sequentially by function approximationdata
compression(C) a second image storage for storing the compressed data;
Storage device; and (d) relating to the previous divided image stored in the second image storage device.
Data to be combined and only the outline of the original
A data compression apparatus (b) for the divided image, wherein (b1) extracting a contour line of a character / graphic figure from the divided image
(B2) two-dimensional coordinates (X, Y) of the extracted contour line
(B3) a contour point sequence storage mechanism for storing the contour point sequence for each groupX,YIndependent coordinates t
Approximate by a piecewise polynomial with variables x and y as dependent variables
A Data Approximation Mechanism for Calculating Approximate Polynomials for Groups of Gut PointsA
And (b4) a point sequence for each group in the x, y space from the above approximation result
And (b5) a perfect circle extraction for extracting a perfect circle from the curvature data for each group.
(B6)Outline point sequence excluding perfect circleFrom the curvature dataInfinity
Or has a curvature greater than a certain valueJoin points
(B9) Same as junction point extraction mechanism that extracts as pointsGroup point sequenceNeighbors withinDoJunctionofA straight line between
Approximate in the order of circular arcs and this does not provide the required approximation accuracyWhen
ComeIs a piecewise polynomial where t is the independent variable and x and y are the dependent variables
Approximate by formula and adjacentDoJunctionofStraight line, circular arc, piecewise
Data approximation mechanism approximated by polynomialBAnd (b10) the coordinates of the junction point and the neighbors for each group of point sequencesDo
JunctionofPressure that stores the parameters of the function that approximates
Compressed data is stored in the second image storage device (c).
By raising the pen or blade at the dividing line.
Is not output.
Split read compression storage synthesis output device. 10. (a) a first image storage device for optically stored data read graphic character to correspond to a finite number arranged pixels vertically and horizontally, 1 at the boundary of the image data (b) storage Without pixel
Data for data compression by sequential function approximation for the divided image divided divided into basic sub-area manner overlapping One
A compression device; (c) a second image storage device that stores the data that has been compressed; and (d) reproduction that reproduces the original image by synthesizing data relating to the pre-divided image stored in the second image storage device. A data compression device (b) for the divided image, wherein: (b1) a contour extraction mechanism for extracting a contour of a character graphic from the divided image; and (b2) a two-dimensional extracted contour. A contour point sequence storage mechanism for storing t as an independent variable and x and y as dependent variables for each continuous group of coordinates (X, Y); and (b4) each point of a point sequence for each group in x, y space. (B5) a perfect circle extraction mechanism for extracting a perfect circle from the curvature data for each group; and (b6) an infinity from the curvature data of the contour point sequence excluding the perfect circle.
Lines and junction position extraction mechanism, between the (b9) adjacent the junction of the sequence of points the same group for extracting a point having a curvature of constant value greater than or some is as junction,
When approximated by an arc of the order which at a given approximation accuracy is not obtained
Can independently a t variable, x, junctions approximation accuracy approximated by piecewise polynomials as the dependent variable y is adjacent repeated a least squares approximation with increasing number of dimensions of the piecewise polynomial to within a predetermined value a data approximation mechanism for approximating a straight line, an arc, piecewise polynomial between the parameters of the function that approximates between <br/> junction adjacent to the junction points of the coordinates for each group of (b10) point sequence Wherein the compressed data is stored in the second image storage device (c).