JPH11144051A - Filtering method filtering device and storage medium stored with filtering program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To perform a filtering processing data in natural and accurate form by filtering the data by using filter characteristics after impulse pesponse components after a 2nd code section and before a 4th code section among 1st to 4th code section are absorbed. SOLUTION: Four sample points A to D read out by a sample point read means are supplied to a curve interpolation processing part of an interpolation processing means, the part between the sample points B and C is set as a section where in curve interpolation processing is performed, and interpolation data are found as to interpolation points obtained by dividing the part between the sections B and C at equal intervals to a virtual time base. Then the curve interpolation processing part multiplies the supplied sample point data by specific interpolation coefficient data. Then the curve interpolation processing part adds the respective sample point data multiplied by the specific interpolation coefficient data to perform interpolation processing between respective sample points, and output the result as print data.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a filtering method, a filtering device, and a storage medium storing a filtering program suitable for a printer device for printing large characters and images on, for example, hanging banners, banners, posters, and the like.

[0002]

2. Description of the Related Art Today, for example, banners, banners,
2. Description of the Related Art A printing apparatus that prints large characters and images on a poster or the like in full color is known. For example, when printing a character, the printing device reads the character to be printed using a scanner device or a camera device, and extracts the outline of the character based on the character image data formed thereby, Based on the extracted character outline, FIG.
A plurality of sample points P1 separated by a predetermined interval as shown in FIG.
ＰPn and store this in a data file. The plurality of sample points P stored in this data file
1 to Pn, a spline curve or a Bezier curve (B
ezier) to perform printing by interpolating between the sample points P1 to Pn.

[0005] Interpolation using a spline curve is shown in FIG.
As shown in (a), when interpolation is performed between a sample point A serving as a start point and a sample point F serving as an end point, sample points B to
A curve in consideration of E, for example, a curve is formed such that the sample points A and B pass through the middle of each of the sample points BE.

[0006] The interpolation by the Bezier curve is performed as shown in FIG. 15B from a sample point A which is a start point to a sample point D which is an end point.
In the case of interpolating up to, a curve is formed so as to pass through the sample points B and C between them and finally connect the sample points A and B. In the interpolation using the Bezier curve, a sample point located on a curve connecting the sample points A and B is called an "anchor point", and a curve is formed so as to pass through the anchor point at the time of interpolation. Is done. Further, for example, "control points" indicated by points E and F are provided at predetermined positions with respect to the "anchor points", and by operating the positions of the control points, a dotted line or one point in FIG. The shape of the curve can be changed as shown by the chain line.

In this manner, several points (the sample points) serving as references for forming the outline of a character or the like to be printed are detected.
At the time of actual printing, by interpolating between some reference points using the above-described interpolation method,
The number of data actually stored for printing can be greatly reduced, and high-quality characters and the like can be printed with a small amount of data.

[0006]

However, the above-described interpolation method using the spline curve or the Bezier curve is good in that the number of data actually stored for printing can be greatly reduced. However, there is a problem that a complicated operation is required.

In an interpolation method using a spline curve, as shown in FIG. 15A, a curve connecting the sample points A and F always passes through intermediate sample points B to E. It is not always formed. For this reason, there has been a problem that the error between the form of characters and the like read by a scanner device and the like and the form of characters and the like actually printed becomes large.

In the interpolation method using the Bezier curve, as shown in FIG.
Is formed so as to pass through the intermediate sample points B and C. However, even when the control point is operated to make a partial change, the effect may be exerted on the entire curve. There was a problem that partial changes were difficult.

The present invention has been made in view of the above problems, and a curve passing through each sample point can be formed by a simple filtering operation, and a partial curve can be easily formed without affecting the entire curve. It is an object of the present invention to provide a filtering method, a filtering device, and a storage medium storing a filtering program that can be changed.

[0010]

According to the filtering method of the present invention, in order to solve the above-mentioned problem, the peak level of the main pulse of an infinitely continuous impulse response is determined in the direction in which time on the time axis advances. A first code section which is a section until the level becomes 0, a second code section which is a section from the time when the pulse level which has become the 0 level in the first code section becomes a minus level to the level 0, and A third code section, which is a section until the peak level of the pulse becomes 0 level in the direction in which the time on the time axis returns, and the pulse level that has become 0 level in this third code section is from the minus level to the 0 level. 4th section, which is the section until
In the code section, data filtering processing is performed using filter characteristics that have absorbed the impulse response components after the second code section and before the fourth code section.

Further, in order to solve the above-mentioned problem, the filtering device according to the present invention has a peak level of a main pulse of an infinitely continuous impulse response becomes zero level in a time direction on a time axis. , A second code section in which the pulse level at which the 0 level is changed from the negative level to the 0 level in the first code section, and a peak level of the main pulse. Is a third code section which is a section until the time on the time axis returns to the 0 level in a direction in which the time returns, and a pulse level from the negative level to the 0 level when the pulse level which has become the 0 level in the third code section. Filtering of filter characteristics in a form of a fourth code section that absorbs an impulse response component after the second code section and before the fourth code section. With a stage.

Further, a storage medium according to the present invention stores a computer program capable of realizing the filtering method according to the present invention in order to solve the above-mentioned problems.

[0013]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Preferred embodiments of a filtering method, a filtering device and a storage medium storing a filtering program according to the present invention will be described below in detail with reference to the drawings.

The storage medium storing the filtering method, the filtering device and the filtering program according to the present invention can be applied to, for example, a printing system as shown in FIG.

A printing system according to an embodiment of the present invention includes, for example, a reading device 1 such as a scanner device or a camera device for reading characters and images as originals, and data for characters and images to be printed. Communication between the stored recording medium driver 2 such as a magneto-optical disk drive and data for externally transmitted data such as characters and images to be printed and / or an interpolation processing program for data such as characters and images. And a line 3.

Further, the printing system is provided as a large output for printing on a large object such as a banner or a poster, and a small output for printing on a relatively small object such as a postcard size or A4 size. Based on the sublimation type printer 7 and the laser printer 8 and data such as characters and images supplied through the reading device 1, the recording medium driver 2 or the communication line 3, directly or through the network 5. It has a plotter 6, a sublimation printer 7, and a computer device 4 for controlling printing of the laser printer 8.

The storage medium storing the filtering method, the filtering device and the filtering program according to the present invention is applied to the computer device 4 in the printing system. That is, the computer device 4 is provided with, for example, a hard disk as a storage medium. By executing an interpolation program of vector data stored in the hard disk, the filtering method and the filtering device according to the present invention are executed. , An interpolation process of each vector data such as a character, an image or a sound is executed (it is obvious that each data such as a character, an image or a sound is vector data having a size and a direction). is there).

The vector data interpolation program is stored in advance on a hard disk provided in the computer device 4, or may be transmitted from an external device via the communication line 3 via the Internet, for example. In such a case, the computer device 4 temporarily stores the interpolation program of the vector data transmitted through the communication line 3 on a hard disk or the like, reads out the program, and executes the interpolation program.

The storage medium for storing the vector data interpolation program may be other than a hard disk or another storage medium such as a magneto-optical disk, an optical ROM disk, a floppy disk, or a semiconductor memory. .

Further, the vector data interpolation program can be applied to, for example, audio data in addition to character data and image data. However, in order to prevent the description from being difficult to understand, the vector data interpolation program will be described below. The description of the interpolation program will be made, for example, assuming that it processes character image data.

Next, FIG. 2 is a machine block diagram schematically showing a mode in which the computer device 4 executes according to an interpolation program of vector data recorded on a recording medium. At this time, not only the interpolation program of the vector data but also a part thereof can be constituted by an electronic circuit.

As can be seen from the machine block diagram shown in FIG. 2, the computer device 4 is supplied with character image data (vector data) via the reading device 1, the recording medium driver 2, or the communication line 3. The input terminal 11 removes noise from the character image data when the character image data is supplied as digital data, and digitizes the noise when the character image data is supplied as an analog signal. And a contour extracting means 13 for extracting a contour of the character based on the character image data from the binarizing / noise removing means 12.

The computer device 4 samples a plurality of (minimum four) sample points by sampling at predetermined intervals along the outline of the character extracted by the outline extracting means 13 and extracts the extracted sample points. If there is an edge or a corner on the outline of the character, a sample point extracting means 14 for extracting sample points of the edge or the corner, and a sample for storing each sample point extracted by the sample point extracting means 14 Each of the sample points stored in the point storage means 15 and the sample point storage means 15 is read out, and an edge or a corner on the outline of the character is detected from the situation of the surrounding sample points and stored in the sample point storage means 15. Edge / corner detection means 16 for giving a flag to an edge or corner sample point
A.

When the edge / corner detecting means 16A detects an edge or a corner on the outline of the character, the computer 4 determines the interpolation method of the edge or the corner as described later. At the same time, an end / corner interpolation method judging means 16B for setting a virtual sample point corresponding to an end or a corner as necessary and storing the virtual sample point in the sample point storage means 15; Sampling point reading for sampling the sampling points of the section to be interpolated along the data and the sampling points before and after the sampling points (including the virtual sampling points if there are virtual sampling points) from the sampling point storage means 15 in total. Means 17.

Further, the computer device 4 calculates each sample along the contour of the character based on each sample point read by the sample point reading means 17 and an interpolation coefficient stored in an interpolation coefficient table 35 shown in FIG. Curve interpolation processing section 18a between points
And an output terminal 19 for outputting interpolation data processed by the interpolation processing unit 18 by performing a linear interpolation by a linear interpolation processing unit 18b when there is an edge or a corner on the contour of the character. And

Next, the operation of the printing system having the above-described configuration according to the embodiment will be described.

First, in FIG. 1, the reading device 1
The character is read from the document on which the character to be printed is represented, and is supplied to the computer device 4 as character image data. Alternatively, the storage medium driver 2 reproduces the character image data stored in the storage medium in the form of character image data in advance, and reproduces the character image data.
To supply. Further, the computer device 4 captures the character image data transmitted from the external device via the communication line 3. When the character image data is supplied as described above, the computer device 4 performs data processing of the character image data according to the function shown in the functional block diagram of FIG.

That is, the data processing function will be described in terms of hardware. The character image data is supplied to the binarizing / noise removing means 12 via the input terminal 11.
When the character image data is digitally supplied, the binarization / noise removing unit 12 supplies the character image data to the contour extracting unit 13 as it is, or regenerates the character image data based on, for example, a system clock. The character image data is sampled and converted into a data processing rate in the computer device 4 and supplied to the contour extracting means 13. When the character image data is supplied in analog form, the character image data is sampled based on, for example, a system clock, converted into digital data, and supplied to the contour extracting means 13.

The outline extracting means 13 extracts the outline of the character based on the character image data, for example, as shown in FIG. 3A (edge detection), and supplies the outline data to the sample point extracting means 14. I do.

The sample point extracting means 14 extracts a plurality of sample points 1 to 19 along the outline of the character based on the outline data, for example, as shown in FIG. More specifically, if the outline of the character is formed by a gentle curve, the sampling point extracting unit 14 determines the sampling point 1 in FIG.
In the case where the interval between the sample points is made large as shown in FIGS. 4 to 4 and the outline of the character is formed by a steep curve, each of the sample points 6 to 9 in FIG. Sample points are extracted at intervals according to the outline of the character, such as by reducing the interval between sample points. Thereby, an error between the outline of the character read by the reading device 1 and the outline of the character to be actually printed can be suppressed as small as possible.

Here, the sampling point extracting means 14 is provided as shown in FIG.
As shown in (a) to (c), when extracting a plurality of sample points based on the contour data displayed two-dimensionally on the XY plane, an XY plane displaying the contour data two-dimensionally is extracted from this. The plurality of sample points are extracted by being divided into a virtual Xt plane and a virtual Yt plane that are orthogonal to each other. Note that this "t"
Represents a virtual time axis, for example. And FIG.
As shown in (b), a plurality of sample points are set on the virtual Xt plane obtained by sampling the contour data displayed on the XY plane along the virtual time axis t at predetermined intervals (equal intervals) in the data input order. As shown in FIG. 3C, a virtual X sampled from the contour data displayed on the XY plane at a predetermined interval (equal interval) in the data input order along the virtual time axis t.
A plurality of sample points are set on the y plane. Sampling point extraction means 1
4 shows a plurality of sample points on the virtual Xt plane and a virtual Xy
A plurality of sample points on the plane are stored and controlled in the sample point storage means 15.

As will be described later, when the outline of the character extracted from the character image data is formed by a gentle curve as shown in FIG.
The curve interpolation unit 18a performs the curve interpolation.

Also, the outline of a character or an image may not be formed by a curve, but may be an end or a corner as shown in FIG. In such a case, the sampling point extracting means 14
The edges and corners are always extracted as sampling points. In particular,
For example, when the character is the character of "ru", FIG.
The part circled by the dotted line corresponds to the end or the corner of the "R" character. In such a case, the sampling point extracting means 14
Is designed to always extract a portion corresponding to the end or the corner as a sampling point.

Next, the sample point storage means 15 reads out a plurality of sample points corresponding to the virtual Xt plane stored in this way and supplies them to the end / corner detection means 16A. This end
When there is an edge or a corner on the contour data among a plurality of sample points corresponding to the virtual Xt plane, the corner detection means 16A determines the sample of the end / corner from the surrounding state of the plurality of sample points. The points are detected, and the sample point storage means 15 is controlled so as to add a flag to the end / corner sample points on the contour data. Similarly, for a plurality of sample points corresponding to the virtual Yt plane, it is detected whether or not there is an end / corner sample point on the contour data, and the end / corner sample on the contour data is detected. Sample point storage means 1 to add a flag to a point
5 is controlled.

Next, the edge / corner detecting means 16A detects that there are edge / corner sample points on the contour data for a plurality of sample points corresponding to the virtual Xt plane and the virtual Yt plane. In this case, the edge / corner interpolation method determining means 16 determines the curve interpolation processing unit 18 of the interpolation processing means 18 for the vicinity of the end / corner.
It is determined on the basis of an interpolation processing program whether to execute “curve interpolation processing based on virtual sample points” in “a” or to execute “linear interpolation processing” in the linear interpolation processing unit 18b.
These “curve interpolation processing based on virtual sample points” and “linear interpolation processing” will be described later, but if it is determined that “curve interpolation processing based on virtual sample points” is to be performed, the A virtual sample point to be described later is set, and this virtual sample point is stored in the sample point storage means 15.

Next, the sample point reading means 17 reads out from the sample point storage means 15 a total of four sample points including the sample points in the section where the curve interpolation is performed along the contour data of the character and the sample points before and after the section. When there are edges and corners on the outline of the character, the sample points in the section where the curve interpolation is performed and virtual sample points set near the edges and corners as necessary are stored in the sample point storage unit 15.
From the sample point storage means 15, and controls reading from the sample point storage means 15 in a section where linear interpolation is performed when there is an edge or a corner on the outline of the character. At this time, since the sections in which the curve interpolation is performed are sequentially changed, four sample points are sequentially read from the sample point storage unit 15 for each section in which the curve interpolation is performed.

Here, in the computer device 4 of the printing system, when the sample point reading means 17 reads out a plurality of sample points near the section where the curve interpolation is performed, a plurality of sample points corresponding to the virtual Xt plane and the virtual Yt plane are read. From among
As shown in FIG. 5, sample points B and C in a section where curve interpolation is performed.
And a total of four first to fourth sample points A to D consisting of sample points A and D before and after having a correlation with these sample points B and C.
D is read out and the minimum required four sample points A to D are used.

Further, a plurality of interpolation points i0 to i9 are set by dividing the interval between the sample points B and C in the section subjected to the curve interpolation at equal intervals on the virtual time axis as shown in FIG. . Then, as described later, the interpolation coefficients stored in the interpolation coefficient table 35 (see FIGS. 2 and 6) are used for a plurality of interpolation points i0 to i9 set between the sample points B and C in the section where the curve interpolation is performed. Thus, the curve interpolation processing section 18a of the interpolation processing means 18 performs the curve interpolation processing.

Next, the interpolation processing means 18 performs a curve interpolation process by the curve interpolation processing unit 18a based on the sample points A to D read by the sample point reading means 17. That is, when performing the “curve interpolation processing”, the interpolation processing means 1
8 replaces each sample point data with a sample point on the virtual Xt plane and a sample point on the virtual Yt plane, respectively, and generates a “curve” based on the sample points on the virtual Xt plane and the sample points on the virtual Yt plane formed by the replacement. Interpolation processing ".

More specifically, for example, sample points 1 to 1 corresponding to a part of the character image data as shown in FIG.
9, the interpolation processing means 18
Detects the level of the virtual Xt plane at each of the sample points 1 to 19. As described above, each sample point is formed at an interval corresponding to the outline of the character. When performing this replacement, as shown in FIG. Are arranged on the virtual time axis t in the order of the number of the sample points (in the order of the number of the sample points) so as to have the same interval, and the replacement with the virtual Xt plane is performed. Similarly, the interpolation processing means 18 detects the level of the virtual Yt plane of each of the sample points 1 to 19. And
As shown in FIG. 3 (c), replacement with the virtual Yt plane is performed by arranging them on the virtual time axis t (in the order of sample point numbers) so that the intervals between sample points are the same.

When the replacement of each sample point with the virtual Xt plane and the virtual Yt plane is completed, the interpolation processing means 18 calculates the sample point and the virtual Yt of the replaced virtual Xt plane.
Each of the sample points on the plane is subjected to a filtering process as a "curve interpolation process" to convert discrete sample point data into continuous sample point data, and supply this to the plotter 6 or the like as print data. .

As described above, when performing the “curve interpolation processing”,
Regarding each sample point as continuous, equally spaced sample data,
These are replaced with a virtual Xt plane and a virtual Yt plane, respectively, and the sampling points on the virtual Xt plane and the sampling points on the virtual Yt plane are subjected to a filtering process to obtain continuous data,
By drawing this on a plane, a more natural smooth curve can be printed.

Next, the specific operation of the "curve interpolation processing" between sample points other than the end and corner portions, which is performed by the interpolation processing means 18, will be described.

FIG. 6 is a functional block diagram corresponding to each of the above-described curve interpolation processing programs executed by the interpolation processing means 18. This functional block diagram shows an interpolation filter of vector data in the interpolation processing means 18, and first to first delays input sample point data by one sample.
It has third delay means 22 to 24 and, for example, an interpolation coefficient table 35 having first to fourth interpolation coefficient memories 31 to 34 each storing ten interpolation coefficient data.

The interpolation processing means 18 is connected to the input terminal 21
A first multiplier 25 for multiplying the sampling point data supplied via the first interpolation coefficient memory 31 by the sampling point data supplied from the first interpolation coefficient memory 31; A second multiplier 26 that performs multiplication processing on the interpolation coefficient data read from the second interpolation coefficient memory 32;
A third multiplier 27 for multiplying the sample point data from the second delay means 23 by the interpolation coefficient data read from the third interpolation coefficient memory 33, and a sample point from the third delay means 24 A fourth multiplier 2 for multiplying the data by the interpolation coefficient data read from the fourth interpolation coefficient memory 34
8 and an adder 20 that forms and outputs print data as interpolation data for interpolating between sample points by performing an addition process on the multiplied outputs from the multipliers 25 to 28. .

The respective interpolation coefficients stored in the respective interpolation coefficient memories 31 to 34 of the interpolation coefficient table 35 are determined and stored as follows.

That is, as a basic method of interpolating between each sample point, it is known to filter each sample point data using an interpolation filter having an impulse response in the form of “sinx / x”. However, this form of interpolation filter is an infinite series, and FIG.
As shown in (b), the code interval T from the peak point of the main pulse.
The level becomes "0" for each c (code section), which is an infinitely continuous impulse response, and therefore, there is a problem that it takes time to calculate the filtering.

As shown in FIG. 7 (a), the interpolation processing means 18 goes to the + time side from the peak point of the main pulse in the infinitely continuous impulse response, and its level becomes 0 level. During this period (hereinafter, this section is referred to as a first code section), from the time when the level of the main pulse becomes 0 level, it proceeds further to the plus time side (+) until the level returns to 0 level. (Hereinafter, this section is referred to as a second code section), from the peak point of the main pulse to the minus time side (-) until its level becomes 0 level (hereinafter, this section is referred to as a third code section). ), And from the time when the level of the main pulse becomes 0 level until the level further returns to the -time side and returns to the 0 level (hereinafter, this section is referred to as a fourth code section). In the second An interpolating filter having a filter characteristic that absorbs the impulse response after the signal section and the impulse response before the fourth code section is provided (in the first to fourth code sections, sections other than these sections are provided). An interpolation filter having a filter characteristic that absorbs the impulse response is provided.)

Specifically, this interpolation filter, when four sample points A to D of the same level are arranged as shown in FIG.
Between the sample points B and C, which are the central sample points, the level (A1 shown by a dotted line in the figure) of the second code section of the filter characteristic corresponding to the sample point A and the sample point B The level of the first code section (B shown by a solid line in the figure)
1), the level of the third code section corresponding to the sample point C (C1 shown by a dashed line in the figure), and the level of the fourth code section corresponding to the sample point D (two-dot chain line in the figure). D shown
The filter characteristic curve is formed such that the values 1 and 2 are respectively flattened as shown in FIG.

The shape of the curve of the filter characteristic of the interpolation filter based on such conditions can be formed innumerably according to the purpose and the like. For example, the shape of the curve when printing characters is shown in FIG. 9 can be formed. In FIG. 9, a curve indicated by a solid line indicates a filter characteristic (fine curve) suitable for printing a sharp character (square character) having a thin line width, and a curve indicated by a dotted line indicates a character having a thick line width and a round character (circular character) Characteristics suitable for printing on paper (thick curve)
Is shown.

Each of the interpolation coefficient memories 31 to 34 shown in FIG. 6 stores an interpolation coefficient reflecting this filter characteristic. That is, the first interpolation coefficient memory 31 stores the interpolation coefficient corresponding to the fourth code section of the filter characteristic curve shown in FIG. 9, and the second interpolation coefficient memory 32 stores the interpolation coefficient corresponding to the filter characteristic curve shown in FIG. The interpolation coefficient corresponding to the code section No. 3 is stored in the third interpolation coefficient memory 33, the interpolation coefficient corresponding to the first code section of the filter characteristic curve is stored in the fourth interpolation coefficient memory 34, and the filter coefficient is stored in the fourth interpolation coefficient memory 34. An interpolation coefficient corresponding to the second code section of the characteristic curve is stored.

More specifically, the first interpolation coefficient memory 31 stores interpolation coefficients k00 to k09 for 10 samples formed by, for example, sampling the curve in the fourth code section at equal intervals. The second interpolation coefficient memory 32 stores the 10th curve formed by sampling the curve of the third code section at equal intervals, for example.
The interpolation coefficients k10 to k19 for the samples are stored respectively. The third interpolation coefficient memory 33 stores an interpolation coefficient k20 for 10 samples formed by, for example, sampling the curve of the first code section at equal intervals.
To k29, respectively, and the fourth interpolation coefficient memory 34 stores eleven samples of interpolation coefficients k30 to k40 formed by sampling the curve of the fourth code section at equal intervals, for example. It is remembered.

More specifically, such interpolation coefficients k00 to k40 are, for example, as shown in FIG. 10 for a thick curve (solid line curve in FIG. 9) and for a thin curve (FIG. 9).
Are respectively programmed in C language or the like. In FIG. 10, interpolation coefficient data A1 to U1 indicate interpolation coefficient data for a thick curve, and interpolation coefficient data A2 to U2 indicate interpolation coefficient data for a fine curve.

The interpolation coefficient data A1 among the interpolation coefficient data A1 to U1 which are interpolation coefficient data for a thick curve.
The interpolation coefficient data k20 to k29 corresponding to the first code section is programmed as to J1, and the interpolation coefficient data k30 to k40 corresponding to the second code section are programmed as the interpolation coefficient data K1 to U1. It has become.

Similarly, among the interpolation coefficient data A2 to U2 which are the interpolation coefficient data for the fine curve, the interpolation coefficient data A
The interpolation coefficient data k20 to k29 corresponding to the first code section are programmed as 2 to J2, and the interpolation coefficient data k30 to k40 corresponding to the second code section are programmed as the interpolation coefficient data K2 to U2. It has become so.

As can be seen from FIG. 7 and FIG.
The curves in the code section and the second code section, and the curves in the third and fourth code sections are symmetrical with respect to the level axis. Therefore, the interpolation coefficient data actually programmed includes interpolation coefficient data k20 to k20 of the first code section and the second code section.
40 are programmed, and the interpolation coefficient data k20 to k4 of the first code section and the second code section are used during the interpolation processing of the third code section and the fourth code section.
The interpolation process is performed by inverting the sign of 0. As a result, the number of interpolation coefficient data to be actually programmed can be reduced to 、, and the amount of stored data can be reduced.

Next, the operation of the "curve interpolation processing" by the curve interpolation processing section 18a of the interpolation processing means 18 having an interpolation filter having such characteristics will be described with reference to the functional block diagrams of FIGS. This is performed using the processing diagram and the flowchart of FIG.

First, as shown in FIG. 11, the four sample points A to D read by the sample point reading means 17 are supplied to the curve interpolation processing section 18a of the interpolation processing means 18, and as described above. A section between the sample points B and C is set as a section for performing the curve interpolation processing, and a plurality of interpolation points i0 to i obtained by dividing the sections B to C for performing the curve interpolation processing at equal intervals on the virtual time axis.
It is assumed that interpolation data is obtained for No. 9. It is assumed that the curve interpolation processing unit 18a of the interpolation processing means 18 sequentially reads out the interpolation coefficients of the interpolation coefficient table 35.

In step S1, the curve interpolation processing section 18a
Confirms that the sample point data has been supplied, and proceeds to step S2. In step S2, the curve interpolation processing unit 1
8a multiplies the supplied sample point data by predetermined interpolation coefficient data, and proceeds to step S3. In step S3, the curve interpolation processing unit 18a performs a curve interpolation process between each sample point by adding each sample point data multiplied by a predetermined interpolation coefficient data, and outputs this as print data. To step S4. Then, in step S4, it is determined whether or not the curve interpolation processing of all sample point data has been completed. If not completed, the process returns to step S1 to execute the above-described routines.
When the processing has been completed, all the routines in the flowchart of FIG. 12 are completed, and the curve interpolation processing is completed.

More specifically, the curve interpolation processing section 18a of the interpolation processing means 18 executes the curve interpolation processing with the configuration shown in the functional block diagram of FIG. The sampling point data is input terminal 2
1 to the first multiplier 25 and the first delay means 22. First to third delay means 2
2 to 23 are, for example, 1 for the supplied sample point data.
A delay corresponding to the number of sample points is applied. Therefore, for example, the sample point A shown in FIG.
When the sample point data of sample points D are sequentially supplied, the fourth multiplier 28 receives the sample point data of the sample point A, and the third multiplier 27 outputs the sample point data of the sample point B. However, the sampler data of the sample point C is supplied to the second multiplier 26, and the sample point data of the sample point D is supplied to the first multiplier 25.

When the sample point data is supplied to the multipliers 25 to 28 in this manner, the fourth interpolation point i0 to i9 obtained by dividing the sample points B and C for performing the curve interpolation in the order of the fourth point. Interpolation coefficient data k00 to k09 are sequentially read from the first interpolation coefficient memory 31 corresponding to the code section and supplied to the first multiplier 25, and the second interpolation coefficient memory 32 corresponding to the third code section. From the interpolation coefficient data k10 to k19
Are sequentially read and supplied to the second multiplier 26, and the first
, The interpolation coefficient data k20 to k29 are sequentially read from the third interpolation coefficient memory 33 corresponding to the code section of, and supplied to the third multiplier 27.
From the interpolation coefficient memory 34 of the interpolation coefficient data k30 to k4
0 is sequentially read and supplied to the fourth multiplier 28.

Each of the multipliers 25 to 28 performs a multiplication process on the supplied sample point data and each interpolation coefficient data sequentially read out from each of the interpolation coefficient memories 31 to 34, and adds the multiplication data for each multiplication process. Is supplied to the vessel 20 (step S2).

Thus, in each of the multipliers 25 to 28, a plurality of correction points i0 between the sample points B and C for performing the curve interpolation are provided.
Four multiplication data corresponding to four sample points A to D are formed for i to i9, and are supplied to the adder 20.

More specifically, sample points B and C for performing curve interpolation
The final correction data as an arbitrary correction point i (i is an arbitrary positive number from 0 to 9) is obtained, for example, as shown by a double circle in FIG. That is, the correction data of the correction point i is as follows: correction data = {Kd (i, 2) + Kd (i, 1) −Kd (i, 0) −Kd (i, 3)} Takes a value between the value of sample point B and the value of sample point C.

At this time, the curve interpolation processing unit 18a calculates each of the sample points on the virtual Xt plane shown in FIG. 3B and the sample points on the virtual Yt plane shown in FIG. Then, the above-described curve interpolation processing method is applied. The interpolation data obtained for each sample point on the virtual Xt plane and each sample point on the virtual Yt plane are supplied to and stored in the interpolation data storage means 41 via the input terminal 19 shown in FIG. Thereafter, a pair of interpolation data corresponding to the virtual Xt plane and the virtual Yt plane is read from the interpolation data storage means 41, and the XY
Inverse transformation is performed so as to remove the virtual time axis t corresponding to the plane, and XY plane interpolation data is synthesized. As a result, it is possible to obtain corrected contour data obtained by smooth curve correction in the XY plane form shown in FIG. Thereafter, the obtained corrected contour data is supplied as an output of the computer device 4 shown in FIG. 1 via the network 5 or directly to the plotter 6, the sublimation type printer 7, the laser printer 8, etc. It will be used for printing posters or postcards.

Next, returning to FIG. 2, the interpolation processing in the case where the character image data described above with reference to FIG. 4A has edges or corners will be described.

When the edge / corner detecting means 16A detects an edge or a corner on the contour data of the character image data, the vicinity of the edge or the corner has a correlation with each other as described above. Since it is difficult to extract four sample points, a special interpolation process is required depending on the end or the corner.

Therefore, the edge / corner interpolation method judging means 16
B indicates that “a linear interpolation process” or “virtual sample” is performed on the corner of the character when the edge / corner detection means 16A detects the edge or the corner on the outline data of the character image data. Of the curve interpolation processing based on points, the interpolation processing of one of these is determined so as to perform the interpolation processing of the programmed one. The two interpolation processes may be selected by the user.

First, the edge / corner interpolation method determining means 16B
Is determined to be "linear interpolation processing", the linear interpolation processing unit 18b of the interpolation processing means 18
The sample points corresponding to the ends and corners are read from among the sample points via the. Then, a plurality of sample points adjacent to the sample points corresponding to the ends and corners are read, and linear interpolation is performed between the sample points corresponding to the ends and corners and the plurality of adjacent sample points. To generate a plurality of interpolation points.

Specifically, as shown in FIG. 4 (b), the sample point corresponding to the corner of the character "R" is the sample point B. Adjacent sample points A and C are detected, and a plurality of interpolation points are linearly interpolated between sample points B and A and between sample points B and C, respectively. Occur. Note that the same processing is performed for the end portion. This makes it possible to linearly interpolate between a sample point corresponding to an end / corner and a sample point adjacent thereto.

Next, edge / corner interpolation method determining means 16B
Is determined as "curve interpolation processing based on virtual sample points", the curve interpolation processing unit 18a of the interpolation processing means 18
Reads out sample points corresponding to the ends and corners from among the sample points via the sample point reading means 17 and extracts a plurality of sample points adjacent to the sample points corresponding to the ends and corners. Then, the linear distance between each adjacent sample point and the sample point corresponding to the end / corner is detected. Further, when a straight line is drawn from each sample point through a sample point corresponding to an end / corner, a straight line distance from the sample point to a sample point corresponding to an end / corner, and A virtual sample point is formed at a position where the linear distance from the sample point corresponding to the edge / corner is the same as the distance from the sample point (the linear distance from the sample point to the sample point corresponding to the edge / corner, The linear distance from the sample point corresponding to the part / corner to the virtual sample point is the same distance.) Then, based on the linear distance from the virtual sample point to the corresponding sample point, a plurality of interpolation points are used to interpolate between the sample points corresponding to the ends and corners and the corresponding sample points using an optimal curve. Occurs.

Specifically, in this example, FIG.
As shown in (c), since the sample point corresponding to the corner of the character "R" is the sample point B, the curve interpolation processing unit 18a calculates the sample points A and C adjacent to the sample point B. Then, the linear distance between the sample points A and B and the linear distance between the sample points C and B are detected. And sample point B
Are formed as virtual points at the same linear distances as the respective linear distances, and the linear distance from the virtual sample point K1 to the sample point A and the virtual sample point K2 to the sample point Based on the linear distance to C, a plurality of interpolation points are interpolated between the sample points B and A corresponding to the corners and between the sample points B and C with an optimal curve. Occur. Note that the same processing is performed for the end portion. This makes it possible to perform a curve-like interpolation process between a sample point corresponding to an end and a corner and a sample point adjacent thereto.

As is apparent from the above description, the printing system according to the present embodiment can interpolate between each sample point with a simple configuration of only multiplication and addition. In addition, this interpolation processing replaces each sample point with a virtual Xt plane and a virtual Yt plane, and then interpolates between sample points on the virtual Xt plane and between sample points on the virtual Yt plane. Thus, more natural smooth curve interpolation can be performed.

Further, since the interpolation points are calculated based on, for example, four sample points, the printed curve can always pass through each sample point. An error between a read character or the like and an actually printed character or the like can be minimized.

Even when the shape of the curve between the sample points is changed, it is only necessary to change the value of the interpolation coefficient data to a desired value. The inconvenience that a temporary change affects the entire curve can be prevented.

The description of the above embodiment is merely an example of the present invention. For this reason, the present invention is not limited to this embodiment, and it is needless to say that various modifications can be made according to the design and the like within a range not departing from the technical idea according to the present invention. .

[0077]

According to the filtering method according to the present invention, the impulse response components in the first to fourth code sections after the second code section and before the fourth code section are calculated. Since the data filtering process can be performed using the absorbed filter characteristics, the data filtering process can be performed in a more natural and accurate manner.

In addition, in the filtering method according to the present invention, a predetermined number of data are sequentially multiplied by each filtering coefficient for each code section when filtering data. , Each multiplication output is added, and the addition output is output as a filtering output, so that the filtering process can be performed so as to always pass through the sampling point of each data, and the filtering operation is multiplied. In addition, it can be realized by simple arithmetic processing of only addition processing. Also, since each of the first to fourth code sections has a filtering coefficient used for filtering processing, it is possible to easily realize a partial change of the filter characteristic by changing the value of this coefficient. can do.

The filtering method according to the present invention as defined in claim 5, wherein the first and second code sections or the third and third code sections
When either one of the filtering coefficients of the fourth code section is used and the filtering coefficient of the first and second code sections is used, the filtering coefficient corresponding to the third and fourth code sections is used. The above multiplication process is performed by inverting the polarities of the filtering coefficients in the first and second code sections, and when the filtering coefficients in the third and fourth code sections are used, the first and second code sections are used. At the time of the corresponding filtering process, the polarity of the filtering coefficients in the third and fourth code sections is inverted and used to perform the multiplication process. Therefore, the amount of the filtering coefficients actually stored is reduced to half. Can be reduced.

Next, the filtering device according to the present invention as set forth in claim 6 to claim 10, is characterized in that, in the first to fourth code sections, the impulse response components after the second code section and before the fourth code section are obtained. Data can be filtered using the filter characteristics in the form of absorbing
Data filtering can be performed in a more natural and accurate manner.

In addition, the filtering apparatus according to the present invention, when filtering data, sequentially multiplies a predetermined number of data by each filtering coefficient for each code section. , Each multiplication output is added, and the addition output is output as a filtering output, so that the filtering process can be performed so as to always pass through the sampling point of each data, and the filtering operation is multiplied. In addition, it can be realized by simple arithmetic processing of only addition processing. Also, since each of the first to fourth code sections has a filtering coefficient used for filtering processing, it is possible to easily realize a partial change of the filter characteristic by changing the value of this coefficient. can do.

A filtering apparatus according to a tenth aspect of the present invention uses the filtering coefficient of one of the first and second code sections or the third and fourth code sections. When the filtering coefficient of the second code section is used, the polarity of the filtering coefficient of the first and second code sections is inverted and used during the filtering process corresponding to the third and fourth code sections. When the multiplication process is performed and the filtering coefficients of the third and fourth code sections are used, the filtering coefficients of the third and fourth code sections are used in the filtering process corresponding to the first and second code sections. Since the multiplication process is performed by inverting the polarity, the amount of filtering coefficients to be actually stored can be reduced to half.

Since the storage medium storing the filtering program according to the present invention described in claim 11 stores the filtering program described in any one of claims 1 to 5, The same effects as those of the filtering method according to the first to fifth aspects of the present invention can be obtained.

[Brief description of the drawings]

FIG. 1 is a block diagram of a printing system according to an embodiment to which a filtering method, a filtering device, and a storage medium storing a filtering program according to the present invention are applied.

FIG. 2 is a functional block diagram of a computer device provided in the printing system of the embodiment.

FIG. 3 is a diagram for explaining an interpolation form of an interpolation processing means provided as one function of the computer device.

FIG. 4 is a diagram for explaining an operation when the interpolation processing means performs an interpolation process on an end portion and a corner portion of a character or the like;

FIG. 5 is a diagram for explaining an operation in which the interpolation processing means extracts a plurality of sample points along a contour of a character or the like.

FIG. 6 is a functional block diagram of the interpolation processing means.

FIG. 7 is a diagram for explaining filter characteristics of an interpolation filter provided as one function of the interpolation processing means.

FIG. 8 is a diagram for explaining filter characteristics of the interpolation filter.

FIG. 9 is a diagram for explaining filter characteristics for square characters and filter characters for round characters of the interpolation filter.

FIG. 10 is a diagram showing an example of a program for configuring the interpolation filter.

FIG. 11 is a diagram for explaining a curve interpolation processing operation in the interpolation processing means.

FIG. 12 is a flowchart for explaining a curve interpolation processing operation in the interpolation processing means.

FIG. 13 is a block diagram for explaining an operation when obtaining one piece of contour data from interpolation data formed by separately interpolating a virtual Xt plane and a virtual Yt plane.

FIG. 14 is a diagram for explaining a conventional interpolation method using a Bezier curve.

FIG. 15 is a diagram for explaining a conventional interpolation method using a spline curve.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... Reading device, 2 ... Storage medium driver, 3 ... Communication line, 3 ... Plotter 4 ... Computer device, 5 ... Network, 7 ... Sublimation printer 8 ... Laser printer, 12 ... Binarization / noise removing means, 1
3 ... Contour extraction means 14 ... Sample point extraction means, 15 ... Sample point storage means, 16A
... End / corner detecting means 16B ... End / corner interpolation method judging means, 18 ... Interpolation processing means, 20 ... Adders 22-24 ... First to third delay means, 25-28 ... First
-Fourth multiplier 31-34 ... interpolation coefficient memory, 35 ... interpolation coefficient table

Claims (11)

[Claims] A first code section in which a peak level of an infinitely continuous main pulse of an impulse response becomes a 0 level in a time direction on a time axis;
The second code section, which is a section from the time when the pulse level that has become 0 level in the first code section becomes a minus level to the 0 level, and the peak level of the main pulse, in the direction in which the time on the time axis returns. A third code section, which is a section until the level becomes 0, and a fourth code section, which is a section from the negative level to the 0 level of the pulse level which has become 0 level in the third code section, A filtering method, characterized by performing a data filtering process using a filter characteristic that absorbs an impulse response portion after a second code section and before the fourth code section. 2. The filter characteristic, when filtering processing is performed on four consecutive first to fourth data at the same level, the level of the second code section of the first data,
The level obtained by adding the level of the first code section of the second data, the level of the third code section of the third data, and the level of the fourth code section of the fourth data is the second level. 2. The filtering method according to claim 1, wherein the filtering characteristic is such that a filtering process is performed between the second and third data with data of the same level as the second and third data. 3. When filtering data, a predetermined number of each data is sequentially multiplied by each filtering coefficient for each code section, each multiplied output is added, and the added output is filtered. The filtering method according to claim 1, wherein the filtering method is output as an output. 4. When filtering data, the first to fourth four consecutive data are sequentially multiplied by a filtering coefficient of the second code section for the first data, Is sequentially multiplied by the filtering coefficient of the first code section, the third data is sequentially multiplied by the filtering coefficient of the third code section, and the fourth data is The multiplication processing is sequentially performed on the filtering coefficients of the fourth code section, the multiplication output sequentially formed by the multiplication processing is added for every four, and this is output as a filtering output. The filtering method according to any one of claims 1 to 3. 5. A filtering coefficient of one of the first and second code sections or a filtering coefficient of one of the third and fourth code sections, and a filtering coefficient of the first and second code sections is used. In the case, when performing the filtering processing corresponding to the third and fourth code sections, the multiplication processing is performed by inverting and using the polarity of the filtering coefficient of the first and second code sections. When the filtering coefficient of the code section is used, the multiplication processing is performed by inverting the polarities of the filtering coefficients of the third and fourth code sections during the filtering processing corresponding to the first and second code sections. The filtering method according to claim 3, wherein the filtering method is performed. 6. A first code section in which a peak level of a main pulse of an infinitely continuous impulse response becomes a 0 level in a time direction on a time axis,
The second code section, which is a section from the time when the pulse level that has become 0 level in the first code section becomes a minus level to the 0 level, and the peak level of the main pulse, in the direction in which the time on the time axis returns. A third code section, which is a section until the level becomes 0, and a fourth code section, which is a section from the negative level to the 0 level of the pulse level which has become 0 level in the third code section, A filter device comprising filtering means for filtering filter characteristics in a form absorbing an impulse response portion after a second code section and before the fourth code section. 7. When the filtering means performs a filtering process between four consecutive first to fourth data at the same level, the level of the second code section of the first data, the second data The level of the first code section of
The level obtained by adding the level of the third code section of the third data and the level of the fourth code section of the fourth data becomes the second and third data between the second and third data. 7. The filter device according to claim 6, wherein the filter device has a filter characteristic for filtering with data at the same level as the data. 8. A coefficient storage means for storing a plurality of filtering coefficients corresponding to each of the first to fourth code sections, and a predetermined number of each data, Multiplying means for sequentially multiplying each of the filtering coefficients, and adding each of the multiplied outputs from the multiplying means,
8. An adding means for outputting the added output as a filtering output.
The filter device according to any of the preceding claims. 9. The multiplying means of the filtering means,
The first to fourth four consecutive data are sequentially multiplied by the filtering coefficient of the second code section read from the coefficient storage means for the first data, and the second data is multiplied. The first value read from the coefficient storage means.
Are sequentially multiplied by the filtering coefficients of the code section, and the third data are sequentially multiplied by the filtering coefficients of the third code section read from the coefficient storage means. Sequentially multiplies the filtering coefficients of the fourth code section read from the coefficient storage means, and the addition means adds the multiplication outputs sequentially formed by the respective multiplication processing every four, 9. The filter device according to claim 8, wherein is output as a filtering output. 10. The coefficient storage means stores, as a filtering coefficient, a filtering coefficient of one of the first and second code sections or the third and fourth code sections. Means for, when filtering coefficients for the first and second code sections are stored in the coefficient storage means, performing filtering for the first and second code sections corresponding to the third and fourth code sections; When the multiplication process is performed by inverting the polarity of the filtering coefficient of the section and the filtering coefficient of the third and fourth code sections is stored in the coefficient storage unit, the first and second code sections are stored. 10. The filter according to claim 8, wherein the filtering process is performed by inverting and using the polarity of the filtering coefficient in the third and fourth code sections. The other apparatus. 11. A storage medium storing a filtering program for executing the filtering process according to claim 1. Description: