JPS6218883A - Image processor - Google Patents

Abstract

PURPOSE:To perform image conversion without spoiling features of a current input image by removing small area images present in an input image and reducing the quantity of converted codes, and also converting the small area images which are removed to the color of images of other areas that the image contacts. CONSTITUTION:The small area image is determined when the number of picture elements of an area encircled with the same color is <=N. When plural areas 12-16 are present, picture elements of each area are counted when there is an unprocessed image area, picture elements in the area are counted; when the counted number (n) is <=N, the area is regarded as the small area image and outer periphery tracking and conversion color specification are carried out. Consequently, when there are small area images 14 and 16 among plural areas 12-16 which are present in an indicated area 11 and has a specific color, those areas are detected and removed automatically.

Description

[Detailed description of the invention] The invention will be explained in the following order.

A. Field of industrial application B. Overview of the invention C. Prior art D. Problem to be solved by the invention E. Means for solving the problem (Fig. 1) F. Effect G. Example G1 Description of Videotex image creation device (Fig. 2) Explanation of G2 small area image removal processing (Figs. 3 to 7) 03 Same as above (Fig. 8 and following) Effects of the H invention A Industrial application field This invention is applicable to, for example, a videotex image creation device. The present invention relates to an image processing apparatus having a small area image removal function suitable for performing data reduction processing before encoding into videotex image data.

B. Summary of the Invention The present invention relates to an image processing device that is suitable for use in a videotex image creation device, for example, when performing data reduction processing before encoding into videotex image data, and particularly relates to an image processing device that is suitable for applying data reduction processing before encoding to videotex image data. When removing an image, this small area image can be removed without damaging the characteristics of the current input image by converting the area to be removed into the color of the image of another area that this small area image is in contact with. It is.

C. Prior Art] Videotex, which is a two-way screen information communication system that combines computer technology and communication technology, is known.

There are several known videotex display methods, one of which is the NAPLPS method.

In this method, graphical information is expressed using geometrical graphical commands as described later, and this is encoded.

NAPLPS code can be broadly divided into control sets and graphic cents, and graphic cents are basic character sets such as alphabets, numbers, and symbols, and PDI, which is graphic description instructions.
Consists of cents, etc.

The feature of the NAPLPS method is this PDI cent, which divides the image components into 5 types: points, lines, arcs, rectangles, and polygons.
It is divided into two basic geometric figures, and the size, position, coordinates, and color are specified and displayed using numerical values.

This PDI cent has the following six types of drawing commands called geometric graph ink primitives.

POINT: Draw a point LINE: Draw a line ARC: Draw an arc 1? ECTANGLE: POLYGON that draws a rectangle
: INCREMENTAL drawing a polygon Image data drawn with a two-handed figure is code-converted from the original image to the figure code of this PDI set, its attribute code, control code, and code of the control set as time-series data in the order of drawing operations. When decoding, the codes are decoded in chronological order and the image is drawn and reproduced.

According to this NAPLPS method, the amount of image information to be transmitted can be significantly reduced compared to the method in which each dot is encoded and transmitted, and it is possible to transmit images with high transmission efficiency, and the position coordinate data is transmitted in the vertical direction. , is a relative coordinate in the horizontal direction, so it has the advantage that the image is displayed with a fineness that corresponds to the resolution of the terminal side.

D Problems to be Solved by the Invention By the way, when converting image data formed by digitizing an input image signal obtained from a video camera, VTR, etc. into a NAPLPS code as described above, a PDT code such as POLYGON or INCREMENTAL
When using drawing instruction codes such as , it is relatively troublesome to code the small area image data included in the image data, and the conversion code requires several bytes.
Encoding one input image signal requires a capacity of 30 to 60 bytes.

In order to solve these problems, it is preferable to limit the code amount of the completed image to 2 to 8 bytes. That is, it is necessary to perform some kind of preprocessing on the input image signal to reduce the amount of code, thereby converting it into image data that can be passed through NAPLPS encoding.

However, conventional image processing devices do not perform such data processing.

Therefore, the present invention solves these conventional problems, and when converting to videotex image data such as NAPLPS code, small area images existing in the input image are removed to reduce the amount of converted code. At the same time, the small area image to be removed is converted to the color of the image of the other area that this image is in contact with, making it possible to perform image conversion without damaging the characteristics of the current input image. be.

E. Means for Solving the Problems In order to solve the above-mentioned problems, the present invention is provided with a small area image removal means (2) as shown in FIG. The image data from which the small area image has been removed is converted into predetermined videotex image data.

F Effect In this configuration, the small area image removal means (2) performs the following data processing.

In other words, the outer periphery of the small area image existing within the area specified by the cursor is tracked, the color that appears most frequently on the outer periphery is detected, and the color of the small area image is set to one of the detected colors described above. The small area image is removed by converting it into .

When converting the color of a small region image to another color, the color of the image of the region adjacent to this small region image is converted, so image processing is possible without damaging the characteristics of the current input image, and this image processing This allows the amount of code for image data to be significantly reduced compared to the conventional method.

G Embodiment G1 Explanation of the Videotex image creation device 5 FIG.
AM12] is a small area image processing means, (3) is a display, (4) is a display controller, and (5) is a video RAM (coordinates that can specify the address of 11 according to the cursor position on the screen of display (3). input means,
(6) is buffer RAM, (7) is video RAM (
11 and buffer RAM (71).

Figure 2 shows an example of the device of the present invention applied to a Videotex image creation device. In this example, the device includes a frame data creation section (100) that converts input color image information into data for each screen, and It consists of a videotex device section (200) that creates videotex image data from data for each screen.

In the frame data creation section (100), (101)
is an analog color video signal input terminal, and an NTSC color video signal from a color video signal generation source (10) such as a video camera, VTR, or video disc is passed through this input terminal (101) to an NTSC decoder (102).
The three primary color signals R, G, and B are obtained from the three primary color signals R, G, and B.

Further, (103) is an input terminal for the three primary color signals R, G, and B, and the input terminal (103) is connected to the source (20) of the three primary color video signals R, G, and H from an image information creation device such as a microcomputer. Through the frame data creation section (1
00), these three primary color signals are input.

The three primary color signals from the NTSC decoder (102) and the three primary color signals from the input terminal (103) are selected and extracted by the input selection circuit (104), and the selected output is supplied to the A/D converter (105). be done. This A
In the /D converter (105), 3
The primary color signals R, G, and B are each converted into 4-bit data and one sample (one dot) is converted into a 12-bit digital signal. This 12-bit digital signal is supplied to a field memory (106) and written into this memory (106) one field at a time.

Writing and reading of this field memory (106) is performed by the Videotex equipment unit (200) as described later.
This is performed based on instructions from the microcomputer (201). Furthermore, switching of the input selection circuit (104) is similarly performed by a command from the Videotex device section (200). (108
) is a timing control circuit for controlling the timing.

Therefore, a signal in units of l field is obtained from the field memory (106), which is returned to the analog three primary color signals R9G and B by the D/A converter (107), and this is converted into the selector (231 ) is supplied to the monitor receiver (230), and the image is displayed on the screen.

A microcomputer (201) is provided in the Videotex equipment section (200).

That is, (202) is its CPU, (203) is a ROM in which a data processing program as described below is stored, and (204) is an R for the work area.
It is AM. Furthermore, (205) is a data bus.

Further, (211) is a buffer RAM for taking in digital data of 1 dot and 12 bits from the frame data creation section (100).

(212) is a converted digital video in which one screen worth of data loaded into this buffer RAM (211) is converted into data representing the color of each dot using a 16-color color bullet created as described below. A video RAM in which data is stored. In this case, this video R
In addition to the video data from the frame data creation unit (100), the AM (212) has a menu for displaying a plurality of commands on the screen as a menu table for performing the processing intended by the operator on this video data. Data is stored. Note that this command menu appears on the screen only when a command is selected with the tablet (216), and normally the data image is displayed on almost the entire screen.

(213) is a controller for this video RAM (212), and each is connected to a data bus (205).

(214) is a color bullet memory, which contains 16 color video data set as described later, part 3.
Data representing primary color signals R, G, and B each as a 4-bit digital signal is stored for 16 colors.

Further, (215) is a RAM that stores data converted from the converted digital video data stored in the video RAM (212) to a NAPLPS PDI code consisting of a geometric figure code and its attribute code.

(216'') is a tablet left as an external input device, and this tablet (216) is a coordinate input device for specifying a position on the screen of the monitor receiver (230).

This coordinate input device may be a so-called mouse or other input device.

When one command from the aforementioned command menu is selected on this tablet (216), that command is executed. That is, video RAM (2
12) can be specified with a stylus on this tablet (216), and the position on the screen specified with the stylus is displayed on the screen by a cursor. This tablet (216)
connects the data bus (221) via the interface (221).
05).

(217) is a floppy disk with RAM (2I5)
The NAPLPS code for each screen stored in the 222
) is connected to a data bus (205).

(218) is a modem, and the NAP of RAM (215)
It modulates the LPS code and transmits it, for example, through a telephone line, which also uses an interface (22
3) is connected to the data bus (205).

And (220) is a frame data creation unit (100)
An interface for importing one screen worth of digital video data from a data bus (
205), the video data supplied to the buffer RAM
(211) is first imported as raw data.

In the Videotex device configured as described above,
First, video data for one screen is transferred from the field memory (106) of the frame data creation unit (100) to the interface (220) in the Videotex device (200) according to the command selected on the menu by the tablet (216). The data is taken into the data bus (205) via the data bus (205).

The data captured in this way is stored in the buffer RAM (211
) and temporarily suspended.

Then, from the data taken into this buffer RAM (211), the frequency of appearance of each color for each dot is stored in the ROM (203).
), and the digital video data (12 bits) of the top 16 colors are written into the color palette memory (214) in order of frequency.

In this way, each dot of the raw data in the buffer RAM (211) is recolored by 16 color data automatically selected from the full screen image data stored in the color palette memory (214).

In this case, the information of each dot is stored in the buffer RAM (212).
The original color of the video data is converted to one of the 16 colors in the color bullet memory (214) that is closest to the original color of the video data, and the data in the conversion table is stored in the video RAM (214).
2) is written as data for each dot. That is, the data stored in this video RAM (212) is not 12-bit video data, but color bullet memory.
(214') is the address data (4 bits) of the video data of the color to be converted.

Then, when the tablet (216) selects a command to display this recolored image in the command menu displayed on the screen of the monitor receiver (230'), the selection circuit (231
) is switched to the opposite state from that shown. and,
The video RAM (
Address data for specifying the color of a given dot is read out sequentially in the horizontal and vertical directions from 212), and the selected color data (i.e., three primary color signals R, G, B) is read out from the color bullet memory (214) using the address data. is 4bee/
This signal is converted back to an analog signal by the D/A converter (233), and these analog three primary color signals R, G, B are sent to the monitor image ti (230) via the selection circuit (231). ), and the recolored image is displayed on the screen.

G2 Description of small area image removal processing The recoloring image data thus created is subjected to small area image removal processing according to the program of the computer (201) as follows.

In FIG. 2, (224) is a high-speed processing circuit for this processing, and (225) is a buffer RAM.

Here, the term "small area image" refers to a case where the number of picture elements in an area surrounded by the same color is N or less. Therefore, isolated images of small areas and noise are also included. N can be set arbitrarily by the user. In this example, it can be set arbitrarily within the range ■≦N≦64.

FIG. 3A shows the designated area (11) of the image data (indicated as input image @! area) specified by the cursor,
Among the plurality of regions (12) to (16) having a predetermined color existing in the designated region (11), a small region image (14),
(16), this region is automatically detected and removed. The input image after small area removal becomes as shown in FIG. 3B.

FIG. 4 is a flowchart showing an example of small image area removal processing.

When the control program starts and it is determined in step (20) that the menu for small area image removal processing has been selected, it is determined in step (21) whether or not to change the image removal area, that is, change the setting of N. If it is checked and there is a change, it is reset in step (23), and after the reset is completed or if there is no change in the setting, the processing area by the cursor is set in step (23).

Once the processing area is set by the cursor, the step (
In step 30), the small area image within the area surrounded by this cursor is removed.

This removal processing step (30) is structured as a subroutine as shown in FIG.

When the subroutine (30) is called, the unprocessed image @! Step (3)
1) is checked. For example, if there are multiple regions (12) to (16) as shown in FIG. 3A, the presence or absence of processing of the image region is checked for each region, and if there is an unprocessed image region, In step (32), the number of picture elements in the area is counted, and in step (33), the size of the counted number n is determined.

If the count number n is less than or equal to N, the area is regarded as a small area image, and outer circumference tracking processing and conversion color designation processing are executed in the subroutine of step (40). When a conversion color is specified, the color of the processing area is converted to the specified color in step (35), a processed flag is set in step (36), and then step (31) is performed.
The process returns to , and the same process is executed again until there are no more unprocessed areas.

FIG. 6 shows a subroutine (40) for outer circumference tracking processing and conversion color designation processing.

First, in step (41), a starting point for outer circumferential tracking in the processing area is determined. The starting point is set at the first coordinate of the processing area with respect to the horizontal and vertical scanning directions that it touches (see Figure 7).

Once this starting point P is determined, the number of identical colors in the four picture element areas on the upper, lower, left and right sides surrounding this starting point P is counted in step (42). When the count is finished,
The outer surface tracking point P moves in the direction indicated by the arrow in FIG. 7, and the colors at that point are counted (step (43)).

The outer tracking point P is checked in step (44) to see if it has returned to the starting point, and if not, it is returned to the starting point in step (44).
Returning to step 2), the colors of the outer periphery of the processing area are sequentially determined. When the outer tracking point P returns to the starting point, the color that appears most frequently among the colors surrounding the outer tracking point is determined as the designated color of the processing area (step (45)). Return to step (35) in FIG. 5. .

As a result, in step (35), the color of the processing area, that is, the small image area, is converted to the above-mentioned specified color.

Through this series of processing, the small area image is converted into a designated color, and the input image shown in FIG. 3A is converted into an image like that shown in FIG. 3B. That is, as a result, the small area image is removed by such image processing, and the color of the removed area is filled in with the above specified color.

Explanation of another example of G3 small area image removal process By the way, the eighth
As shown in Figure A, the small area image is located in another image area (17).
and (18), and the length of contact with one image area (I7) is longer than the length of contact with the other image area (18), the length of contact with image area (17) Since color is more dominant, the series of image processing described above results in processing as shown in FIG. 8B, resulting in a very unnatural processing result.

In such a case, it is preferable to process as shown in FIG.

In the removal process of a small area image as shown in C in the same figure, the color of one pixel (picture element at the processing point) in the small area image is changed to A for all colors existing at the k and l area boundary points. : Calculated as a function of the distance from the processing point to the boundary of the small area image, like a number, and calculate the pixel of the processing point from the size of the sum of the calculation data for each color executed for the colors existing at the area boundary point. All you have to do is decide on the color. If you convert using the calculated color,
Since regions approximate to each image boundary in FIG. 9 are newly found, image processing can be realized without damaging the characteristics of the current image even if the small region image is removed.

As shown in FIG. 10, the color of the processing point may be calculated from the distance of the intersection of the horizontal and vertical axes passing through the picture element of the processing point and the image area. Although the calculation method for the simplified form is more complicated than the process shown in Figure 9, the image processing is less unnatural than the process shown in Figure 8B, and the image processing is simpler than that shown in Figure 9. be done.

The image processing shown in FIG. 11 and subsequent figures is a specific example for realizing the processing shown in FIG.

The processing routine (30) for small area removal in this case is as shown in FIG. 11, and compared to the case in FIG. 5, only the subroutine of step (5o) is different. This subroutine (5o) is a processing step for determining the color of the processing point.

FIG. 12 shows an example of this subroutine (50). When the subroutine (50) is called, the process first moves to the processing step (subroutine configuration) (60) of creating a data table by tracking the outer circumference, and The color is determined at , and the distance between this picture element and the boundary of the small area image is calculated.

When the creation of the data table is complete, go to Step 7 (51).
), the presence or absence of an unprocessed point in the small area image is checked, and if an unprocessed point exists, the specified color of the pixel at the processed point is determined in step (70) using the above-mentioned data table. Ru. This step (6o) also has a subroutine configuration.

After determining the conversion color, in step (52), the color of the pixel at the processing point is converted to the specified color calculated in step (70) described above, and a processing flag is set.

When such scanning is completed, step (3) in FIG.
Return to 6).

In the subroutine (60) shown in FIG. 13, the color of the picture element at the processing point is designated by arithmetic processing. That is, in step (61), the outer circumferential tracking point P shown in FIG. 7 is set,
In step (62), the color of the outer circumference tracking point P and the coordinates of this outer circumference tracking point P are set in the data table.

Next, in step (63), the outer circumferential tracking point P is moved one position clockwise, and if it is determined in step (64) that the moving point is not the starting point of the outer circumferential tracking point, the process returns to step (62). The color and its coordinates at the outer circumferential tracking point P that has been moved back are set in the data table.

FIG. 15 shows an example of a data table, in which the color of the outer circumferential tracking point and the coordinates (X, Y coordinates) of that point are marked for each outer circumferential tracking point.

Such a data table creation process is executed at all the outer circumference tracking points, and when the outer circumference tracking point P returns to the starting point, the data table becomes an end of table, and the process returns to step (51).

FIG. 14 is an example of a processing step (70) in which the color of a pixel at a processing point is determined from the coordinates of the processing point and the coordinates of the outer tracking point using a data table.
), then step (72)
The presence or absence of data set in the data table is checked in step (73) if the data exists.
From the coordinates of the processing point and the coordinates of the outer tracking point, the distance between the processing point and the outer tracking point is calculated, and the ratio Z between this distance and the constant A is calculated.

This ratio Z is set in the histogram in step (74), and once set, the process returns to step (72), where the same process is executed for each outer circumferential tracking point and added to the data set in the above-mentioned histogram. . This addition is performed for each outer tracking point of the same color. Therefore, such a series of processes is a process for calculating the fl1 formula.

Now, since the ratio Z mentioned above indicates which image area the pixel at the processing point is close to, the calculation result of equation (1) for the color of each image area with respect to the pixel at the processing point is determined. By doing this, it is possible to know to which image area the color of the pixel at the processing point should be converted to the color of the current input image without impairing the characteristics of the current input image.

Therefore, after performing the above processing on all picture elements in the small image area, the color with the largest data among the histogram data (data in equation (1) for each color) is selected in step (75). It is used as the color of the pixel at that processing point.

For example, if we pay attention to an arbitrary picture element Qi in FIG.
The total sum of data in area (17) in picture element Qi is 71,
If the sum of data in area (18) is 22 and the sum of data in area (19) is 73, then this picture element Qi is in area (17).
), so Z 1 > 23 > 2
The relationship is as shown in 2. Therefore, the color of picture element Qi is the area (1
7) is specified as the color.

Such color conversion processing is executed for all picture elements included in the small area (14).

For this reason, in the case of an input image as shown in FIG. 8A, the small area image to the left of the broken line in FIG. Since the small area image on the right side will be filled in with the color of the image area (18), if the above-mentioned small area image removal process is performed, it will be converted into an image as shown in FIG.

Therefore, unnatural image processing as shown in FIG. 8B can be significantly improved.

The image data after the redundant data reduction process stored in the video RAM (212) as described above is further manually corrected by filling in unnecessary image parts by input scanning with the tablet (216), and the resulting image is The data is a microcomputer (201
NAPL consists of a figure code consisting of the above-mentioned five types of basic geometric figures and an attribute code indicating color, size, coordinate position, etc. according to the program in the ROM (203) of
It is encoded into a PS system code, and the encoded code is written into the RAM (215). Furthermore, the image data of this NAP1 and PS code is stored as a data image file on the floppy disk (217) through the interface (222) depending on the case.

In the above description, the present invention is applicable to a processing device that converts input image information into NAPLPS code, but it is also applicable to any image conversion process that describes an input image by dividing it into basic geometric figures. Therefore, in addition to the above, PRES
The present invention can be applied to videotex image processing devices such as , 5EPT, and CAPTEN systems.

As described in detail of the invention H, in this invention, when an input image is digitized and converted into predetermined videotex image data such as NAPLPS code, small objects existing in the image area specified by the cursor are extracted from the input image information. A region image is removed by taking into consideration the mutual relationship with the adjacent image region, and the color of the removed region is converted to correspond to the color of other image regions adjacent to this small region image. It is.

Therefore, according to the present invention, the amount of code when converting to NAPLPS code or the like can be significantly reduced compared to when such processing is not performed. In addition, since small area images can be removed without significantly impairing the characteristics of the current input image, image processing can be realized in a more natural state.

Therefore, the present invention is extremely suitable for use in preprocessing such as the above-mentioned NAPLPS code conversion.

[Brief explanation of the drawing]

FIG. 1 is a system diagram showing an example of an image processing device according to the present invention, FIG. 2 is a system diagram showing an example of the main part of the invention that further embodies FIG. 1, and FIG. An explanatory diagram showing an example of a removal process, FIGS. 4 to 6 are flowcharts showing an example of a small image area processing operation, FIG. 7 is an explanatory diagram of an outer circumference tracking operation, and FIGS. FIGS. 11 to 14 are explanatory diagrams for explaining other examples, and FIGS.
FIG. 5 is an explanatory diagram showing an example of a data table. In (1), the video RAM 121 is a processing means for removing small area images, (3) is a display, (5) is a coordinate input means, (14) and (16) are small area images, and P is an outer circumferential tracking point. . Hidemori Matsukuma ``1'' Figure 3 ^ Figure 3 B

Claims (1)

[Claims] An image signal output from a video camera or the like is digitally converted and then supplied to a small area image removal means, which eliminates the small area image existing within the area specified by the cursor. On the other hand, the small area image can be removed by tracking its outer periphery, detecting the color that appears most often on the outer periphery, and converting the color of the small area image to one of the detected colors. Image processing device.