JPH04356867A - Area recognizing system for recorder - Google Patents

Abstract

PURPOSE:To recognize a rectangular area and to reduce cost by recognizing the rectangular area with the forward scan of prescan, further recognizing the area again by back scan and deciding the inside/outside of the area. CONSTITUTION:Input/output processing parts 70-1 and 70-3 store AR color data separated by a color separation circuit 50 in a line buffer 60. Judgement processing parts 70-2 and 70-4 execute the forward scan of the prescan to the data in the buffer 60, judge the inside/outside of the area and store the result in a line memory 61. The memory 61 is read out, and the data in the preceding line and the data at the adjacent picture element are referred to and stored in a page buffer memory 62 after the judgement processing. Next, the back scan is executed, the memory 62 is read out and the judgement is executed again. Further, the AR color data in the buffer 60 are fetched by copy scan, and the judged result of the prescan is corrected. Thus, the rectangular area can be recognized, and the cost is reduced.

Description

[Detailed description of the invention]

0001

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an area recognition system for recording devices such as digital copying machines and facsimiles.

0002

2. Description of the Related Art The specific structure of an image reading system, particularly a scanning section, in an image processing apparatus such as a digital copying machine or a facsimile machine is shown in FIG. 24(a), for example.

[0003] This is a platen 1 on which a document 13 is placed.
A light source 15, a light guide member 16 including a rod lens array, etc., and a one-dimensional image sensor 10 such as a CCD are disposed below the image sensor 2, and together constitute a scanning section. Then, in the process of optically scanning the document by moving parallelly (in the direction of the arrow in the figure) with the platen cover 14 placed on top of the document 13, the scanner 13 is scanned in parallel with the amount of received light output from the image sensor 10. Based on the detection signals in units of cells, image information in units of predetermined pixels is generated corresponding to a gray scale image, line diagram, characters, etc. drawn on the document 13. In this way, the pixel information obtained by optically scanning and reading the original is sequentially subjected to correction processing, various pixel editing processing, etc. on a pixel-by-pixel basis, and then, for example, is printed or reproduced as an image on a display device.

[0004] In this type of image processing device, as shown in FIG.
As shown in , special processing, such as shading, coloring, etc., is performed only on the inner area Ei (shaded area) of the closed-loop image L that specifies one point inside the image drawn in the original document 13, as shown in FIG. There is a method that does this, but this includes:
An image closed loop area recognition function is required to determine whether the pixel whose pixel information has been read is inside or outside the currently specified closed loop L. Therefore, for example, the function to recognize the entire closed loop L area by specifying one point inside is the function to draw the closed loop with a marker and recognize the closed loop area (AR function: Area Recognition).
For example, as a more advanced function compared to the ADAR function (Advanced Area Recogni
tion), and has conventionally been realized as follows.

As shown in FIG. 25, the coordinate input device 18 (editor pad) has a plate-shaped input pad corresponding to each position in the document placement area of the platen 12, and register positions (registration positions on the platen 12) are provided. The original 13 is set on the paper (corresponding to 1), and in this state, an appropriate position on the input pad corresponding to the inner region of the closed-loop image L of the original 13 is specified and input by a pen touch with the editor pen 19. Then, information on the designated input position on the input pad is transmitted from the coordinate input device 18 to the image processing device, and the image processing device recognizes the designated position in the corresponding document placement permissible area of the platen 12. In this state, the original 13 is transferred from the coordinate input device 18 onto the platen 12, and the scanning system scans the original and stores the obtained image information (eg, dot data) in the page memory. Then, area determination is performed for each dot sequentially starting from the specified position on the page memory where the image dot data has been developed, and dots determined to be within the closed loop L are subjected to special processing, such as coloring processing (white color processing). (converting dots to black dots) is performed as processing on the page memory. Specifically, for example, CRT
C-LSI (for example, Hitachi product HD 6348
4) etc.

[0006]

[Problem to be solved by the invention] By the way, CRTC-
Those that use a CRT controller such as LSI,
The CPU writes data to the VRAM, sends a command to the CRT controller, and instructs it to fill in the area surrounded by the marker color in the VRAM.Based on that instruction, the CRT controller
The RT controller accesses the VRAM and performs calculations while viewing the data individually to fill in the closed loop.When the CPU sends a command to the CRT controller, it cannot do anything until the processing is completed. As a result, there is a problem that a lot of processing time is required, and there is also a problem that the cost increases because a highly versatile and expensive CRT controller is used.

[0007] In the conventional rectangle recognition method, the coordinates Xmin and Xm of the vertices of the rectangular area are also
The area is recognized by latching ax, Ymin, and Ymax and importing it as area data, but this method is limited in the number of areas, and also in the Y direction (main scanning direction).
There were drawbacks such as the inability to recognize separate areas in which multiple areas were lined up.

The present invention is intended to solve the above problems.The number of rectangular areas to be recognized is unlimited, a plurality of rectangular areas lined up in the Y direction can be recognized separately, and a hardware latch circuit is not required. An object of the present invention is to provide an area recognition device for a recording device that can realize rectangular area recognition and reduce costs by using most of the conventional closed-loop recognition circuit and switching the algorithm to rectangular recognition. shall be.

[0009]

[Means for Solving the Problems] As shown in FIG. 1, the present invention is a scan timing generating means which is controlled by a control means 1 and generates each scan timing of a forward scan, a back scan of a prescan, and a copy scan. 2
and rectangular area recognition means 3 that executes determination processing in synchronization with the scanning timing, and the rectangular area recognition means captures area recognition color data in a forward scan of pre-scan and performs area determination while storing the results in memory. 4, the memory 4 is read from the reverse direction again in the backscan of the pre-scan, and while determining the area, the result is saved and the rectangular area circumscribed to the closed-loop marker is recognized. The present invention is characterized in that the area recognition color data is read in again at the time, and the judgment result at the time of pre-scanning is corrected using the read data, so that a rectangular area circumscribing the closed-loop marker is recognized.

0010

[Operation] The present invention recognizes a rectangular area by the forward scan of the pre-scan, and also recognizes the rectangular area by the back scan to determine the inside/outside of the area. If the pre-scan determination result is corrected during the determination, it becomes possible to perform more precise area recognition.

[0011]

[Examples] Examples of the present invention will be described below.

FIG. 2 is a block diagram showing the overall configuration to which the present invention is applied.

In FIG. 2, reference numeral 10 denotes a one-dimensional full color sensor of the document scanning section, and reference numeral 20 converts the read signal output from the full color sensor 10 in a time-division manner cell by cell into color component data (green: G, blue) for each predetermined pixel. :B, red:R) and outputs them efficiently, and this full color sensor 10 and sensor interface circuit 20 constitute an image input section. 50
is a color separation circuit that generates density information and color information for each pixel from each color component data (GBR) from the sensor interface circuit 20, as well as marker color information for closed-loop designation. As a result, a sub color flag SCF corresponding to the sub color "red" and a main color flag MCF corresponding to the main color "black" are generated. Reference numeral 70 denotes an area recognition circuit, and this area recognition circuit 70 detects the closed-loop image on the document whose read pixels are specified by the marker based on the mark color information (ARCF1, ARCF2) output from the color separation circuit 50. It determines whether it is inside or outside and outputs the result for each pixel. The determination result here is ARDT1 in the first AR area indicating the inside of the closed loop area specified by the first AR color marker, and ARDT2 in the second AR area indicating the inside of the closed loop area specified by the second AR color marker. , outside area A indicating the outside of both the first AR area and the second AR area
There are three types of ROUT. 150 is a color separation circuit 50
A correction/filter circuit 160 performs various corrections and filter processes on the density information D and color information (SCF, MCF) from the correction/filter circuit 150; to enlarge, reduce,
This is an editing/processing circuit that performs editing and processing such as color inversion, and in this correction/filter circuit 150 and editing/processing circuit 160, the above-mentioned area recognition circuit 70 Judgment information from (A
It has a function of performing predetermined editing and processing processing only on the specified area based on RDFT1, ARDT2, AROUT).

As described above, the correction/filter circuit 1
The density information D and color information (SCF, MCF) that have undergone various processing in the image forming apparatus 50 and the editing/processing circuit 160 are provided to a specific image forming device via an interface circuit 230. The image forming equipment includes a laser printer 240 that reproduces two colors, and an image transmitter/receiver 270.
In addition, the density information D and color information are provided to the computer 280, stored in an auxiliary storage device (magnetic disk device, etc.) of the computer 280, and the information is used in various end devices. Aspects are also possible. When the laser printer 150 is connected, the entire system constitutes a two-color copying machine, and when the image transceiver 270 is connected, the entire system constitutes a facsimile machine.

A specific configuration of the editing/processing circuit 160 is shown in FIG. 3, for example. This example can be edited
The processing circuit 160 includes a color inversion circuit 161 that performs processing such as conversion between a main color (black) and a sub color (red), a whiteout circuit 190 that converts black characters into white characters, and a shading circuit 200 that performs shading processing. , negative/positive inversion circuit 2 that inverts the main color (black) or sub color (red) and white.
10, this editing/processing circuit 160
The density data D, sub color flag SCF, and main color flag MCF supplied pixel by pixel are serially processed in each circuit. Note that if the processing is not particularly necessary, the density data D is not processed in the corresponding circuit and becomes the data path as it is.

It is also possible, of course, to include other processing circuits for editing/processing in the editing/processing circuit 160. A CPU (not shown) is connected to each processing circuit (color inversion circuit 161, outline circuit 190, shaded circuit 200, negative/positive inversion circuit 210) of the editing/processing circuit 160.
, the first area setting signal (closed loop area specified by the marker of the first AR color) AR1, the second area setting signal (the second AR
A closed loop area designated by a color marker) AR2, and an outside area setting signal (outside area of both the first AR area and the second AR area) AROUT are given. Further, from the area recognition circuit 70, first area data ARDT1, second area data ARDT2, and out-of-area data AROUT are sent to each processing circuit. In addition, out-of-area data A
ROUT is generated as a signal that becomes active when both the first area data ARDT1 and the second area data ARDT2 become 0. In each processing circuit, when a region setting signal is input from the CPU, only the region set based on the region data from the region recognition circuit 70 is set, and data is passed for other regions.

FIG. 4 is a diagram showing the configuration of the area recognition circuit 70. The area recognition circuit 70 includes an input/output processing section 70-1 that performs input/output processing of first AR color data obtained through the color separation circuit 50, a determination processing section 70-2 that performs area determination, and an input/output processing of second AR color data. Input/output processing unit 70-
3. It has a determination processing section 70-4, and is capable of recognizing two-color markers. In addition, the area recognition circuit 70
Controlled by the CPU 30 via the U interface 75,
It includes an AR enable circuit 71, a line timing control circuit 72, a clock timing control circuit 73, a line counter 76, a determination address counter 74, and a page buffer address counter 77, which will be described later.

The input/output processing units 70-1 and 70-3 store the AR color data separated by the color separation circuit 50 in the line buffer 60, and perform input correction to be described later. Further, the determination processing units 70 - 2 and 70 - 4 determine whether the data written in the line buffer 60 is inside or outside the area in prescan and forward scan, and send the determination result to the address specified by the determination memory address counter 74. is stored in the judgment line memory 61. That is, in forward scanning, while sequentially reading out the contents of the judgment line memory 61, performing judgment processing on the judgment pixel by referring to the data one line before and the data one pixel adjacent, and storing the results in the page buffer memory 62. Store. Next, during the back scan of the pre-scan, the page buffer memory 6
Read the contents of 2 and judge again. Then, the content of the page buffer memory 62 is determined again using the AR color data obtained through the color separation circuit 50 during copy scanning,
The rough judgment results obtained during pre-scanning are finely corrected using copy scanning to perform precise area recognition.

FIG. 5 is a block diagram of the AR enable circuit of FIG. 4. In the present invention, the pre-scan makes a rough judgment and stores the result, and the pre-scan judgment result is corrected using the data read at fine intervals during the copy scan, so the data read during the pre-scan and copy scan is Correction cannot be made if the position changes. However, if the number of lines is counted as an address during copy scanning, the position will change depending on the magnification. Therefore, in the present invention, the speed is fast when the magnification is reduced, and it is slow when the magnification is enlarged, and the distance per clock is changed, and the total number of motor clocks does not change depending on the magnification. I am trying to use the sub-scanning address. Therefore, 1/
Pass the motor clock every 16 mm through the inverter 80,
The frequency is divided by 16 by a frequency divider 88 to obtain an address signal in units of 1 mm. Then, selector 81 selects scan 1 (forward scan of prescan) and scan 3 (copy scan) using a page synchronization signal (page sync, PS).
, 82, and the motor clock signal is latched with the signal obtained through the OR circuit 90 to obtain the sub-scanning address. In addition, the signal of scan 2 (backscan of prescan) is set to VV, which indicates the effective range of data.
The AD signal is latched by a latch circuit 91 to obtain a scan 2 signal.

Next, the outline of the area recognition algorithm will be explained. In the present invention, it is possible to perform area determination for two colors, the first AR color and the second AR color, and the AR mode is set by the CPU before prescanning starts. As shown in FIG. 6, the AR mode in the present invention is a free type mode that recognizes a closed loop area on the specified color image data and by the color image data, and a free type mode that recognizes the closed loop area on the specified color image data and surrounded by the color image data. It has four AR modes: a fill mode that recognizes the image data area, a rectangle mode that recognizes the circumscribed rectangular area of the specified color image data, and a one-point pointing mode that recognizes the image closed loop area that includes the specified color. The 2-bit data is (00) for free type mode and (01) for filled mode.
), rectangle mode is (10), one point instruction mode is (11)
The default is (00), and different AR modes can be set for the two colors, the first AR color and the second AR color.

FIG. 7 is a diagram for explaining the processing at the time of forward scan of pre-scan. The copy cycle starts with a forward scan of pre-scan, and when the carriage advances to the tip of the platen glass, it starts capturing AR color data and image data, and the page sync signal input to the area recognition circuit is based on this data capture. It starts up at the starting point (■ in Figure 7). From then on, each time the carriage advances a certain distance, the algorithm described below uses the AR color data and image data to determine whether the area for one line is inside or outside, and the results are saved in the page memory (see Figure 7). ■ area). The above processing is performed until the carriage reaches the rearmost position (for example, 432 mm for an A3 document platen glass) when data is taken in from the front end of the platen glass, or the page sync signal becomes LOW (■ in FIG. 7). As explained in Fig. 5, the carriage position is detected by counting the carriage motor clock after the page sync signal rises.In this copier, the accuracy of the page memory in the sub-scanning direction is 1 mm, and the carriage position is Every time the motor clock count reaches 1 mm, it is determined whether an area corresponding to one line in the main scanning direction is inside or outside.

FIG. 8 is a diagram for explaining processing at the time of backscan of prescan. During the backscan of the prescan, no image data is input, and the judgment results from the forward scan saved in the page memory are read from the rear edge of the platen glass, and the judgment is performed from the opposite direction from the forward scan. , save the result to page memory again. In this case, input data such as AR color data is not input, so when backscanning starts and the backscan signal rises, judgment starts immediately from the last line (431) (Fig. 8
■). After that, unlike the forward scan,
Judgment processing for each line is performed according to internal processing timing regardless of the position of the carriage (■ in FIG. 8). This process is continued until all lines have been processed (■ in FIG. 8). During backscanning, the determination process proceeds regardless of the position of the carriage, so as shown in the figure, the actual determination process ends earlier than the carriage returns to the tip position of the platen glass.

FIG. 9 is a diagram for explaining processing at the time of forward scan of copy scan. During copy scanning, capture of AR color data and image data starts from the leading edge of the document (the rising edge of the page sync signal).
For every two lines (VVAD) of input data, the inside/outside of the area is determined based on the AR color data and black image data and the determination result read from the page memory, and the result is output. Note that VVAD is a signal indicating an effective range as image data in the sub-scanning direction. Through such processing, the accuracy of the determination result during backscanning is as low as 1 mm, but since the reading during copy scanning is fine, correction is performed to smooth the contour of the area. In this case, the read line address of the page memory is different from the number of lines of input data, and it is necessary to count the carriage motor clock from the rise of the page sync signal and add 1 to the line address every 1 mm.

[0024]Furthermore, functions such as ``free registration'' which aligns the registration position with the leading edge of the original even when the original is placed in any position, ``movement'' which moves the registration to a specified position, and ``centering'' are selected. In this case, the rising edge of the page sync signal is not necessarily at the tip of the platen glass, which is the line address = 0, so the page sync The read line address of the page memory at the rising edge of the page sync signal (A
RSLA) from the CPU. Therefore, the read line address of the page memory is ARSLA,
The line addresses increase as ARSLA+1, ARSLA+2, etc., and such line addressing is continued until 432 is reached or the page sync signal becomes LOW.

Next, a basic determination algorithm will be explained. The basic algorithm is common to all modes: free type mode, fill-in mode, rectangle mode, and one-point pointing mode. The input data is judged from , and the judgment result is saved in the page memory. During backscanning of pre-scanning, the judgment results saved in the page memory are read out, judgments are made again from three directions: the main scanning direction, the anti-main scanning direction, and the anti-sub-scanning direction, and the judgment results are saved in the page memory. Finally, during the forward scan of the copy scan, final judgment is performed on the input data and the judgment results saved in the page memory from three directions: main scanning direction, anti-main scanning direction, and sub-scanning direction, and determines the inside/outside of the area. . In this way, in any scan, determination is basically made from three directions, and the determination from these three directions is performed in two determinations: a primary determination and a secondary determination, as described below.

FIG. 10 is for explaining the primary determination. First, primary determination is performed from two directions: the main scanning direction and the sub-scanning direction (or the anti-sub-scanning direction). That is, FIG.
As shown in (a), the primary determination result of one pixel adjacent obtained by latching the primary determination result and the determination line memory 61
Judgment is made for the judgment pixel based on the judgment result of the previous line read from the input data and the input data, and as shown in FIG. The inside/outside of the area is determined with reference to the next determination result, and the result is written into the determination line memory 61.

FIG. 11 is for explaining the secondary determination. The secondary determination adds information from the opposite main scanning direction to the result of the primary determination to determine the determination result for one line. In this case, there is no input data from the anti-main scanning direction, so the judgment line memory 6 saves the results of the primary judgment.
1 is read out again from the anti-main scanning direction in the opposite direction from when it was saved, thereby adding information from the anti-main scanning direction. In other words, as shown in FIG. 11(b), in order to make a determination based on the secondary determination result of one pixel adjacent to the determined pixel,
1(a), the primary judgment result read from the judgment line memory 61 and the latched secondary judgment result
Judgment is made based on the secondary judgment result of the adjacent pixel, and the result is written into the line memory for judgment.

Based on such a determination algorithm, at least in the main scanning direction, line memory reading, determination,
Writing to the line memory requires 3 VCLK (VCLK: a signal indicating the period in which data for one image is fixed) and 2 VVAD processing time for the primary judgment and secondary judgment in the sub-scanning direction. Considering simplification and memory cost, the accuracy of judgment/output is set to main scanning direction = 4VCLK,
The sub-scanning direction is set to 2VVAD. Therefore, assuming that the data for determination is 2 bits per color, the required capacity of the determination line memory 61 is 4672/4×2×2=116
Since it is 8 words x 4 bits, a 2K word x 8 bit SRAM is used. On the other hand, the precision of the page memory 62 for saving the prescan determination results is set to be equivalent to 8 VCLK in the main scanning direction and 1 mm in the sub-scanning direction in order to reduce costs. As a result, the required capacity of the page memory 62 is 46
Since 72/8 x 432 x 2 x 2 = 246K words x 4 bits, a 256K words x 4 bits DRAM is used.

The above-mentioned algorithms for primary determination and secondary determination basically include ``marker color pixels and non-density pixels (white pixels) adjacent to marker color pixels within a closed loop.''
However, the specific judgment result is "O"...
Outside the image closed loop, “M”…area containing the marker, “I”
...becomes inside the image closed loop. This determination result is handled within the system as status information, and finally pixels with status "I" are inside the image closed loop specified by the marker, and other pixels are outside the image closed loop.

Next, referring to FIG. 12, a case will be described in which the sub-scanning direction is M lines and the main scanning direction is determining the Nth pixel (target pixel). For example, in the forward scan of the prescan, the determination is made based on the image data (density data) of the target pixel, the marker color information, and the determination results of the adjacent pixels, as shown in FIG. In other words, the main scanning direction determination result (temporary status) of the previous pixel (N-1st) on the same line (M line)
The target pixel (shaded area in the figure) is determined based on the determination result of the same pixel (Nth) of the previous line (M-1). The determination result in the main scanning direction becomes the primary determination result and is stored in the memory as a temporary status (KS). After the determination of each pixel in the main scanning direction is completed in this manner, determination is made from the opposite side, that is, determination in the opposite main scanning direction. As shown in FIG. 12(b), the target pixel (shaded area) is determined based on the determination result of the previous pixel (N+1st) on the same line as the target pixel, and the determination result is the secondary determination result. The temporary status (KS) of the target pixel stored in the memory becomes the status (S
) can be rewritten into information. In this way, determinations from both the main scanning direction and the anti-main scanning direction are performed sequentially for each line, and the determination during backscanning is completed with determinations from line 0 to line 431.

[0031] At the time of backscanning of pre-scanning, determination is made regarding the previous pixel (N-1st) on the same line (M line) in the main scanning direction, as shown in FIG.
The main scanning direction judgment result (temporary status) and the previous line (
The target pixel is determined based on the determination result of the same position pixel (Nth) of M+1 line). The determination result in the main scanning direction becomes a temporary status (KS), and the determination result stored in the memory is rewritten to this temporary status (KS). After the determination of each pixel in the main scanning direction is completed, determination in the opposite main scanning direction is performed. As in the case of the forward scan described above, as shown in FIG. 12(d), the target pixel (shaded area) is determined based on the determination result of the previous pixel (N+1st) on the same line as the target pixel, and The determination result becomes the secondary determination result and becomes the final status (S) information.

With this determination method, the main scanning direction and anti-main scanning direction are determined during forward scanning and back scanning, respectively, so that one pixel is determined from four directions: top, bottom, left, and right. , accurate judgment results can be obtained.

As a result of the above determination, for example, FIG.
For the closed loop shown in a), the pixels on the marker have the status "M", and the pixels inside the area have the status "I".
”, but if it is outside the area, the status “O” will be given. As shown in FIG. 13(b), in the main scanning direction, for example, the pixel in the shaded area has the status "M" for both the pixel at the same position one line before and the pixel one pixel before. It is determined that the status is "I", and the status "I" is assigned to the subsequent pixels.
may be granted. In this case, as shown in FIG. 13(c), it is impossible for the first pixel in the anti-main scanning direction to be given the status "I", so this is rewritten to the status "O" to ensure a correct determination. It turns out.

Next, noise removal and connection correction will be explained. For example, in a single point instruction mode, it is undesirable to recognize even small points such as dust. Therefore, in order to recognize only those larger than a certain size as markers, for example, as shown in FIG. AND circuit 100 when continuous
Detect and remove noise. Of course, the number of dots is not limited to three, and it may be possible to detect two or more consecutive dots. Through such processing, it is possible to prevent objects with a diameter of 1 mm or less from being recognized as markers, for example. Furthermore, as shown in FIG. 14(b), the FIFOs 105 and 106 are also shifted one line at a time in the sub-scanning direction so that an output can be obtained when a plurality of consecutive lines occur. By doing this, it is possible to avoid picking up markers that are 1 mm or less in the sub-scanning direction, for example.

[0035] Also, free type mode, fill mode,
In the case of rectangular mode, if markers smaller than 1 mm are not allowed, thick markers must be prepared. Therefore, as shown in FIG. 14(c), in the main scanning direction, the latch circuits 111 and 112 shift one pixel at a time, and the OR circuit 110 performs the OR of three pixels, for example.
Regarding the sub-scanning direction, as shown in FIG. 14(d), the FIF
Shift one line at a time with O115 and 116, and move several lines of O
By taking R, it becomes possible to recognize even a thin marker of 1 mm or less. In this way, in the present invention,
The correction method is changed depending on the AR mode at the input/output section of the area recognition circuit, so that input correction suitable for the mode is performed.

By the way, as shown in FIG. 15(a), even if you intend to specify a closed loop area with a marker, if there is black image data under the marker, the AR color cannot be detected on the black image data, and as a result, As shown in FIG. 15(b), there are cases in which the marker is broken and an incomplete closed loop is formed, making it impossible to detect the region. Therefore, free type mode,
Figure 14(c) in case of fill mode and rectangle mode,
The disconnection of the AR color data is corrected by simply calculating the OR for 1 mm using the OR circuit in (d).

Next, the free type mode determination algorithm will be explained. In the free type mode, the input AR color and the closed loop region of the AR color are detected. When a plurality of AR colors are adjacent or intersect, the closed loop is detected using the outermost AR color data. For example, if there is AR color data as shown in FIG. 16(a), first, in a forward scan of the pre-scan, the area is roughly recognized and stored in the page memory as shown in FIG. 16(b). In this case, an area surrounded by AR color is detected when viewed from three directions: main scanning direction, anti-main scanning direction, and sub-scanning direction, but the area that is shaded by AR color when viewed from three directions is inside the area. I made a wrong judgment. However, during the backscan of the pre-scan, the judgment result at the forward time is read out from the page memory as described above, and the judgment as seen from the anti-sub-scanning direction is performed again to determine the inside/outside of the area. The misjudgment in the part is eliminated. However, since the judgment accuracy of prescan is rough on the order of 1 mm, it is not output as is,
By comparing this with the read data at fine intervals during the copy scan and making a determination based on this, it is possible to perform fine area recognition as shown in FIG. 16(d).

The operation rule for determining the free type mode is shown in FIG. 17 (
As shown in a), the status of the secondary judgment result one line before is “O” (outside the closed loop), “M” (on AR color), “I
” (in a closed loop excluding the top of the AR color) and the status “O”, “M”, and “I” of the primary judgment result of one pixel to the left,
As shown in FIG. 17(b), if the primary judgment result of one pixel adjacent is "O" and the judgment result of the same position one line before is "M", the judgment pixel is set as "O". As shown in (c), if the judgment result of one line before is "M" and the primary judgment result of one pixel next to it is "M", the judgment pixel is "I", and the judgment result of one line before is " I”, the primary judgment result of one pixel adjacent is “
In the case of “I”, the judgment pixel is “I”, the judgment result of one line before is “O”, and the primary judgment result of the next pixel is “O” or “
In the case of "M", the determination pixel is set to "O", and the status of the determination pixel is determined one by one.

The determination calculation for this free type mode is shown in FIG. 18(a).
) to (e) are used.

- At the time of forward scan of pre-scan 1
In the case of the next judgment, the AR color data = 1, that is, when the data on the marker comes in, the calculation table of Fig. 18(a) is used, and the status of the secondary judgment result one line before, the primary one pixel left The status of the judgment result is "O", "M", "I"
In either case, an operation is performed to set the status of the determined pixel to "M". In addition, when AR color data = 0, that is, when no marker color is detected, Fig. 18(b)
), the secondary judgment result of one line before is “O”,
Or, if the status of the primary judgment result one pixel to the left is “O”, the judgment pixel is set to “O”, and the status of the secondary judgment result one line before, and the primary judgment result one pixel left When the status is "M" or "I", the determination pixel is determined to be "I". Therefore, if one line before or one pixel to the left is outside the closed loop, the determination pixel is always outside.

Further, in the case of secondary determination, the calculation table shown in FIG. 18(e) is used. Since the secondary judgment is performed in the anti-main scanning direction, it is determined based on the status of the secondary judgment result of one pixel to the right and the status of the primary judgment result of the judgment pixel, and the status of the primary judgment result of the judgment pixel is " If it is "O", the judgment pixel is set as "O" regardless of the status of the secondary judgment result one pixel to the right, and if the status of the primary judgment result of the judgment pixel is "M", it is set as 1. It is set as "M" regardless of the status of the secondary determination result on the right side of the pixel. However, the status of the secondary judgment result of one pixel to the right is "O"
In this case, even if the status of the primary determination result of the determined pixel is "I", it is rewritten as "O".

- At the time of backscan of pre-scan, the judgment result stored in the page memory is read and the first judgment is carried out;
The next decision will be made. First, in the first judgment, if the content of the page memory is "M", the operation table shown in FIG. 18(a) is used, and if the content of the page memory is "I",
The calculation table b) is used, and if the page memory content is "O", the calculation table shown in FIG. 18(c) is used. Figure 18(a),
Regarding (b), it is the same as the case of forward scan of pre-scan, but in the calculation table of FIG. 18(c), when the content of page memory is "O", the
Regardless of the status of the next determination result and the status of the primary determination result one pixel to the left, it is set to "O". Therefore, pixels that are determined to be "O" in the forward scan of the prescan are always "O" even in the back scan of the prescan.
”. The secondary determination uses the calculation table shown in FIG. 18(e) as in the case of forward scan.

- In the primary judgment during forward scanning of copy scanning, when the AR color data is 1, that is, when data on the marker is received, the status is always set to "M" using the calculation table of FIG. 18(a), The AR color data is 0, that is, there is no marker color, and the page memory contents are “M
'' and the image data is 1, use the calculation table in Figure 18(d), and only when the status of the secondary judgment result one line before and the status of the primary judgment result one pixel to the left are both "O". "O", otherwise "M". In addition, if the AR color data is 0, the page memory content is "M", and the image data is 0, using the calculation table in FIG. 18(b), the secondary judgment result of one line before is "O", Or, only when the status of the primary judgment result of one pixel to the left is “O”, the judgment pixel is “
"O", and the status of the other secondary judgment results one line before and the status of the primary judgment results one pixel to the left are "M".
Alternatively, in the case of "I", the determination pixel is determined to be "I". The calculation table in FIG. 18(b) is also used when the AR color data = 0 and the content of the page memory is "I", and when the AR color data = 0 and the content of the page memory is "O", the calculation table in FIG. Set all to "O" using the calculation table in c). In the case of secondary determination, the calculation table shown in FIG. 18(e) is used as in the case of prescan.

FIG. 19 is for explaining the filling mode determination algorithm. In fill mode, the AR color data area and the black or red image data area surrounded by the AR color data are filled with AR color, and the image data is filled with AR color data, for example, as shown in FIG. 19(a). and Figure 19(b)
The area shown in is recognized. In this case, the area filled in by the marker is different from the area where the marker color was detected. For example, even if you intend to fill in a black image with a blue marker, only black will be detected on the black image,
When a color image is filled with a blue marker, it is unclear whether the detected color is blue, the color of the original document image, or the composite color of the marker and document image. Therefore, in the fill mode, the AR color data area and the black or red image data area surrounded by the AR color data are filled with the AR color. In this case, if detected in free type mode, it will be recognized as shown in Figure 19(c), and if this is ANDed with the image data, it will be recognized as shown in Figure 19(d), but fill mode is different from such detection. I can see that

In the calculation rule for determining the fill mode, AR
Define the status as "M" for the top of the color, "I" for inside the filled area (excluding the top of the AR color), and "O" for outside the filled area, and use the calculation table in Figure 18 to determine the status as in the free type mode. I do.

- At the time of forward scan of pre-scan 1
In the next determination, if AR color data = 1, FIG. 18 (a
), if AR color data = 0 and image data = 1, use the calculation table in Figure 18(b); if AR color data = 0 and image data = 0, use the calculation table in Figure 18(c) . The calculation table shown in FIG. 18(e) is used for the secondary determination.

・If the content of the page memory is "M" in the primary determination during backscan of prescan, then FIG. 18(a)
) operation table, when the page memory content is “I” Figure 18
When the content of the page memory is "O", the calculation table shown in FIG. 18(c) is used. The secondary judgment is shown in Fig. 18(e).
Use the calculation table.

- In the primary judgment during forward scanning of copy scanning, when AR color data = 1, the operation table in Figure 18 (a), AR color data = 0, image data = 1, and the page memory contents are "M". If the AR color data = 0, the image data = 1, and the content of the page memory is "I", use the calculation table of FIG. 18(d), and use the calculation table of FIG. 18(b).
If color data = 0, image data = 1, and the page memory content is "O", the calculation table in Figure 18(c), AR color data =
If the image data is 0, the calculation table shown in FIG. 18(c) is used. The calculation table shown in FIG. 18(e) is used for the secondary determination.

FIG. 20 is for explaining the rectangular mode determination algorithm. In the case of rectangular mode, a page memory is always required because rectangular areas that can be detected only by forward scanning and rectangular areas that can only be detected by back scanning are different. That is, Figure 2
When AR color data such as 0(a) is detected, the area shown by diagonal lines in Figure 20(b) is recognized in the forward scan of the pre-scan, and the determination result is further applied to the area from the backscan direction. By determining the inside/outside, the shaded area as shown in FIG. 20(d) is determined. However, as mentioned above, the judgment accuracy of pre-scanning is poor in this area, so it is not output as is, but the data captured during forward scanning of copy scanning is used as shown in Figure 2.
The determination result of 0(d) is corrected, and a fine region determination as shown in FIG. 20(c) is performed.

The rectangular mode determination algorithm is as follows:
Three statuses are defined: "M" for on the AR color, "I" for inside the rectangular area (excluding on the AR color), and "O" for outside the rectangular area.
, "O" that has the possibility of becoming "I" (exists only in the first judgment, and is determined as "O" or "I" in the second judgment) is changed to "O".
'' and the determination is made using an arithmetic table as shown in FIG.

・Forward scan of prescan 1
In the next judgment, when AR color data = 1, the calculation table shown in Fig. 21(a) is used, and regardless of the status of the secondary judgment result one line before and the status of the primary judgment result one pixel to the left, the judgment The pixel is designated as "M". In addition, when AR color data = 0, the calculation table shown in FIG. “O” regardless of the status of
If the status of the secondary determination result one line before is "M", "I", and the status of the primary determination result one pixel to the left is "O", "O'", "O'" is determined. It was determined that 1
The status of the secondary judgment result before the line is "M", "I"
, the status of the primary judgment result of one pixel to the left is "M", "I"
”, the judgment pixel is “I”.

[0052] Then, in the secondary judgment, the operation table shown in Fig. 21(e) is used, and the area judged as "O'" is rewritten as "O" or "I", and the area as shown in Fig. 20(b) is rewritten. The area is determined by secondary determination.

・If the content of the page memory is "M" in the primary determination during backscan of prescan, then FIG.
1(a) is used, and the contents of the page memory are
In the case of "I", the operation table shown in FIG. 21(c) is used. In the operation table of FIG. 21(c), if the status of the primary determination result one pixel to the left is "I", it becomes "I" regardless of the status of the secondary determination result one line before, and as shown in FIG. 20(b) The missing part ("O") of the rectangular area is rewritten as "I". In addition, if the content of the page memory is "O",
1(b) is used. In the case of secondary determination, the calculation table shown in FIG. 21(e) is similarly used.

- In the primary judgment during forward scanning of copy scanning, if AR color data = 1 and the content of the page memory is "M" or "I", judgment is made using the calculation table shown in FIG. 21(a). All pixels are "M", AR color data 1
If the content of the page memory is "O" in Figure 21 (d
) calculation table is used, and regardless of the status of the primary determination result of one pixel to the left, if it is "O", it will be "O", if it is "M", it will be " M”
, in the case of "I", a determination is made to replace it with "M". Also, if AR color data = 0, the page memory is "M" or "I".
In the case of AR
If the color data is 0 and the page memory is “O”, the
) calculation table is used. Similarly, in the case of secondary judgment, see Figure 2.
1(e) is used.

FIG. 22 is for explaining the determination algorithm for the one-point instruction mode. In one point instruction mode,
A closed loop is detected in the binary data of the image, and within the closed loop, the white image area adjacent to the AR color data and the AR
The color area is detected as being inside the area. In the case of single-point indication mode, as in the case of rectangular mode, there are different areas that can be detected only by forward scan and areas that can only be detected by back scan, so a page memory is required, and the judgment result of forward scan is transferred in the back scan direction. The inside/outside of the area is determined by re-judging from the above. In other words, when an AR color exists in the closed-loop image data as shown in FIG. 22(a), the diagonal area determined by the image data and the AR color is first detected by forward scanning as shown in FIG. 22(b). Recognized,
Next, by backscan determination, the white area in FIG. 22(b) is detected as inside the area, and an area as shown in FIG. 22(d) is determined. Then, in the forward scan of the copy scan, the determination result of the pre-scan is corrected, and a fine area is determined as shown in FIG. 22(c).

[0056] The status defined in the case of the one-point instruction mode is "I" for the area containing the AR color within the image closed loop;
The image area is defined as "M", the area outside the image closed loop is defined as "O", and the area inside the image closed loop but not including the AR color is defined as "O'", and the determination is made using the calculation table shown in FIG. 23.

・Forward scan of prescan 1
In the next determination, when image data=1, the conversion table shown in FIG. 23(a) is used, and all determination pixels are set to "M". When image data = 0 and AR color data = 1, the image data shown in Fig. 23 (b
) is used, and unless either the status of the secondary determination result one line before or the status of the primary determination result one pixel to the left is "O", the determined pixel is set to "I". When image data = 0 and AR color data = 0, the image is shown in Fig. 23(c).
When the status of the secondary judgment result one line before is "I", the judgment pixel becomes "I" unless the status of the primary judgment result one pixel to the left is "O",
Further, if the status of the primary determination result one pixel to the left is "I", the determination is made to be "I" unless the status of the secondary determination result one line before is "O". In the case of secondary determination, the calculation table shown in FIG. 23(d) is used.

・If the content of the page memory is "M" in the primary determination during backscan of prescan, then FIG.
23(a) is used, and all judgment pixels are set to "M", and when the page memory is "I", the calculation table of FIG. 23(b) is used, and when the page memory is "O", the calculation table is used. Figure 2
3(c) is used. The secondary judgment is shown in Figure 23 (d
) calculation table is similarly used. In the case of secondary determination, the calculation table shown in FIG. 23(d) is used.

・In the primary judgment at the time of forward scan of copy scan, if image data = 1, the result shown in FIG. 23 (
The arithmetic table in a) is used and all the pixels to be determined are set to "M", and when the image data = 0 and the content of the page memory is "I", the arithmetic table in FIG. All determined pixels are set to "I" except when the status of the secondary determination result before the line is "O" and the status of the primary determination result one pixel to the left is "O". Furthermore, when the image data = 0 and the page memory is not "I", FIG. 23(c) is used, and 1
If the status of the secondary judgment result before the line is “I”, the judgment pixel becomes “I” unless the status of the primary judgment result one pixel to the left is “O”, and the status of the primary judgment result one pixel to the left is “I”. If the status is "I", the judgment pixel is "I" unless the status of the secondary judgment result one line before is "O".
It is said that In the case of secondary determination, the calculation table shown in FIG. 23(d) is used.

[0060] In this way, the present invention is provided with a plurality of AR modes, which can be arbitrarily selected by simply changing the algorithm (selecting the calculation table), and can independently recognize areas for two specified colors. , you can select a separate area recognition mode for each specified color, so
For example, if the 1st AR color is blue and the free type mode is selected, the 2nd AR color is set to blue.
It is possible to set the R color to green and set the rectangular mode.

[0061]

As described above, according to the present invention, the number of rectangular areas to be recognized is unlimited, a plurality of rectangular areas lined up in the Y direction can be recognized separately, and a hardware latch circuit is not required. Most of the conventional closed-loop recognition circuit is also used,
By simply switching the algorithm to one for rectangular recognition, rectangular area recognition is possible and costs can be reduced.

[Brief explanation of the drawing]

FIG. 1 is a diagram showing the configuration of an area recognition circuit according to the present invention.

FIG. 2 is a block diagram showing the overall configuration to which the present invention is applied.

FIG. 3 is a diagram showing a specific configuration of an area recognition circuit and an editing/processing circuit.

FIG. 4 is a diagram showing the configuration of an area recognition circuit.

FIG. 5 is a block diagram of the AR enable circuit of FIG. 4;

FIG. 6 is a diagram for explaining an AR mode.

FIG. 7 is a diagram illustrating processing during forward scan of prescan.

FIG. 8 is a diagram illustrating processing during backscan of prescan.

FIG. 9 is a diagram illustrating processing during forward scan of copy scan.

FIG. 10 is a diagram illustrating primary determination.

FIG. 11 is a diagram illustrating secondary determination.

FIG. 12 is a diagram illustrating determination of an N-th pixel in an M line in the sub-scanning direction and an N-th pixel in the main scanning direction.

FIG. 13 is a diagram illustrating a region recognition method.

FIG. 14 is a diagram illustrating input correction.

FIG. 15 is a diagram illustrating connection correction.

FIG. 16 is a diagram illustrating a free type mode.

FIG. 17 is a diagram illustrating a determination method.

FIG. 18 is a diagram showing an operation table used in free type mode.

FIG. 19 is a diagram illustrating a fill-in mode.

FIG. 20 is a diagram illustrating rectangular mode.

FIG. 21 is a diagram showing an operation table used in rectangular mode.

FIG. 22 is a diagram illustrating a one-point instruction mode.

FIG. 23 is a diagram showing an arithmetic table used in the one-point instruction mode.

FIG. 24 is a diagram illustrating a conventional scanning optical system and area recognition.

FIG. 25 is a diagram illustrating a conventional editor pad.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1...control means, 2...scan timing generation means, 3...rectangular area recognition means, 4...memory.

Claims (2)

[Claims] [Claim 1] Scan timing that generates each scan timing of forward scan, back scan of prescan, and copy scan in a recording device that performs color separation from read image data, captures area recognition color data, and performs area recognition. and a rectangular area recognition unit that executes a determination process in synchronization with a scanning timing signal, and the rectangular area recognition unit captures area recognition color data in a forward scan of a pre-scan, and outputs the result while determining the area. Save it in memory, read the memory from the reverse direction again in the backscan of the prescan, and save the result in the memory while judging the area. An area recognition method for a recording device characterized by recognizing an area. 2. The area recognition method according to claim 1, wherein said rectangular area recognition means further captures area recognition color data during copy scanning, and corrects determination results during pre-scanning using the captured data.