JPH046302Y2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Industrial Field of Application] The present invention relates to a cursor recognition device for a scanning machine.

[Prior Art] A scanner that places a paper face down on a paper glass and optically scans and reads the characters and images on the paper, scans the image in a range specified by marks printed in advance on the paper. There are some that scan using an optical sensor.

Some scanners have a movable cursor attached to the machine frame beside the paper instead of the marks on the paper, and scan the image within the range specified by the cursor. FIG. 2 shows how to move the movable cursors 7 and 8 from inside the glass on which the paper is placed on such a scanner.
2 is a view of the image area reference marker 9. When placing a sheet of paper in the image area 5 and wishing to partially scan a partial scanning area 6 within it, the operator selects CL, CR to determine the left and right boundary positions of the scanning area.
A pair of cursors 7 indicated by are moved left and right while being guided by a rail or the like as appropriate. Also, the upper and lower boundary positions are
Decide by moving the pair of cursors 8 indicated by CT and CB up and down as appropriate. These cursors 7 and 8 are white so that they can be easily recognized against the black area of the background.
A white image area reference marker 9 is provided to provide a reference position for the image area. Vertical lines indicated by H, R, and L of this marker 9 provide reference lines for horizontal scanning, and horizontal lines indicated by V, R, and L provide reference lines for vertical scanning.
The image of the cursor or marker is read through scanning reading means such as a CCD element, and by recognizing this, the area 6 to be partially scanned is determined. That is, before scanning the partial scanning area 6, use the same scanning means to check the positions of the cursors 7 for CL and CR, memorize the positions, and then scan the partial scanning area 6 while searching for the positions of the cursors 8 for CT and CB. scan at the same time.

Conventionally, when marks on paper are used to define a partial scanning area, the density of the marks is relatively stable, there is little deterioration due to adhesion of dust, and the recognition pattern obtained by scanning is There was continuity. It was not very difficult to recognize such a mark with one scan, that is, with one scanning line.

[Problems that the invention aims to solve] However, when a cursor is used instead of the above-mentioned marks on paper, it becomes difficult to recognize the cursor stably because dirt and dust accumulates with repeated use. things become difficult. Also, since the cursor is located outside the image area, noise also increases.

From another point of view, when scanning with pseudo halftones using the dither method as an image area recognition method, the cursor image obtained with a single scanning line becomes a discontinuous black and white pattern.

Therefore, it is an object of the present invention to provide a cursor recognition device that can stably recognize a cursor even with the above-mentioned problems.

[Means for Solving the Problems] The present invention includes a scanning means 1 that scans the cursor and its background area multiple times, instead of recognizing the cursor and its background area by scanning only once as in the conventional method. Then, the black pixels (the number of white pixels may be used) in each small area are counted not as lines but as planes. The number of black pixels in each small area may be stored as is in Table 2-1, or the number of black pixels in each small area and a predetermined number of adjacent small areas (two on both sides in the embodiment) may be stored as is. The total value may be obtained by the totalizing means 2-2 and then stored in the table 2-1 as the number of black pixels for each small area. When the number of black pixels of each small area is stored in the table 2-1, the summation means 2-2 obtains the total value. By summing the number of black pixels in a predetermined number of small areas in this way, even if there is black dust that happens to cover one small area, the total value of black pixels including adjacent small areas can be calculated. are averaged to reduce the influence of dust adhesion. The total value obtained in this way is compared with the threshold value in the threshold value storage means 3 by the comparison means 4,
Each subregion produces either white or black output.

The threshold value storage means 3 may store two threshold values, but this will be explained later along with FIG. 4.

[Example] In this example, in addition to a pair of left and right cursors 7 indicated by CL and CR, a pair of upper and lower cursors 8 indicated by CT and CB are also scanned, but their configurations and operations are almost the same. Therefore, the former cursor 7 will be explained.

As shown in the upper part of Figure 3, 1 indicated by CL and CR
The pair of movable cursors 7 undergoes 30 horizontal scans with the image area fiducial marker 9 in a fixed position and the black area of its background. One small area is divided into 8 pixels or 8 bits in the horizontal direction and 8 bits in the vertical direction.
Set to 30 pixels or 30 bits, totaling 240 bits,
The number of black bits obtained by scanning is stored in table 2-1 for each small area. The size of one small area is set to be smaller than the size of one cursor. Ideally, if you set cursor width >> small area width >> dust size, you can clearly and stably read the cursor position. For simplicity, in FIG. 3, the width of one cursor 7 is shown as four small areas, but in this embodiment, it is approximately 6.5 small areas.

If the contents of Table 2-1 are shown graphically, the third
It will look like A in the figure. If all pixels in one small area are black, a value of 240 on the vertical axis is plotted. The horizontal axis corresponds to the position of the cursor etc. shown in the upper part of FIG. If black dust is attached to one small area on the white cursor, as shown in Figure 3A.
The number of black bits in part a of the horizontal axis stands out.
If this is directly compared with a threshold value, such as 120, between 240 and 0, there is a possibility that this small area with black dust will be determined to be a black background part.

In the present invention, in addition to the position of a, the adjacent b
Summation means 2 for summing the number of black bits at the positions of and c
(See Figures 1 and 4) to create plot A as shown in graph B in Figure 3. Similarly, for the position b, plot B is created of the total number of black bits at positions c and d, and for the position c, plot C is created. As a result, 40% of the maximum number of black bits (if 240 per small area, the number of black bits in the three small areas, 720)
When compared with the threshold value, the influence of dust is reduced or averaged by three small areas, and the comparison result is correctly output as "white".

In the embodiment, the number of small areas to be totaled is three, but it can be set to any predetermined number depending on the size of the small area with respect to the cursor.

Further, in this embodiment, the number of black bits for each small area is stored in the table 2-1, but a total value of the number of black bits for three small areas may be stored. Furthermore, instead of counting the number of black bits in each small area and then summing them, the total value may be obtained by counting the number of black bits in the first three small areas at once.

The dimensions and pixel density of the three small areas in the example are approximately 3 mm and 24 bits in the horizontal direction, and 24 bits in the vertical direction.
It is set to approximately 3.8mm and 30 bits (30 scanning lines).

The above settings are for cursor 7 (CL, CR), but for cursor 8 (CT, CB), the horizontal direction is 16 bits at approximately 2 mm, and the vertical direction is 24 bits at approximately 3 mm.
bit (24 scanning lines).

In the embodiment shown in FIG. 4, two threshold storage sections 3-1 and 3-2 are stored in advance in the threshold storage means 3, and one of the thresholds is selected by the selection means 3-3. , is compared by the comparison means 4 with the total value from the means 2 for generating the total value. The black-and-white threshold used when checking whether there is a transition from white to black and the black-white threshold used when checking whether there is a reverse transition are shown in graph B in Figure 3. 2 large and small
By providing two, it is possible to increase the margin 1 for dust in the black area and the margin 2 for dust in the white area. Margin is the allowable range of dust.

For example, from the black small area that is the background area on the right side of FIG. The number of bits is compared with the threshold value. Since the first comparison is performed on a small area in a known black background area, the threshold value from the black-and-white threshold storage section 3-2 is compared with the number of black bits of the first small area in the initial setting. Since the black-and-white threshold is set low, even if some whitish dust is attached to a black background area, it is determined to be black. That is, margin 1 is large.

The black or white output from the comparison means 4 is fed back to the selection means 3-3, and if it is a black output, a black-and-white threshold value is given to the comparison means 4, and if it is a white output, a black-and-white threshold value is given to the comparison means 4, and the black or white output for the next small area is given to the comparison means 4. compared to the number of bits. In the graph B of FIG. 3, the number of black bits in the small area at position B has become lower than the low black-white threshold, so a white output occurs. This white output is
It represents the right end of the cursor 7 shown as CR.

Point A in the graph B in FIG. 3 is similarly determined to be white, but the threshold to be compared at this time is the high black-and-white threshold. This is because the comparison output at point B was white. The margin 2 can be increased by comparing with a high black-and-white threshold. In other words, even if there is some blackish dust on the white cursor, it will not be displayed correctly as "white".
It is determined that In this way, it is found that the first small area for which the number of black bits exceeding the black-and-white threshold is given to the comparing means 4 has a black background color, and thus the small area immediately before it is shown by CR on the right side. This will represent the left end of the white cursor 7. One cursor 7 is identified between the above-mentioned right end and this left end. Although FIG. 2 shows an example of using the right edge of the cursor as the boundary of the image scanning area, other locations, such as halfway between the two edges, can be used as the boundary.

Although the cursor recognition device for a scanner according to the present invention has been described above, those skilled in the art will readily understand that the functions described can be performed by microcode or software in various ways.

The cursor may be black, the background may be white, or the number of white pixels may be counted. It will also be understood that the present invention is applicable not only to optical sensing devices but also to magnetic sensing devices.

[Effects of the invention] In this invention, the cursor and background area are identified by scanning not with a single scanning line but with multiple scanning lines, and by counting the number of black bits in a small area as a plane rather than a line. , moreover, the total number of black bits in a predetermined number of small areas is compared with a predetermined threshold value. Therefore, even if some dust is attached, the influence of the dust is reduced and the cursor can be recognized stably.

When scanning and recognizing using the dither method, the cursor can be recognized as a continuous pattern because it is compared as a plane.

Since a table is provided, if the number of black bits is stored in the table, applications such as tracking the cause of cursor unreadability and considering adjustment of the threshold value are also possible.

The embodiment in which two threshold values are stored has the effect of providing a large margin (tolerable range) against noise such as dust.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing a cursor recognition device for a scanner according to the present invention. FIG. 2 is a view of the cursor and image area reference marker viewed from inside the glass on which the paper is placed. FIG. 3 is a diagram for explaining the concept of the device of the present invention. FIG. 4 is a block diagram of an embodiment of the present invention. 1...Scanning means, 2-1...Table, 2-2
...Summing means, 3...Threshold storage means, 4...Comparing means.

Claims (1)

[Claims for Utility Model Registration] (1) A scanning means for scanning a cursor area and its background area with a plurality of scanning lines; , stores the number of black pixels obtained as a result of scanning for each small area whose size is set to have a vertical width that allows multiple scanning lines to pass through and a horizontal width that includes multiple pixels per scanning line. a table for calculating the total number of black pixels in each small region and a predetermined number of adjacent small regions; a threshold storage means for storing a threshold value; and a total value obtained by the summation means and the above. Comparing means for producing an output indicating whether each small region is white or black by comparing the threshold value from the threshold value storage means, thereby distinguishing a white or black cursor from a black or white background region. A cursor recognition device for scanning machines that can recognize (2) A utility model registration request characterized in that the threshold value storage means stores at least two threshold values, and generates one of the two threshold values depending on a white or black output from the comparison means. A cursor recognition device for a scanner according to scope (1). (3) The number of black pixels for each small area stored in the table is the number of black pixels in each small area, and the summing means generates a total value from the number of black pixels obtained from the table. A cursor recognition device for a scanner according to claim (1) of the claimed utility model registration. (4) Scanning according to claim (1) of claim 1, wherein the number of black pixels for each small area stored in the table is the number of black pixels of the total value generated by the summing means. Machine cursor recognition device. (5) The cursor recognition device for a scanner according to claim (1), wherein the scanning means is an optical or magnetic scanning means.