JPH0366673B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION (A) Field The present invention relates to a scanning method of a pixel matrix in a display or recording device expressed by a pixel matrix, and in particular to a scanning line width corresponding to one pixel pitch at the scanning edge of scanning a pixel matrix. The width of the spot is divided into multiple overlapping units of scanning edge, and one pixel pitch width is scanned as needed. Mark at the scanning edge is controlled to vary the width of the scanning line. Variable scanning line width display record. Regarding the scanning method for

(B) Prior Art Conventionally, in methods for displaying or recording characters, figures, etc. by a combination of brightness-modulated main scanning lines and sub-scanning, pixels for forming scanning lines are used for each scanning pitch. Writing point (spot)
By making the diameter larger than the scanning pitch and slightly overlapping each scanning direction, a scan slightly thicker than the scanning pitch is formed.

This is because if the scanning line width is equal to or less than the scanning pitch, errors in the scanning pitch will cause unfilled areas to be filled in to represent characters or figures. However, when this method is adopted, the line width of the actual writing (black in positive drawing, called a mark) is thick, and the line width is blank (white in the same drawing).
The line width becomes thinner. That is, even if actual writing is performed by one scanning pitch, the line width of the blank portion will be less than the scanning pitch, and this line width will be affected by various errors inherent in the device as described above.

Therefore, with this method, it is not possible to record characters and figures that require one scanning pitch blank for expression, but this problem is particularly problematic in phototypesetting, where characters and figures must be recorded in negative form. In the case of original plates, etc., the blank area becomes black when it finally becomes a positive (positive image), so the black line width becomes thinner, and in the worst case, including errors, lines may be missing. . Of course, the above problem can be solved by making the pitch of the main and sub-scans finer and intervening multiple scanning pitch widths to express the blank part, but the number of pixels to be dealt with will depend on the scanning pitch. This increases at a square rate, and the memory to be backed up and the number of scans also increase correspondingly. In any case, once the pixel arrangement pitch is set, the blank scanning line width cannot be freely selected.

(C) Purpose The present invention was developed in view of this background, and its purpose is to scan a single pixel pitch in a shared manner without changing the pixel density (vertical pixel pitch x horizontal pixel pitch) constituting the display recording surface. The purpose of the present invention is to make it possible to variably control the width of the scanning line during scanning, and the feature of the present invention is that in order to achieve the above object, the scanning direction array pitch P is divided in units of one pixel in each of the main and sub-scanning directions. scanning means for recording marks or blanks corresponding to pixels of pixel data as image data on the recording surface while performing main and sub-scanning on pixel points on a display recording surface; and distribution of pixel data in synchronization with scanning. and mark pixel recording means arranged in the sub-scanning direction corresponding to the arrangement pitch P and larger than the arrangement pitch P in order to record marks with an overlap margin for at least pixel points adjacent in the sub-scanning direction. In a scanning device for display recording, which is configured in a sub-scanning bundle with a plurality of one-pixel scan ends adjacent in the sub-scanning direction as a unit of group sub-scanning, the one pixel is shared. One pixel scanning edge is composed of ξ unit scanning edges arranged adjacently in the sub-device direction with an array pitch P that is one integer ξ of the above-mentioned array pitch P, and each of the unit scanning edges has a skid width q. The values of are the feed error δ in the group sub-scanning, α the mark recording width, and β the correction value necessary to determine the skid width q (however, β is 0<β≦p). ), (q-p)=ε+α, then ε+
By configuring the skid width q to satisfy α=p−β(), ε+α>δ(), the mark width change amount (of scanning in the sub-scanning direction by setting the relationship (relationship), and ensuring (relationship) that the feed error δ is filled in between adjacent marks without any gaps, and monitoring changes in the pixel pattern in the sub-scanning direction. If the pixel pattern changes from blank to mark in sub-scanning order, or from mark to blank to mark, the change point at which the pixel pattern changes to that mark is detected and the pixel pattern changes from blank to next. The recording scan is performed by changing a blank in one pixel scan end for recording marks located in the sub-scanning order to a blank and an adjacent unit scan end to a blank.

(D) Embodiment FIG. 1 is an explanatory diagram of the concept of the present invention, and is an explanatory diagram of a scanning plane and a spot formed by the concept. In the figure, x and y are the main and sub-scans, respectively, the center of the grid is the pixel center, the numbers added in the x and y directions are addresses to define the pixel position of each scanning system, and Px and Py are the main and sub-scans, respectively. The pixel arrangement pitches in the scanning direction, Dx and Dy, indicate the dimensions of the pixel point (mark spot when the scanning end writes a mark) at which the pixel address is scanned, as seen from each direction.

It is assumed that pixels filled with circles in the figure represent mark spots, and areas without circles represent blanks, which are written while being scanned by main and sub-scanning, as shown in FIG. 1, for example. In the conventional technology, in such matrix scanning writing, Dx and Dy are larger than Px or Py, respectively, so (Dx − Px) = εx,
If (Dy−Py)=εy is defined, the line width of the blank part is nPx−εx or mPy−εy, and the line width of the marked part (actual writing side) is nPx+εx or
It becomes closer than mPy+εy. Note that n and m are the numbers of intervening pitches.

FIG. 2 is an explanatory diagram of one embodiment of the present invention, and is for explaining a scanning end at which pixel scanning is performed. In the figure, A indicates a unit scanning end, and B indicates a pixel scanning end. In the figure, Y indicates one scanning bundle in the sub-scanning direction; that is, the sub-scanning pitch division width of the scanning edge where multiple pixel addresses in the sub-scanning direction are bundled and simultaneously scanned, and U1 to U12 are each unit scanning edge. shows,
One of the unit pixels defined as one of the pixel addresses in FIG. It is assumed that one group of unit scanning ends is assigned and controlled so that each scans one pixel when viewed from a pixel scanning perspective (see FIG. 2B). Note that the explanation will mainly be given using the sub-scanning direction as an example. Therefore, Py is abbreviated as P.
Note that the size of the spot at the unit scanning end is defined as q in the y direction, P/ξ=p is defined, and q-p=
It is defined as ε+α. However, ε is the difference between the width of the mark spot at the end of the unit scan and p when there is no bleeding,
α is the blurring margin when actually written. When performing surface scanning with ξ pixel scanning edges based on such a configuration, the width of the mark spot made up of ξ unit scanning edges is ξp + (ε + α), and the width of the blank is ξp - (ε + α). It is obvious that the number of unit scan edges that share one pixel pitch can be made variable under certain setting conditions corresponding to a certain control purpose: For example, if the number of unit scanning edges that share one pixel pitch can be made variable under certain setting conditions: If you want to make the scanning width Dy of the blank the same as the scanning width 2P-Dy of the blank, in this example, to make Dy-P=0, for example, p=
This can be achieved by controlling the number of unit scanning edges that constitute one pixel scanning edge to be ξ-1, by configuring it so that ε+α and only when scanning a mark under the above setting conditions. , if the width of the mark is controlled to be less than 2 or more, the width of the mark as one pixel width can only be made larger than the width of the blank due to the overlapping margin, which is the constraint in the conventional example. It is also possible to make the width of the blank to be scanned larger than the width of the mark. As a modification of this relationship, we have now considered the condition of P = ξp, but when one pixel pitch is shared by k times the unit scanning end of ξ, kp = ε + α, and the scanning width is controlled, P(k±j) as a unit (where j=1, 2, 3...
...), of course, does not deviate from the intention of the present invention. As another modification, the above embodiment is
We have explained how to realize the condition Dy-P=0. Under the condition that the pixel pitch P is determined, the 1-pitch line width Dy of the mark and the 1-pitch line width (2P-Dy) of the blank are the sum of 2P. As in the previous example, one pixel is divided into unit pixels of kξ, and when one pixel scan edge scans a mark, the unit scan edge is divided into kξ unit pixels. kξ−i (where |i|=1, 2, 3,
), but in general, by defining p-(ε+α)=β, when β=0, if correction is made by p, the scanning line widths of marks and blanks will match. When β≠0, the scanning line widths of the mark and the blank do not match. However, it is necessary to set a value smaller than the above p as the setting amount of β, that is, 0
By setting β in the range <β≦p, the pixel pitch p can be set without changing, and without increasing the number ξ or kξ of the unit scan edges that share one pixel scan edge, the size is less than or equal to p. It is possible to change the scanning width of the mark spot portion and the blank portion in units of . In addition, I bundled some p's y
If we consider that the indexing error when sub-scanning the scanning bundle is ±δ/2 in the overlapping part of the sub-scanning by the scanning bundle of pitches, the mark at the scanning end by the previous scanning bundle and the mark of the next scanning bundle due to the above error are In order to ensure an overlapping margin between the mark and the mark due to the scanning end of the mark and fill in the marks without any gaps, the overlapping margin α+ε must be larger than δ. That is, the relationship α+ε>δ is required.

While scanning the scanning plane with the unit scanning edge and scanning bundle that satisfy the above conditions, if necessary, the one-pixel scanning width can be changed from the normal scanning width of ξ lines to ξ-1 lines or (- 2, -3... books)
Or, by performing control such as (+1, +2, +3...), it is possible to scan a predetermined pixel portion with a predetermined mark width or blank width, and under this condition, p=ε+α or p-β=ε+α However, since 0<β≦p, if the number of ξs selected differs in the scanning area of one device, that is, if the pixel scanning pitch is partially changed within the scanning area, Even if it is necessary to change the resolution, the mark and blank widths can be kept constant if β = 0, and 0 < β.
Even in the case of ≦p, the difference between the mark width and the blank width can be kept constant, and to meet the demand for variable resolution depending on the scanning area, it is necessary to set the overlap margin and size of the scanning edge as in the conventional example. In the conventional example, the unit scan edge is distributed and another one-pixel scan edge is set, and the scan bundle formed thereby does not require any design consideration or modification work such as setting the width or replacing the scan edge in the device. However, changes can be made simply by providing the necessary pattern information for pixel writing for scanning the area. In addition, if there is no requirement for P itself to be variable during scanning as described above, and there is only one component number ξ in one system, only the unit scan edge located between pixels ξ
The pitch of each unit pixel is P, and among the unit pixels adjacent to the scan edge, if it is on the front side, it is the front side, or if it is on the next side, it is the next side.This and the mark width of the previous unit scan edge are overlapped and synthesized. If the composite mark width is equivalent to 2p + α + ε defined in the example above,
In other words, even if the value of q, which is the width of the mark spot, is increased from the center of the unit scan edge to the side that is not diagonal, the overlapping margin of the mark spot at the unit scan edge on the side that is not the boundary will only increase. As long as the part in question does not share the boundary between pixels; that is, as long as it does not protrude beyond either boundary within a pixel, it can take any value; however, at the boundary, the relationship in the previous example is applied. This will also be included in the present invention.

FIG. 3 is an explanatory diagram of one embodiment of the present invention, and is an explanatory diagram of the embodiment in terms of a device image.

In the figure, 1 is a rotating drum that forms the surface to be scanned.
Ay is the scanning end of one scanning bundle, y is the sub-scanning unit of the scanning bundle, 2 is the scanning head, 3 is the optical system, and 2A is a writing element group that forms the scanning end, where a plurality of unit writing elements correspond to one pixel. One or more of these are further assembled to form a write element group for forming a scanning bundle.

The scanning bundle consists of a plurality of one-pixel scanning edges, and as this scans in the main scan x, the sub-scanning is performed in a bundle corresponding to the plurality of pixels, which is a so-called multi-scanning edge.

Note that 5 and above are writing and scanning control related parts, 5 is a pattern memory such as a character pattern generator, 6 is a data file that stores writing or display information, and 7 is a data file that stores writing or display information.
8 is an editing processing unit that edits pixel pattern information from 5 and 6 above; 8 is a video buffer that stores the edited pattern information at least for one scan bundle at the same main scanning time; 9 is a unit that forms a unit scan edge. The writing elements are divided into groups such that each one forms one pixel scanning end, and the scanning pattern information from the video buffer 8 is applied to the unit writing elements that form the same pixel as one group in one sub-scanning order in the main-scanning order. The allocation distribution unit 13 is a sub-scanning monitor which monitors pattern information stored in the video buffer in main-scanning order for at least one sub-scanning bundle in the sub-scanning direction for the same main scanning. This is the part that detects that there is a change pattern. Reference numeral 14 denotes a scanning width correction unit which, when a specific pattern is detected on the sub-scanning monitor 13, determines the number ξ of unit scanning edges of one pixel scanning edge in which marks of the corresponding pattern are written, on the side adjacent to the blank scanning edge next to it. It is a work that increases or decreases. Also 1
0 is a time correction unit, which calculates the amount of spatial misalignment in time in the main scanning direction when writing elements for forming a unit scan that requires an overlap margin cannot be formed in a straight line on the same substrate as shown in Figure 2. After correcting the deviation, the pixels scanned for recording or display on the rotating drum or the unit scan end positions are aligned the same, and the pixels on the scanning surface are set so that they correspond to the addresses in Figure 1. It is used as a pixel scanning plane. In such a configuration, while writing is in progress from the scanning end, patterns in the video buffer 8 stored in the sub-scanning bundle in main-scanning order are monitored on the sub-scanning monitor 13 in the order of sub-scanning addresses for the same main scanning, and patterns of changes in the sub-scanning direction are monitored. For example, when detecting a mark from a blank or a mark, a blank, a mark, the scanning width correction unit 14
If the unit write element that forms the single scanning edge located on the side that is in contact with the blank of the mark that appears next to the blank does not write marks, the scanning width on the blank side becomes mp - (α + ε) + p and overlaps. It becomes possible to set the minimum line width of characters and graphics based on pixels in accordance with design requirements, without being bound by constraints on width and pixel pitch.

(E) Supplement: The pixel width at the end of the scanning bundle is equal to both widths of the feed error ±δ/2 of the width y of the scanning bundle, that is, the amount corresponding to the error in the feed amount of the scanning head 2 that sub-scans the scanning bundle. The overlapping margin between the scanning ends changes, but when looking at the pixel pattern being scanned in order in the sub-scanning direction... mark, blank, mark, etc., when the pattern exactly comes to the overlapping part of the scanning bundle, that blank is at the last address of the previous scan in the scanning bundle (Case 1), or at the first address of the next scan (Case 2). If the scanning is +δ/2 and the subsequent scanning is -δ/2, even if the correction in the embodiment is performed, the blank width during scanning (remaining width to be filled in) becomes ξp-(ε+α)+p-δ, which is in the scanning bundle. The scanning width is narrower by δ than the scanning width of This will require some ingenuity. On the other hand, if we limit our thinking to the scanning pattern used in photolithography, for example, if we consider the pixel pattern configuration of typeface in Mincho typeface, we can see that compared to the thin horizontal lines of Mincho typeface (blank in the original plate), the white parts are The width of the (mark in the original version) is usually designed to be large, so in actual operation, as long as you secure the line width of the thin line, there is no need to renormalize the effect on the white part to achieve it. There is little practical adverse effect. In addition, the overlap due to the feed error becomes wider, and the scanning width of the blank becomes ξp−(ε+α)+p
+δ is also acceptable in practice. In such a case, as shown in FIG. 4, by slightly changing the arrangement of the unit scan ends constituting one pixel scan on the side with the smaller sub-scan address among the one pixel scan ends at the end of the scan bundle, the scan width in the embodiment can be changed. The scanning width of the blank at the end can be corrected without changing the control.

FIG. 4 is a supplementary diagram of one embodiment of the present invention and is compared with FIG. Note that the one pixel scanning end in this figure corresponds to ξ=3. Further, in the figure, A is expressed by a unit scanning edge, and B is expressed by a scanning bundle and one pixel scanning edge. The arrowed portion indicates the end of the scanning bundle, that is, the section of the scanning by the sub-scan y, which is affected by the feed error ±δ/2. What is different from Fig. 2 is that in the scanning bundle diagram, the address of the one-pixel scanning edge is The spot widths at the unit scanning end at the end adjacent to the younger side of the image are q+γ and q−γ, respectively, when viewed in the forward address direction, and the overlap is shifted by the correction amount γ.

With this configuration, for example, as in Case 1, if a pattern is scanned with a mark before the first scan, a blank at the end, and a mark at the beginning of the second scan, and the correction of the previous embodiment is applied, the correction will be made. The change in mark scanning width is not p but p+γ, and the blank scanning width in this case is ξp−(ε+α)+p
Therefore, even if δ is negative, that is, the blank scanning width is narrowed, the line width is increased by γ, and the minimum line width can be secured by the set value of γ. Also, as in case 2, if the end of the first scan is a mark, the beginning of the second scan is a blank, and the next is a mark, similarly, the second one pixel scan from the youngest end in the address order of the one pixel scan end Similarly to case 1, the minimum line width can be ensured by setting the spot widths to q+.gamma. and q-.gamma., respectively, for the inside of the end. Using this method, although it is not possible to reduce the feed error itself, it is possible to compensate for fluctuations in line width that occur due to feed errors in the scanning bundle in the scan connection area without changing the correction control method or pixel pitch. If possible, the minimum line width at the joint can be corrected as a problem at the joint.

(F) Effect As described above, according to the present invention, the pixel density of the display or recording surface, that is, the number of unit area dots in the pattern memory, can be
The minimum line width or desired line width for marks or blanks can be achieved using a relatively simple method of pattern monitoring and scanning width correction, and even if the resolution requirements change during scanning, the group at the unit scanning edge can be changed. Furthermore, by changing the grouping, characters and figures can be enlarged or reduced, and even if the number of groups ξ is changed, the unit line width of marks and blanks remains unchanged. This produces the characteristic effect of being able to make the two parts the same.

[Brief explanation of drawings]

Fig. 1 is an explanatory diagram of the concept of the present invention, Fig. 2 is an explanatory diagram of an embodiment of the invention, Fig. 3 is an explanatory diagram of an embodiment of the invention, and Fig. 4 is an illustration of an embodiment of the invention. Supplementary figure,
In the figure, x is the main scanning, y is the sub-scanning, Px and Py are the scanning pitches in the main and sub-scanning directions, Dx and Dy are the diameters of the same mark spot, U1 to U12 are the unit scanning ends, and q is the mark at the unit scanning end. The dimension of the spot, p is the arrangement pitch of the unit scanning end, 1 is the rotating drum, 2 is the scanning head, Ay is the scanning end, 9 is the layout distribution section, 13 is the pattern monitor, and 14 is the scanning width correction section.

Claims (1)

[Claims] 1. A scanning direction array pitch P in each of the main and sub-scanning directions.
a scanning means for recording marks or blanks corresponding to pixels of pixel data on the recording surface as image data while performing main and sub-scanning on pixel points on a display recording surface divided in units of one pixel; a means for supplying and distributing pixel data in synchronization with pixel data, and a means for distributing pixel data that is arranged in the sub-scanning direction corresponding to the arrangement pitch P in order to record marks with an overlap margin for at least pixel points adjacent in the sub-scanning direction; A scanning device for display recording, which is configured to have a mark pixel recording width larger than the pitch P, and is composed of a sub-scanning bundle in which a plurality of one-pixel scan ends adjacent in the sub-scanning direction are units of group sub-scanning. In the above, one pixel scanning edge that shares one pixel is composed of ξ unit scanning edges arranged adjacently in the sub-scanning direction with an array pitch P that is an integer ξ of the array pitch P, and the above unit The value of each skid width q at the scanning end is determined by the feed error δ of the group sub-scanning, α the bleeding width during mark recording, and β the correction value necessary to determine the value of the skid width q (however, β is 0<β≦p), when (q-p)=ε+α, ε+α
= p - β (), ε + α > δ () When one unit scan end changes from mark to blank by configuring the skid width q so that it satisfies ) and
In addition, for the feed error δ, it is ensured that adjacent marks are filled in without any gaps, and one pixel pattern in the scanning order in the sub-scanning direction is changed by means of monitoring changes in the pixel pattern in the sub-scanning direction. Monitors the change, and if the pixel pattern changes from blank to mark in sub-scanning order, or from mark to blank to mark, detects the point of change to that mark and positions it in the next sub-scanning order of blank. A scanning method for recording, characterized in that recording scanning is performed by changing a blank in one pixel scan edge where a mark is recorded and an adjacent unit scan edge to a blank. 2. The scanning method for display and recording according to claim 1, characterized in that the correction value β is 0. 3. The arrangement pitch P of adjacent one-pixel scanning edges constituting the sub-scanning bundle is selectable and realized by changing the number ξ of unit scanning edges constituting one pixel. A scanning method for display recording according to item 1 or 2. 4 Arrangement pitch P of one pixel scanning end in the above sub-scanning bundle and number of unit scanning ends ξ constituting one pixel
is fixed, at least one unit scan end located at the division of one pixel scan end is arranged every ξ with an arrangement pitch P, and at least the end of the sub-scan bundle on the younger side in the sub-scan order and the next The skid width of the unit scan end located at the beginning of one pixel scan end is the positive correction amount γ added (q + γ), the width of the next unit scan end is (q - γ), and both unit scan ends are The skid width of the unit scan end located at the younger end of the sub-scanning bundle in the sub-scanning order and the skid width of the next unit-scanning end are maintained while maintaining the condition that the size of the combined mark width is (2p + ε + α). A scanning system for display recording according to claim 1, 2, or 3, characterized in that the scanning system is configured by changing the dimensions of the display.