Multi-level to bi-level raster shape converter
    21.
    发明授权
    Multi-level to bi-level raster shape converter 失效
    多级到双级光栅形状转换器

    公开(公告)号:US5589851A

    公开(公告)日:1996-12-31

    申请号:US210869

    申请日:1994-03-18

    摘要: A method and apparatus for converting multi-level raster shapes into bi-level raster shapes while preserving as much of the visual character of the shape is disclosed. The method and apparatus solve two main problems arising in the task: that of broken continuity (drop-outs) and that of large changes in the width of horizontal and vertical lines on the bi-level raster shape as a result of minute changes on the multi-level shape (stem width aliasing). Drop-outs are handled by identifying certain boundaries between pixels as drop-out warnings. Pixels adjacent to dropout warnings have their coverage values converted from multi-level to bi-level by a process that guarantees that at least one of them will be rounded up. Dropout warnings are found from the outline of the ideal shape used to generate the multi-level shape and represent an approximation to said outline. In a preferred embodiment, the set of drop-out warnings for an outline is the set of pixel boundaries containing the nearest pixel boundary to each intersection of the outline with mid-pixel lines. Stem width aliasing occurs because of rounding errors clustering along horizontal and vertical lines. The present invention minimizes stem width aliasing by a variable threshold technique in which the threshold value used to turn the multi-level coverage value of a pixel into a bi-level value depends on the pixel position. The multiple threshold values are used to scatter the rounding errors over neighboring pixels so that they do not cluster along horizontal or vertical lines and are therefore less apparent to a viewer. In a preferred embodiment, two different threshold values are used in a checkerboard pattern, i.e., the threshold value for every other pixel in the horizontal or vertical direction alternates between the two values.

    摘要翻译: 公开了一种用于将多级光栅形状转换为双层光栅形状同时保留形状的视觉特征的多数的方法和装置。 该方法和装置解决了任务中出现的两个主要问题:断续连续性(drop-out)和双层光栅形状上的水平和垂直线宽度变化大的变化是由于 多级形状(杆宽度混叠)。 通过将像素之间的某些边界识别为辍学警告来处理辍学。 与辍学警告相邻的像素,其覆盖值由多层次转换为双层级,过程保证至少其中一个将被舍入。 从用于产生多级形状的理想形状的轮廓找到压差警告,并表示对所述轮廓的近似。 在优选实施例中,轮廓的一组丢弃警告是包含轮廓与中间像素线的每个交点的最近像素边界的像素边界集合。 发生干宽度混叠是因为沿着水平和垂直线聚类的舍入误差。 本发明通过可变阈值技术最小化干宽度混叠,其中用于将像素的多级覆盖值转换为双电平值的阈值取决于像素位置。 多个阈值用于分散相邻像素上的舍入误差,使得它们不沿着水平或垂直线聚类,因此对于观看者来说不太明显。 在优选实施例中,在棋盘图案中使用两个不同的阈值,即,水平或垂直方向上每隔一个像素的阈值在两个值之间交替。

    Method and apparatus for filling contours in digital typefaces

    公开(公告)号:US5068803A

    公开(公告)日:1991-11-26

    申请号:US407911

    申请日:1989-09-15

    CPC分类号: G09G5/24

    摘要: A method and apparatus is described which fills in the pixels missed when drop-out occurs during the fill process of character contours. The character contour is decomposed into a series of rook moves. The pixel selected is the one more covered by the actual shape of the portion of the curve where the rook moves are coincident and dropout occurs. This is dependent upon the slope of each of the curves at the location of dropout. Preferably the length of the sequence of colinear consecutive rook moves is used to approximate the slopes of the curves. The target pixel of the longest sequence of colinear rook moves is more covered than its opposite pixel and therefore the target pixel is set. The target pixel for a rook move is the pixel in the winding direction (i.e., left or right direction) along the rook move. Thus the target pixel of the stronger rook moves will be set and added to the bit map image generated using a outline fill process. In the preferred embodiment, the rook moves are examined to determine the length of a sequence, that is the number of rook moves that are sequentially in alignment with one another. A strength value is then assigned to each of those rook moves in a sequence equal to the number of rook moves in the sequence. The strength value of the rook move is then compared to the strength value of the opposite pixel, that is the pixel opposite the target pixel for a particular rook move. If the strength value of the target pixel of the rook move is greater than the strength of the opposite pixel, the opposite pixel value if reset to zero and the target pixel is set to equal the strength of the sequence. This process is performed for each rook move in a sequence and for each sequence of rook moves around the contour of the character. Once all rook moves have been evaluated, the value is assigned to the pixels of the character according to this process, are adjusted to reflect a bit map image to be displayed. This is achieved by turning on all bits which are greater than zero. The bit map generated is then logically ORed with the bit map generated using the contour fill process to provide a filled character in which drop out is eliminated and the distortion to the character is minimized.

    Envolvent approximation using accurate slope information
    24.
    发明授权
    Envolvent approximation using accurate slope information 失效
    使用准确的斜率信息进行逼近

    公开(公告)号:US6025851A

    公开(公告)日:2000-02-15

    申请号:US7971

    申请日:1998-01-16

    IPC分类号: G06T11/20 G06F15/00

    CPC分类号: G06T11/203

    摘要: A trajectory and a footprint of a hypothetical pen are each divided into sequences of segments, each segment having a slope associated therewith. The segments of the footprint form a closed sequence and those of the trajectory form a sequence with a beginning and an end. Each of the two approximate envolvents of the thick line are calculated in the same way, from the trajectory segments and the footprint segments. An envolvent is initialized to include the first segment of the trajectory starting at an appropriate offset from the trajectory start point, followed by zero or more footprint segments and then followed by the next trajectory segment until each of the trajectory segments are concatenated. In determining how many footprint segments to concatenate before concatenating the next trajectory segment, the slopes of the last concatenated trajectory segment and the next trajectory segment are compared to the slopes of the footprint segments. If any footprint segments have slopes between the slopes of the last and next trajectory segments, those footprint segments are concatenated before the next trajectory segment is concatenated. The selection of one of the two sets of footprint slope vectors, either the clockwise set or the counterclockwise set, determines which envolvent of the two envolvents are generated.

    摘要翻译: 假设笔的轨迹和足迹每一个分为段的序列,每个段具有与其相关联的斜率。 足迹的片段形成一个封闭的序列,轨迹的片段形成一个具有开始和结尾的序列。 从轨迹段和足迹段以相同的方式计算粗线的两个近似值。 一个envolvent被初始化为包括从与轨迹开始点的适当偏移开始的轨迹的第一段,随后是零个或多个覆盖区段,然后跟随下一个轨迹段,直到每个轨迹段被级联。 在确定在连接下一个轨迹段之前要连接多少足迹片段时,将最后连接的轨迹段和下一个轨迹段的斜率与足迹段的斜率进行比较。 如果任何足迹片段在最后一个和下一个轨迹段的斜率之间具有斜率,则在下一个轨迹段连接之前,这些足迹片段被级联。 选择两组脚印斜率矢量中的一组,即顺时针设置或逆时针旋转,都决定了两个输入轨迹的生成过程。

    Method and apparatus for minimizing the visual degradation of digital
typefaces-vertical adjustment
    25.
    发明授权
    Method and apparatus for minimizing the visual degradation of digital typefaces-vertical adjustment 失效
    用于最小化数字字体视觉退化的方法和装置 - 垂直调整

    公开(公告)号:US5304989A

    公开(公告)日:1994-04-19

    申请号:US846586

    申请日:1992-03-05

    摘要: A method and apparatus for minimizing the visual degradation of a typeface wherein the need for the manual input of skilled technicians is eliminated and the important visual components of each character and the relationship of the visual components of each character with the other characters of the typeface are preserved. The characters are analyzed to determine the visual components in the horizontal and vertical direction which comprise each character and the priority of adjustment. Using the visual components determined, starting with the highest priority component, the coordinates describing each character are adjusted according to rules which are dependent upon whether the strokes formed are oriented in the horizontal or vertical direction. The adjustment of the lower priority visual components is dependent on the previously aligned components. The important visual components in the diagonal direction are subsequently adjusted in accordance with similar sized visual components in the horizontal and vertical direction wherein the dimensions of the adjusted horizontal and vertical components are maintained.

    摘要翻译: 用于最小化字体的视觉退化的方法和装置,其中消除了熟练技术人员的手动输入的需要,并且每个角色的重要视觉部件以及每个角色的视觉部件与字体的其他字符之间的关系是 保存 分析字符以确定包括每个字符和调整优先级的水平和垂直方向上的视觉分量。 使用从最高优先级分量开始的视觉部件,描述每个角色的坐标根据依赖于所形成的笔画是否沿水平或垂直方向取向的规则来调整。 较低优先级的可视组件的调整取决于先前对齐的组件。 随后根据水平和垂直方向上相似尺寸的视觉部件调整对角线方向上的重要视觉部件,其中维持经调节的水平和垂直分量的尺寸。