首页> 外国专利> Non-integer scaling of raster images with image quality enhancement using an anamorphically scaled intermediate bitmap

Non-integer scaling of raster images with image quality enhancement using an anamorphically scaled intermediate bitmap

机译:使用变形比例缩放的中间位图提高图像质量的光栅图像的非整数缩放

摘要

A method for scaling an input bitmap by a non-integer factor with improved image quality comprises the steps of: selecting a pel in the input bitmap; selecting a neighborhood surrounding the selected pel; searching a set of image feature patterns to find an image feature pattern that matches the selected neighborhood; and writing a blocks of bits to an anamorphically scaled intermediate bitmap based on the match. The intermediate bitmap has a resolution in a first direction (row or horizontal direction) which is greater than that of the desired output resolution and has resolution in a second direction (column or vertical direction) which is equal to that of the desired output. The appropriate block of bits is obtained by subsampling an image enhancing tile which corresponds to the matched image feature pattern. The process of matching neighborhoods and writing a subsampled image enhancing tile (i.e., block of bits) to the intermediate bitmap is repeated for each pel in the input bitmap. When the image enhancing tiles are subsampled to the proper size and shape and placed in the proper locations in the intermediate bitmap, the jagged "stair-step" look of curved and diagonal image features in the input image is smoothed, resulting in improved quality. The intermediate rows of the bitmap are then scaled down to a selected output resolution and written into an output bitmap with improved image quality and resolution greater than that of the input bitmap. This method is particularly useful for scaling raster font characters from a given input resolution to the resolution required by an output device when the resolution of the output device is a non-integer multiple of the input resolution.
机译:一种通过具有改善的图像质量的非整数因子缩放输入位图的方法,包括以下步骤:在输入位图中选择像素;选择所选象素周围的邻域;搜索一组图像特征图案以找到与所选邻域匹配的图像特征图案;并根据该匹配将一个位块写入一个按比例缩放的中间位图。中间位图在第一方向(行或水平方向)上的分辨率大于期望输出分辨率的分辨率,并且在第二方向(列或垂直方向)上具有等于期望输出的分辨率的分辨率。通过对与匹配的图像特征图案相对应的图像增强块进行二次采样来获得适当的比特块。对于输入位图中的每个像素,重复进行邻域匹配并将子采样的图像增强图块(即位块)写入中间位图的过程。当对图像增强图块进行二次采样以达到适当的大小和形状并将其放置在中间位图中的适当位置时,输入图像中弯曲和对角图像特征的锯齿状“阶梯式”外观将变得平滑,从而提高了质量。然后将位图的中间行按比例缩小到选定的输出分辨率,并以比输入位图更好的图像质量和更高的分辨率写入输出位图。当输出设备的分辨率是输入分辨率的非整数倍时,此方法对于将光栅字体字符从给定的输入分辨率缩放到输出设备所需的分辨率特别有用。

著录项

相似文献

  • 专利
  • 外文文献
  • 中文文献
获取专利

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号