首页> 外国专利> A METHOD FOR EXTENDING AN IMAGE AT THE IMAGE BOUNDARY AND A METHOD AND APPARATUS FOR FILTERING THE IMAGE IN A DIGITAL SIGNAL PROCESSING SYSTEM

A METHOD FOR EXTENDING AN IMAGE AT THE IMAGE BOUNDARY AND A METHOD AND APPARATUS FOR FILTERING THE IMAGE IN A DIGITAL SIGNAL PROCESSING SYSTEM

机译:在图像边界处扩展图像的方法以及在数字信号处理系统中滤波图像的方法和装置

摘要

T is a plurality of odd integers, each having a T kernel function weighting coefficient, the first segment being one of the segments is the first unfiltered plurality of pixels, and the second segment near the segment is the second unfiltered A plurality of pixels, wherein a predetermined one of the unfiltered pixels includes at least two of the at least two symmetric T-tap digital filters common to both the first and second plurality of pixels at the image boundary of the two segments. In a multi-rate digital signal processing system for resizing an image to a segment, (T-2) / of the plurality of unfiltered first pixels and the unfiltered plurality of second pixels near a predetermined one of the unfiltered pixels. Multiply 2 by two times one of each kernel function weighting factor, so that the first and second mirrored half kernel middle Generating an accumulation, multiplying a predetermined pixel of the pixel by the center tap kernel function weighting coefficient of the digital filter, to generate another intermediate product, and generating the first mirrored half kernel product and the first mirrored half kernel product. Adding another intermediate product to produce a first segment estimate boundary filtering output pixel, and adding the second mirrored half kernel intermediate product and the another intermediate product to add a second segment estimate boundary filtering output pixel. And averaging the first segment estimate boundary filtering output pixel and the segment, thereby seamlessly extending the image across the image boundary.
机译:T是多个奇整数,每个均具有T核函数加权系数,其中,第一段是所述多个段中的一个,是第一未滤波的多个像素,第二段附近是第二未滤波的多个A像素,其中未过滤的像素中的预定一个包括至少两个对称的T抽头数字滤波器中的至少两个,这两个对称的T抽头数字滤波器对于两个分段的图像边界处的第一多个像素和第二多个像素都是相同的。在用于将图像调整为片段大小的多速率数字信号处理系统中,多个(T-2)/个未滤波的第一像素和未滤波的多个第二像素中的一个接近未滤波像素中的预定一个。将每个内核函数加权因子乘以2,从而使第一和第二个镜像半内核中间产生一个累加,将像素的预定像素乘以数字滤波器的中心抽头内核函数加权系数,以生成另一个中间产品,并生成第一个镜像的半内核产品和第一个镜像的半内核产品。添加另一个中间乘积以生成第一段估计边界滤波输出像素,并且添加第二镜像半内核中间乘积和另一个中间乘积以添加第二段估计边界滤波输出像素。并对第一段估计边界滤波输出像素和该段的平均值,从而跨图像边界无缝扩展图像。

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