公开(公告)号：US08787653B2

公开(公告)日：2014-07-22

申请号：US13426203

申请日：2012-03-21

Applicant: Xing Li ,  Meng Yao ,  Barbara L. Farrell

Inventor： Xing Li ,  Meng Yao ,  Barbara L. Farrell

IPC: G06K9/00

CPC classification number: H04N1/58 ,  G06T1/20

Abstract: What is disclosed is a system and method for determining an orientation direction of a color edge at a given pixel location in a binary color image. The orientation direction of the color edge is determined from eight pixel counts with each pixel count being a total number of pixels in each of eight regions of a window centered about a candidate pixel which resides along the color edge. The eight regions are associated with 8 compass points. The orientation of the edge is determined by a 1st, 2nd and 3rd tier control bits which are based upon the pixel counts of each region. The 1st, 2nd and 3rd tier control bits collectively form a 3-bit word. The 3-bit word defines the orientation direction. The teachings hereof provide an efficient way of performing binary edge orientation detection by making uses of intermediate results to simultaneously encode the edge orientation.

Abstract translation: 所公开的是用于确定二值彩色图像中的给定像素位置处的颜色边缘的取向方向的系统和方法。 颜色边缘的取向方向由八个像素计数确定，每个像素数量是围绕沿着颜色边缘驻留的候选像素的中心的窗口的八个区域的每个区域中的像素总数。 八个区域与8个罗盘点相关联。 边缘的方向由基于每个区域的像素计数的第一，第二和第三层控制位确定。 第1，第2和第3层控制位一起形成3位字。 3位字定义方向。 本发明的教导提供了一种通过利用中间结果同时编码边缘取向来执行二进制边缘取向检测的有效方法。

公开(公告)号：US08611671B2

公开(公告)日：2013-12-17

申请号：US12841663

申请日：2010-07-22

Applicant: Meng Yao ,  Xing Li

Inventor： Meng Yao ,  Xing Li

IPC: G06K9/46

CPC classification number: H04N1/4092 ,  G06K15/1874

Abstract: A methodology for thin line detection and enhancement in electronic images is disclosed. The methodology includes associating an electronic image with at least one basic context window that is less than the size of the electronic image based on the input image resolution of the electronic image; detecting one or more predefined patterns which correspond to thin lines in the electronic image using the at least one basic context window; excluding patterns for the one or more detected patterns which are halftone patterns; and adding at least one pixel to the electronic image based on at least one of the remaining patterns so as to enhance thin line features in the electronic image. In some implementation, the methodology may be configured to handle electronic images having different resolutions. A system for thin line detection and enhancement in electronic images having different resolutions is also disclosed.

Abstract translation: 公开了一种用于电子图像中细线检测和增强的方法。 该方法包括基于电子图像的输入图像分辨率将电子图像与至少一个小于电子图像大小的基本上下文窗口相关联; 使用所述至少一个基本上下文窗口来检测对应于所述电子图像中的细线的一个或多个预定义模式; 排除作为半色调图案的一个或多个检测图案的图案; 以及基于所述剩余图案中的至少一个向所述电子图像添加至少一个像素，以便增强所述电子图像中的细线特征。 在一些实现中，该方法可以被配置为处理具有不同分辨率的电子图像。 还公开了一种用于具有不同分辨率的电子图像中的细线检测和增强的系统。

公开(公告)号：US20120020570A1

公开(公告)日：2012-01-26

申请号：US12841663

申请日：2010-07-22

Applicant: Meng Yao ,  Xing Li

Inventor： Meng Yao ,  Xing Li

IPC: G06K9/54

CPC classification number: H04N1/4092 ,  G06K15/1874

Abstract: A methodology for thin line detection and enhancement in electronic images is disclosed. The methodology includes associating an electronic image with at least one basic context window that is less than the size of the electronic image based on the input image resolution of the electronic image; detecting one or more predefined patterns which correspond to thin lines in the electronic image using the at least one basic context window; excluding patterns for the one or more detected patterns which are halftone patterns; and adding at least one pixel to the electronic image based on at least one of the remaining patterns so as to enhance thin line features in the electronic image. In some implementation, the methodology may be configured to handle electronic images having different resolutions. A system for thin line detection and enhancement in electronic images having different resolutions is also disclosed.

Abstract translation: 公开了一种用于电子图像中细线检测和增强的方法。 该方法包括基于电子图像的输入图像分辨率将电子图像与至少一个小于电子图像大小的基本上下文窗口相关联; 使用所述至少一个基本上下文窗口来检测对应于所述电子图像中的细线的一个或多个预定义模式; 排除作为半色调图案的一个或多个检测图案的图案; 以及基于所述剩余图案中的至少一个向所述电子图像添加至少一个像素，以便增强所述电子图像中的细线特征。 在一些实现中，该方法可以被配置为处理具有不同分辨率的电子图像。 还公开了一种用于具有不同分辨率的电子图像中的细线检测和增强的系统。

公开(公告)号：US20110051197A1

公开(公告)日：2011-03-03

申请号：US12548783

申请日：2009-08-27

Applicant: Meng Yao ,  John A. Handwork ,  Xing Li

Inventor： Meng Yao ,  John A. Handwork ,  Xing Li

IPC: H04N1/60

CPC classification number: H04N1/58

Abstract: Provided are bitmap based trapping methods, apparatus and systems. According to one exemplary method, black trapping color image data is performed by estimating the continuous tone values associated with non-black pixels near a qualified black pixel and subsequently, the estimated continuous tone values are halftoned at the qualified black pixel locations and ORed with the original bitmap data.

Abstract translation: 提供了基于位图的陷印方法，装置和系统。 根据一个示例性方法，通过估计与合格的黑色像素附近的非黑色像素相关联的连续色调值来执行黑色捕获彩色图像数据，随后将估计的连续色调值在合格的黑色像素位置处半色调并与 原始位图数据。

公开(公告)号：US09307119B2

公开(公告)日：2016-04-05

申请号：US13587331

申请日：2012-08-16

Applicant: Xing Li ,  David Jon Metcalfe ,  Meng Yao ,  Michael Dale Stevens ,  Paul W. Phillippi

Inventor： Xing Li ,  David Jon Metcalfe ,  Meng Yao ,  Michael Dale Stevens ,  Paul W. Phillippi

IPC: H04N1/56 ,  G06K1/12

CPC classification number: H04N1/56

Abstract: Embodiments relate to systems and methods for a computation-efficient image processing system architecture. Image data can be transmitted from a computer, online service, and/or other image source to an output device having a set of image processing modules in two or more image paths, including an edge detection module and a video decoding module. The edge detection module can produce edge tag output, and the video decoding module, operating in parallel, can generate decoded video output. The edge tag output and decoded video output can be transmitted to a set of downstream image processing modules, including modules for color trapping, edge smoothing, and other operations. Because earlier processing stages share information with downstream modules which require the same or related data, redundant processing can be reduced or eliminated. Complex image operations can therefore be carried out, and high-quality output can be generated, without sacrificing responsiveness.

Abstract translation: 实施例涉及用于计算高效图像处理系统架构的系统和方法。 图像数据可以从计算机，在线服务和/或其他图像源发送到包括边缘检测模块和视频解码模块的两个或多个图像路径中的具有一组图像处理模块的输出装置。 边缘检测模块可以产生边缘标签输出，并行的视频解码模块可以产生解码的视频输出。 边缘标签输出和解码视频输出可以传输到一组下游图像处理模块，包括用于色彩捕捉，边缘平滑和其他操作的模块。 因为较早的处理阶段与需要相同或相关数据的下游模块共享信息，所以可以减少或消除冗余处理。 因此，可以执行复杂的图像操作，并且可以产生高质量的输出，而不牺牲响应性。

公开(公告)号：US08792706B2

公开(公告)日：2014-07-29

申请号：US13426308

申请日：2012-03-21

Applicant: Xing Li ,  Meng Yao ,  Barbara L. Farrell

Inventor： Xing Li ,  Meng Yao ,  Barbara L. Farrell

IPC: G06K9/00

CPC classification number: H04N1/58

Abstract: What is disclosed is a system and method for determining an orientation direction of a color edge at a given pixel location in a binary color image. The orientation direction of the color edge is determined from eight pixel counts with each pixel count being a total number of pixels in each of eight regions of a window centered about a candidate pixel which resides along the color edge. The eight regions are associated with 8 compass points. The orientation of the edge is determined by a 1st, 2nd and 3rd tier control bits which are based upon the pixel counts of each region. The 1st, 2nd and 3rd tier control bits collectively form a 3-bit word. The 3-bit word defines the orientation direction. The teachings hereof provide an efficient way of performing binary edge orientation detection by making uses of intermediate results to simultaneously encode the edge orientation.

公开(公告)号：US20140050269A1

公开(公告)日：2014-02-20

申请号：US13587331

申请日：2012-08-16

Applicant: Xing Li ,  David Jon Metcalfe ,  Meng Yao ,  Michael Dale Stevens ,  Paul W. Phillippi

Inventor： Xing Li ,  David Jon Metcalfe ,  Meng Yao ,  Michael Dale Stevens ,  Paul W. Phillippi

IPC: H04N7/26

CPC classification number: H04N1/56

Abstract: Embodiments relate to systems and methods for a computation-efficient image processing system architecture. Image data can be transmitted from a computer, online service, and/or other image source to an output device having a set of image processing modules in two or more image paths, including an edge detection module and a video decoding module. The edge detection module can produce edge tag output, and the video decoding module, operating in parallel, can generate decoded video output. The edge tag output and decoded video output can be transmitted to a set of downstream image processing modules, including modules for color trapping, edge smoothing, and other operations. Because earlier processing stages share information with downstream modules which require the same or related data, redundant processing can be reduced or eliminated. Complex image operations can therefore be carried out, and high-quality output can be generated, without sacrificing responsiveness.

Abstract translation: 实施例涉及用于计算高效图像处理系统架构的系统和方法。 图像数据可以从计算机，在线服务和/或其他图像源发送到包括边缘检测模块和视频解码模块的两个或多个图像路径中的具有一组图像处理模块的输出装置。 边缘检测模块可以产生边缘标签输出，并行的视频解码模块可以产生解码的视频输出。 边缘标签输出和解码视频输出可以传输到一组下游图像处理模块，包括用于色彩捕捉，边缘平滑和其他操作的模块。 因为较早的处理阶段与需要相同或相关数据的下游模块共享信息，所以可以减少或消除冗余处理。 因此，可以执行复杂的图像操作，并且可以产生高质量的输出，而不牺牲响应性。

公开(公告)号：US08437045B2

公开(公告)日：2013-05-07

申请号：US12548783

申请日：2009-08-27

Applicant: Meng Yao ,  John A. Handwork ,  Xing Li

Inventor： Meng Yao ,  John A. Handwork ,  Xing Li

IPC: H04N1/405 ,  H04N7/30

CPC classification number: H04N1/58

Abstract: Provided are bitmap based trapping methods, apparatus and systems. According to one exemplary method, black trapping color image data is performed by estimating the continuous tone values associated with non-black pixels near a qualified black pixel and subsequently, the estimated continuous tone values are halftoned at the qualified black pixel locations and ORed with the original bitmap data.

Abstract translation: 提供了基于位图的陷印方法，装置和系统。 根据一个示例性方法，通过估计与合格的黑色像素附近的非黑色像素相关联的连续色调值来执行黑色捕获彩色图像数据，随后将估计的连续色调值在合格的黑色像素位置处半色调并与 原始位图数据。

公开(公告)号：US08830520B2

公开(公告)日：2014-09-09

申请号：US12884311

申请日：2010-09-17

Applicant: Meng Yao

Inventor： Meng Yao

IPC: B41B15/00 ,  H04N1/407 ,  H04N1/60

CPC classification number: H04N1/407 ,  H04N1/6022

Abstract: A computer-implemented method and system for enhancing black density of a halftoned bitmap are provided. The method includes receiving a halftoned bitmap into computer memory, and, using a computer, identifying at least one black-only pixel in the halftoned bitmap. The method further includes for each of the identified black-only pixels, identifying at least one black-only pixel as a candidate for adding color based at least in part on the location of the black-only pixel with respect to an edge in the halftoned bitmap, modifying the halftoned bitmap by adding color to at least one of the candidate black-only pixels, and outputting the modified halftoned bitmap.

Abstract translation: 提供了一种用于增强半色调位图的黑色密度的计算机实现的方法和系统。 该方法包括将半色调位图接收到计算机存储器中，并且使用计算机识别半色调位图中的至少一个仅黑色像素。 该方法还包括针对每个识别的仅黑色像素，至少部分地基于黑色像素相对于半色调像素中的边缘的位置，将至少一个仅黑色像素识别为用于添加颜色的候选 位图，通过将颜色添加到候选的仅黑色像素中的至少一个来修改半色调位图，并输出修改的半色调位图。

公开(公告)号：US20120069357A1

公开(公告)日：2012-03-22

申请号：US12884311

申请日：2010-09-17

Applicant: Meng Yao

Inventor： Meng Yao

IPC: H04N1/60

CPC classification number: H04N1/407 ,  H04N1/6022

Abstract: A computer-implemented method and system for enhancing black density of a halftoned bitmap are provided. The method includes receiving a halftoned bitmap into computer memory, and, using a computer, identifying at least one black-only pixel in the halftoned bitmap. The method further includes for each of the identified black-only pixels, identifying at least one black-only pixel as a candidate for adding color based at least in part on the location of the black-only pixel with respect to an edge in the halftoned bitmap, modifying the halftoned bitmap by adding color to at least one of the candidate black-only pixels, and outputting the modified halftoned bitmap.

Abstract translation: 提供了一种用于增强半色调位图的黑色密度的计算机实现的方法和系统。 该方法包括将半色调位图接收到计算机存储器中，并且使用计算机识别半色调位图中的至少一个仅黑色像素。 该方法还包括针对每个识别的仅黑色像素，至少部分地基于黑色像素相对于半色调像素中的边缘的位置，将至少一个仅黑色像素识别为用于添加颜色的候选 位图，通过将颜色添加到候选的仅黑色像素中的至少一个来修改半色调位图，并输出修改的半色调位图。