Image processing method and apparatus

    公开(公告)号:US09781400B2

    公开(公告)日:2017-10-03

    申请号:US15065061

    申请日:2016-03-09

    CPC classification number: H04N9/76 G06T7/12 G06T7/194 H04N9/77

    Abstract: An image processing method and apparatus, where the method includes determining m boundary points of a target image, acquiring a color component of a non-boundary point in a jth area, where the jth area is a neighborhood of the jth boundary point in the m boundary points, 1≦j≦m, and m is a positive integer greater than or equal to 1, and performing synthesis processing according to the color component of the non-boundary point in the jth area, to obtain a color component of the jth boundary point, and by means of processing on a boundary point of a target image, precision of image matting and synthesis processing may be improved, and may be applied to a real-time image matting and synthesis process.

    Image Anti-Aliasing Method and Apparatus
    2.
    发明申请
    Image Anti-Aliasing Method and Apparatus 审中-公开
    图像抗锯齿方法和装置

    公开(公告)号:US20160110847A1

    公开(公告)日:2016-04-21

    申请号:US14820767

    申请日:2015-08-07

    CPC classification number: G06T5/00 G06F3/013 G06T5/002 G06T2200/12

    Abstract: An image anti-aliasing method and apparatus are provided, where the method includes acquiring a to-be-displayed image in a terminal, acquiring a current distance parameter of the terminal, where the current distance parameter is used to indicate a distance between an eye of a user of the terminal and a display screen of the terminal, determining a current anti-aliasing algorithm according to the current distance parameter by using a correspondence between a distance parameter and an anti-aliasing algorithm, and performing anti-aliasing processing on the to-be-displayed image according to the current anti-aliasing algorithm. In embodiments of the present disclosure, anti-aliasing processing is performed on a to-be-displayed image based on a distance parameter, that is, an anti-aliasing processing manner of the image is dynamically adjusted according to the distance parameter.

    Abstract translation: 提供了一种图像抗混叠方法和装置,其中该方法包括在终端中获取要显示的图像,获取终端的当前距离参数,其中当前距离参数用于指示眼睛之间的距离 终端的用户和终端的显示屏幕,通过使用距离参数和抗混叠算法之间的对应关系,根据当前距离参数确定当前的抗混叠算法,并且对 根据当前的抗锯齿算法来显示图像。 在本公开的实施例中,基于距离参数对待显示的图像执行抗混叠处理,即,根据距离参数动态地调整图像的抗混叠处理方式。

    Method, apparatus, and device for detecting convex polygon image block
    3.
    发明授权
    Method, apparatus, and device for detecting convex polygon image block 有权
    用于检测凸多边形图像块的方法,装置和装置

    公开(公告)号:US09489595B2

    公开(公告)日:2016-11-08

    申请号:US14572210

    申请日:2014-12-16

    CPC classification number: G06K9/6217 G06K9/4604 G06K9/6202 G06T7/70

    Abstract: The present invention discloses a method and an apparatus for detecting a convex polygon image block, and a device. The method includes: acquiring an image collected by an image sensor; acquiring a corner set, where the corner set includes a corner in the image, and the corner is an intersection point at which any two straight line sides that are not parallel to each other in the image intersect; and detecting a convex polygon image block in the image according to the corner set, where a shape of the convex polygon image block is a preset shape and a similarity between the convex polygon image block and a preset image block exceeds a preset first threshold. The apparatus includes: a first acquiring module, a second acquiring module, and a detecting module. In the present invention, the convex polygon image block can be detected from the image.

    Abstract translation: 本发明公开了一种用于检测凸多边形图像块的方法和装置以及装置。 该方法包括:获取由图像传感器收集的图像; 获取角集合,其中角集合包括图像中的角,并且角是与图像中彼此不平行的任何两条直线侧相交的交点; 并且根据角集合检测图像中的凸多边形图像块,其中,所述凸多边形图像块的形状为预设形状,并且所述凸多边形图像块和预设图像块之间的相似度超过预设的第一阈值。 该装置包括:第一获取模块,第二获取模块和检测模块。 在本发明中,可以从图像中检测凸多边形图像块。

    Image Processing Method and Apparatus
    4.
    发明申请
    Image Processing Method and Apparatus 有权
    图像处理方法和装置

    公开(公告)号:US20160269708A1

    公开(公告)日:2016-09-15

    申请号:US15065061

    申请日:2016-03-09

    CPC classification number: H04N9/76 G06T7/12 G06T7/194 H04N9/77

    Abstract: An image processing method and apparatus, where the method includes determining m boundary points of a target image, acquiring a color component of a non-boundary point in a jth area, where the jth area is a neighborhood of the jth boundary point in the m boundary points, 1≦j≦m, and m is a positive integer greater than or equal to 1, and performing synthesis processing according to the color component of the non-boundary point in the jth area, to obtain a color component of the jth boundary point, and by means of processing on a boundary point of a target image, precision of image matting and synthesis processing may be improved, and may be applied to a real-time image matting and synthesis process.

    Abstract translation: 一种图像处理方法和装置,其中所述方法包括确定目标图像的m个边界点,获取第j个区域中的非边界点的颜色分量,其中第j个区域是m个第j个边界点的邻域 边界点1≤j≤m,m是大于或等于1的正整数,并且根据第j区域中的非边界点的颜色分量进行合成处理,以获得第j个区域的颜色分量 边界点,并且通过对目标图像的边界点进行处理,可以提高图像消隐和合成处理的精度,并且可以应用于实时图像消光和合成处理。

Patent Agency Ranking