利索-全球专利检索

  1. 登录
  2. 注册

搜索结果

8 条记录 (耗时 0.016 秒)

    RESOLUTION ENHANCEMENT BY NEAREST NEIGHBOR CLASSIFIED FILTERING
    1.
    发明公开
    RESOLUTION ENHANCEMENT BY NEAREST NEIGHBOR CLASSIFIED FILTERING 有权
    提高分辨率分类别近邻滤波

    公开(公告)号：EP1346311A4

    公开(公告)日：2008-07-02

    申请号：EP01986017

    申请日：2001-11-21

    申请人： SONY ELECTRONICS INC

    发明人： KONDO TETSUJIRO ,  CARRIG JAMES J ,  FUJIMORI YASUHIRO ,  CAREY WILLIAM KNOX

    IPC分类号： G06T3/40 ,  G06K9/62 ,  G06T5/20 ,  H04N7/01

    CPC分类号： G06K9/6272 ,  G06T3/4007 ,  G06T5/20

    摘要： A method and apparatus for enhancing image resolution by determining a nearest neighbor class for an input image vector from a plurality of spatial classes are disclosed. In one embodiment, the nearest neighbor class is determined by first receiving the input image vector that is to be classified into one of several spatial classes. Each spatial class has a corresponding normalized mean class vector. The input image vector is normalized. Then, the weighted distances from the normalized image vector to each normalized mean class vector are determined. The class vector which is the nearest neighbor class to the input image vector is determined based on the weighted distances. A filter corresponding to the nearest neighbor class is applied to the input image vector to enhance the resolution of the image.

    CLASSIFIED ADAPTIVE MULTIPLE PROCESSING SYSTEM
    7.
    发明公开
    CLASSIFIED ADAPTIVE MULTIPLE PROCESSING SYSTEM 审中-公开
    分类，自适应，多处理系统

    公开(公告)号：EP1188127A4

    公开(公告)日：2006-05-03

    申请号：EP00908618

    申请日：2000-02-10

    申请人： SONY ELECTRONICS INC

    发明人： KONDO TETSUJIRO ,  FUJIMORI YASUHIRO ,  GHOSAL SUGATA ,  CARRIG JAMES J

    IPC分类号： G06F17/10 ,  G06T1/20 ,  G06F17/17

    CPC分类号： G06F17/10

    摘要： An apparatus, method and computer-readable medium for selectively performing, in parallel structures, different classified adaptive processes on an input image, sound data or other correlated data. An input (901) is configured to receive the input data. A class generator (913) is coupled to the input (901) and configured to generate a multiple class ID for each classified adaptive process based upon the input data and a control signal. A filter tap selector (921) is coupled to the input (901) and the class generator (913). The filter tap selector (921) is configured to select filter tap data for each classified adaptive process based upon the input data, the control signal, and the multiple class ID. A coefficient memory (925) is coupled to the class generator (913) and configured to provide coefficient data for each classified adaptive process for each data based upon the control signal and the multiple class ID. A filter (933) is coupled to the filter tap selector (921) and the coefficient memory (925). The filter (933) is configured to generate output data (941) based upon the filter tap data and the coefficient data.

    METHOD AND APPARATUS FOR ADAPTIVE CLASS TAP SELECTION ACCORDING TO MULTIPLE CLASSIFICATION
    8.
    发明公开
    METHOD AND APPARATUS FOR ADAPTIVE CLASS TAP SELECTION ACCORDING TO MULTIPLE CLASSIFICATION 有权
    方法和设备的学习给符合条件的学生选课MEHRERN分类

    公开(公告)号：EP1222614A4

    公开(公告)日：2005-09-21

    申请号：EP00908627

    申请日：2000-02-11

    申请人： SONY ELECTRONICS INC

    发明人： KONDO TETSUJIRO ,  FUJIMORI YASUHIRO ,  GHOSAL SUGATA ,  CARRIG JAMES J

    IPC分类号： G06T5/00 ,  G06T5/20 ,  H04N5/21 ,  H04N19/895 ,  G06K9/40 ,  H04N5/235 ,  H04N7/26 ,  H04N7/36 ,  H04N7/68

    CPC分类号： H04N5/21 ,  G06T5/20 ,  H04N19/503 ,  H04N19/80 ,  H04N19/895 ,  H04N19/98

    摘要： A method, apparatus, and article of manufacture for restoring a deteriorated signal (201) to an undeteriorated signal (213). A deteriorated signal (201) consists of a plurality of deteriorated and undeteriorated data points. For each deteriorated data point, a plurality of class types is created based upon characteristics of the area containing the deteriorated data point. The data point is classified with respect to one of the plurality of class types and assigned a corresponding input signal class (205). The undeteriorated signal (213) is generated by adaptively filtering the deteriorated input signal (201) in accordance with the input signal classification result (209). More than one classification method is used to create the plurality of class types. Created classes may include a motion class (1809), an error class (1811), a spatial class (1807) or a spatial activity class. An adaptive class tap structure (1813) is used to create the plurality of class types for improvement of classification quality. The spatial class taps can be selected adaptively according to the motion and/or error class. The spatial activity class taps can be selected adaptively according to the motion and/or error class.