专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

91. 发明专利

JP2003296741A Method, program and device for cutting object image 有权
标题翻译：用于切割对象图像的方法，程序和设备
公开(公告)号：JP2003296741A
公开(公告)日：2003-10-17
申请号：JP2002097760
申请日：2002-03-29
申请人： Toshiba Corp , 株式会社東芝
发明人： MATSUMOTO NOBUYUKI , HORI OSAMU , IDA TAKASHI , TAKESHIMA HIDENORI
IPC分类号： G06T1/00 , G06T3/00 , G06T7/60 , H04N5/272 , H04N5/275
CPC分类号： H04N5/275 , H04N5/272
摘要： PROBLEM TO BE SOLVED: To provide an object image cutting device for generating an alpha image representing respective object areas for a plurality of source images. SOLUTION: This object image cutting device is provided with a storing means 101 for storing a plurality of source images including objects to be cut and a plurality of reference alpha images prepared in advance and representing object areas, a deformation parameter determining means 103 for generating an output alpha image of a reference source image to be reference by deforming a standard reference alpha image corresponding to the reference source image and determining a deformation parameter on the basis of the output alpha image, and an output alpha image preparing means 104 for generating output alpha images of remaining source images by deforming remaining reference alpha images on the basis of the deformation parameter. COPYRIGHT: (C)2004,JPO
摘要翻译：要解决的问题：提供一种用于生成表示多个源图像的各个对象区域的α图像的对象图像切割装置。解决方案：该对象图像切割装置设置有存储装置101，用于存储包括预先准备并表示对象区域的待切割对象和多个参考α图像的多个源图像，变形参数确定装置103 用于通过使与参考源图像相对应的标准参考α图像变形并基于输出的α图像确定变形参数来生成参考源图像的输出alpha图像;以及输出α图像准备装置104，用于通过基于变形参数使剩余的参考α图像变形来产生剩余源图像的输出α图像。版权所有（C）2004，JPO

92. 发明专利

JP2010211300A Image processing device and method 有权
标题翻译：图像处理装置和方法
公开(公告)号：JP2010211300A
公开(公告)日：2010-09-24
申请号：JP2009053900
申请日：2009-03-06
申请人： Toshiba Corp , 株式会社東芝
发明人： SAITO KANAKO , TAGUCHI YASUNORI , IDA TAKASHI
IPC分类号： G06T3/00 , H04N1/387
CPC分类号： G06T11/001
摘要： PROBLEM TO BE SOLVED: To improve texture by adding a new texture to an image while maintaining the shadow of an image to be converted and the direction of the texture. SOLUTION: The target image 106 is transferred first to a sample texture image generation part 101. The sample texture image generation part 101 generates a sample texture image 107 from which the shadow as an image component of a smooth change pattern of intensity is removed, from the target image 106. The texture image generation part 102 generates a texture image 108 of the same size as a composite image 110 to be output, from the sample texture image 107. A composition part 104 combines an input base image 105 and a texture image 108 which is generated by the texture image generation part 102 to generate the composite image 110. COPYRIGHT: (C)2010,JPO&INPIT
摘要翻译：要解决的问题：通过在保持要转换的图像的阴影和纹理方向的同时向图像添加新的纹理来改善纹理。解决方案：将目标图像106首先传送到样本纹理图像生成部101.样本纹理图像生成部101生成样本纹理图像107，作为强度的平滑变化图案的图像成分的阴影为从目标图像106移除。纹理图像生成部102从样本纹理图像107生成与要输出的合成图像110相同大小的纹理图像108.组合部104将输入基底图像105和由纹理图像生成部102生成的纹理图像108，以生成合成图像110.（C）2010，JPO＆INPIT

93. 发明专利

JP2010198499A Image processing device and method 有权
标题翻译：图像处理装置和方法
公开(公告)号：JP2010198499A
公开(公告)日：2010-09-09
申请号：JP2009044765
申请日：2009-02-26
申请人： Toshiba Corp , 株式会社東芝
发明人： TAGUCHI YASUNORI , IDA TAKASHI , KANEKO TOSHIMITSU , MATSUMOTO NOBUYUKI , TAKESHIMA HIDENORI , ONO TOSHIYUKI
IPC分类号： G06T3/40 , H04N1/393
摘要： PROBLEM TO BE SOLVED: To improve restoration performance while holding the rapidity of processing. SOLUTION: This image processing device includes: a determining part 102 for determining the size of a low block, the size of a high block, and the size of a processing block including the low block; a setting part 103 for sequentially setting the positions of low blocks; a searching part 104 for searching temporary correspondence blocks whose difference from the low block is within a first threshold from a low resolution image with integer accuracy, searching a correspondence range whose difference from the low block is within a second threshold and which is different from the range of the low block from surroundings of the temporary correspondence block with decimal accuracy, and calculating a relative position between the temporary correspondence block and a correspondence range; a determining part 105 for determining a coefficient for calculating the high block from the correspondence block which is included in the correspondence range and is a partial block of the temporary correspondence block and a processing block from the relative position so that the high block may satisfy a reconfiguration restriction to the low block and the correspondence block; and a preparing part 106 for preparing a high block with convolution of the processing block, the correspondence block and the coefficient. COPYRIGHT: (C)2010,JPO&INPIT
摘要翻译：要解决的问题：提高恢复性能，同时保持加工的快速性。解决方案：该图像处理装置包括：确定部件102，用于确定低块的尺寸，高块的尺寸以及包括低块的处理块的尺寸; 用于顺序地设置低块的位置的设置部103; 搜索部104，用于从具有整数精度的低分辨率图像中搜索与低块的差异在第一阈值内的临时对应块，搜索与低块不同的对应范围在第二阈值内，并且与并且计算临时对应块与对应范围之间的相对位置;以及计算临时对应块与对应范围之间的相对位置; 确定部分105，用于从对应范围中包括的对应块中确定用于计算高块的系数，并且是临时对应块的部分块和来自相对位置的处理块，使得高块可满足对低块和对应块的重配置限制; 以及准备部分106，用于准备具有处理块，对应块和系数的卷积的高块。版权所有（C）2010，JPO＆INPIT

94. 发明专利

JP2010034696A Resolution converting device, method, and program 有权
标题翻译：分辨率转换设备，方法和程序
公开(公告)号：JP2010034696A
公开(公告)日：2010-02-12
申请号：JP2008192533
申请日：2008-07-25
申请人： Toshiba Corp , 株式会社東芝
发明人： TAKESHIMA HIDENORI , KANEKO TOSHIMITSU , IDA TAKASHI
IPC分类号： H04N7/01 , G06T3/40 , H04N1/387
CPC分类号： G06T3/4069
摘要： PROBLEM TO BE SOLVED: To obtain high restoring precision while suppressing the increase of calculation quantity and a noise. SOLUTION: A resolution converting device is provided with: a means 1005 for setting a temporary corresponding point density function for inputting an N-dimensional coordinate value on an N-dimensional space, and for outputting the density value of a temporary corresponding point designated by the N-dimensional coordinate value; a means 1005 for setting the set of temporary corresponding points with the density value satisfying the temporary corresponding point density function; a means 1002 for acquiring, from target data, an input pattern including pixel values within a first range with reference to each of the temporary corresponding points; a means 1002 for acquiring, from search pattern data, a candidate pattern including pixel values within a second range; a means 1005 for finding, from the search pattern data, an estimated pattern as a candidate pattern of the candidate patterns which has a smaller difference from the input pattern than other candidate patterns; a means 1005 for setting the pixel values of positions corresponding to the temporary corresponding points in the estimated pattern as pixel values at the temporary corresponding points; and a means 1005 for converting the resolution of processing target data by using the set pixel values and the N-dimensional coordinate values of the temporary corresponding points. COPYRIGHT: (C)2010,JPO&INPIT
摘要翻译：要解决的问题：为了在抑制计算量和噪声的增加的同时获得高的恢复精度。解决方案：分辨率转换装置设置有：用于设置临时对应点密度函数的装置1005，用于在N维空间上输入N维坐标值，并输出临时对应点的浓度值由N维坐标值指定; 用于设置具有满足临时对应点密度函数的密度值的一组临时对应点的装置1005; 参考每个临时对应点从目标数据获取包括在第一范围内的像素值的输入图案的装置1002; 用于从搜索图案数据获取包括第二范围内的像素值的候选图案的装置1002; 用于从搜索模式数据中找出作为候选模式的候选模式的估计模式的装置1005，其与输入模式的差异小于其他候选模式; 用于将与估计图案中的临时对应点相对应的位置的像素值设置为临时对应点处的像素值的装置1005; 以及用于通过使用设定的像素值和临时对应点的N维坐标值来转换处理目标数据的分辨率的装置1005。版权所有（C）2010，JPO＆INPIT

95. 发明专利

JP2009211626A Image processing device and method 有权
标题翻译：图像处理装置和方法
公开(公告)号：JP2009211626A
公开(公告)日：2009-09-17
申请号：JP2008056338
申请日：2008-03-06
申请人： Toshiba Corp , 株式会社東芝
发明人： IDA TAKASHI , MATSUMOTO NOBUYUKI
IPC分类号： G06T7/60
CPC分类号： G06T7/13 , G06T2207/20192
摘要： PROBLEM TO BE SOLVED: To provide an image processing device capable of detecting only a long edge area which is not a texture area. SOLUTION: In an arrangement part, each of pixel values of a template block containing a determination pixel is calculated for the determination pixel to be sequentially set up within an image, and a plurality of reference blocks are arranged so that the template block is surrounded. Block matching errors between the respective pixel values of the plurality of reference blocks and the respective pixel values of the template block are obtained, respectively. When the minimum value among the block matching errors is a deviation value in the whole of the plurality of block matching errors, the determination pixel is determined to exist in the edge area. COPYRIGHT: (C)2009,JPO&INPIT
摘要翻译：要解决的问题：提供一种能够仅检测不是纹理区域的长边区域的图像处理装置。解决方案：在布置部分中，针对要在图像内顺序建立的确定像素计算包含确定像素的模板块的每个像素值，并且布置多个参考块，使得模板块被包围分别获得多个参考块的各个像素值与模板块的各个像素值之间的块匹配误差。当块匹配误差中的最小值是整个多个块匹配误差中的偏差值时，确定确定像素存在于边缘区域中。版权所有（C）2009，JPO＆INPIT

96. 发明专利

JP2009100407A Image processing apparatus and method thereof 审中-公开
标题翻译：图像处理装置及其方法
公开(公告)号：JP2009100407A
公开(公告)日：2009-05-07
申请号：JP2007272295
申请日：2007-10-19
申请人： Toshiba Corp , 株式会社東芝
发明人： TAGUCHI YASUNORI , IDA TAKASHI , KANEKO TOSHIMITSU , MATSUMOTO NOBUYUKI , TAKESHIMA HIDENORI
IPC分类号： H04N1/387 , G06T3/40 , H04N1/46 , H04N7/01
摘要： PROBLEM TO BE SOLVED: To provide an image processing apparatus capable of accelerating processing of resolution improvement. SOLUTION: An image processing apparatus includes a temporary high-resolution pixel value calculation unit 101, an importance calculation unit 102, a corresponding position calculation unit 103, and a pixel value conversion section 104. An estimated pixel value of each pixel of a temporary high-resolution image is calculated by interpolation processing based on pixel values of pixels of a reference frame, importance is set to each pixel of the reference frame, and the pixels of one frame of a moving image are sequentially set one by one as an attention pixel. A corresponding position, within the reference frame, corresponding to each attention pixel is calculated with decimal accuracy, each estimated pixel value is corrected to calculate corrected pixel values, respectively, and as the importance corresponding to a partial region including the corresponding position is high, the number of search positions is increased. COPYRIGHT: (C)2009,JPO&INPIT
摘要翻译：要解决的问题：提供能够加快分辨率改善处理的图像处理装置。解决方案：图像处理装置包括临时高分辨率像素值计算单元101，重要度计算单元102，对应位置计算单元103和像素值转换部分104.每个像素的估计像素值通过基于参考帧的像素的像素值的内插处理来计算临时高分辨率图像，将重要性设置为参考帧的每个像素，并且将运动图像的一帧的像素逐个依次设置为注意像素。以十进制精度计算对应于每个注意像素的参考帧内的相应位置，校正每个估计像素值以计算校正像素值，并且由于对应于包括相应位置的部分区域的重要性高，搜索位置的数量增加。版权所有（C）2009，JPO＆INPIT

97. 发明专利

JP2009060464A Video resolution increasing apparatus and method 有权
标题翻译：视频分辨率增加装置和方法
公开(公告)号：JP2009060464A
公开(公告)日：2009-03-19
申请号：JP2007226995
申请日：2007-08-31
申请人： Toshiba Corp , 株式会社東芝
发明人： MATSUMOTO NOBUYUKI , IDA TAKASHI , KANEKO TOSHIMITSU , TAKESHIMA HIDENORI , TAGUCHI YASUNORI
IPC分类号： H04N7/01 , G06T3/00 , G06T3/40 , G09G5/00 , G09G5/36 , G09G5/391 , H04N1/387
摘要： PROBLEM TO BE SOLVED: To provide a video resolution increasing apparatus, and a method capable of performing high-resolution processings, using small throughput. SOLUTION: A video resolution increasing apparatus includes means for setting one frame contained in an image as a reference frame; calculating the offset amount at a pixel position of the reference frame by inputting a field order and a dilation ratio; sequentially setting a plurality of pixels contained in the image one by one as a target pixel; calculating a temporary high-resolution pixel value with the number of pixels more than the number of pixels contained in the reference frame, through interpolation processing based on image data constituted of a pixel value of the target pixel and the offset amount; calculating at least one or more corresponding points, with decimal precision, of the target pixel with respect to the reference frame, by inputting the image data and the offset amount; and calculating the high-resolution pixel value, in such a way as to reduce an error of an estimated pixel value generated using the pixel value of the target pixel, the temporary high-resolution pixel value and the corresponding positions. COPYRIGHT: (C)2009,JPO&INPIT
摘要翻译：要解决的问题：提供一种使用小吞吐量的视频分辨率提高装置和能够执行高分辨率处理的方法。解决方案：视频分辨率提高装置包括用于将包含在图像中的一帧设置为参考帧的装置; 通过输入场顺序和扩张比来计算参考帧的像素位置处的偏移量; 顺序地将包含在图像中的多个像素逐个设置为目标像素; 通过基于由所述目标像素的像素值构成的图像数据和所述偏移量的内插处理来计算具有比所述参考帧中包含的像素数多的像素数的临时高分辨率像素值; 通过输入图像数据和偏移量来计算目标像素相对于参考帧的十进制精度的至少一个或多个对应点; 以及计算高分辨率像素值，以减少使用目标像素的像素值，临时高分辨率像素值和相应位置生成的估计像素值的误差。版权所有（C）2009，JPO＆INPIT

98. 发明专利

JP2008262331A Object tracking device and object tracking method 审中-公开
标题翻译：对象跟踪设备和对象跟踪方法
公开(公告)号：JP2008262331A
公开(公告)日：2008-10-30
申请号：JP2007103712
申请日：2007-04-11
申请人： Toshiba Corp , 株式会社東芝
发明人： MITA TAKESHI , KANEKO TOSHIMITSU , IDA TAKASHI
IPC分类号： G06T7/20
摘要： PROBLEM TO BE SOLVED: To improve performance to make tracking continue without losing sight of an object when detecting and tracking a photographic object (the object) inside a video.
SOLUTION: This object tracking device has: an observation means 2 for obtaining a plurality of different observation series; an evaluation means 3 for evaluating a certainty factor of each of the observation series; and a propagation means 4 for propagating a certainty factor distribution by use of a particle between the observation series. The observation means 2 includes: an offline identification device previously performing learning; and an online identification device performed with learning update during the tracking. The online identification device learns a positive case and a negative case with an image area having a high certainty factor as the positive case, with an image area having a low certainty factor as the negative case of an observation result by the offline identification device, by use of the identification device suitable to identify them.
COPYRIGHT: (C)2009,JPO&INPIT
摘要翻译：要解决的问题：在检测和跟踪视频内的摄影对象（对象）时，提高性能以使跟踪继续，而不会忽视物体。该对象跟踪装置具有：用于获得多个不同观察序列的观察装置2; 用于评估每个观察系列的确定性因子的评价装置3; 以及传播装置4，用于通过使用观察系列之间的粒子来传播确定因子分布。观察装置2包括：先前执行学习的离线识别装置; 以及在跟踪期间用学习更新执行的在线识别装置。在线识别装置以具有高确定性因子的图像区域作为肯定情况学习正面情况和负面情况，具有低确定性因子的图像区域作为离线识别装置的观察结果的否定情况，由使用识别装置适合识别它们。版权所有（C）2009，JPO＆INPIT

99. 发明专利

JP2008228336A Image encoding method and device 审中-公开
标题翻译：图像编码方法和设备
公开(公告)号：JP2008228336A
公开(公告)日：2008-09-25
申请号：JP2008118996
申请日：2008-04-30
申请人： Toshiba Corp , 株式会社東芝
发明人： YAMAGUCHI NOBORU , IDA TAKASHI , KIKUCHI YOSHIHIRO , WATANABE TOSHIAKI
IPC分类号： H04N19/50 , H04N19/13 , H04N19/137 , H04N19/17 , H04N19/176 , H04N19/189 , H04N19/21 , H04N19/46 , H04N19/463 , H04N19/503 , H04N19/517 , H04N19/59 , H04N19/60 , H04N19/61 , H04N19/70 , H04N19/91
摘要： PROBLEM TO BE SOLVED: To provide a device for efficiently encoding the profile information of an object. SOLUTION: The image encoder comprises a means for encoding a square region of an alpha map signal for each block, a means for assigning a label representing whether to include only the pixel of an object region, only the pixel of background region or both pixels in the object region and the background region for each block; a means for determining a prediction value by predicting the label information at the position of a reference block with reference to the label information in an encoded frame; a label encoding means for outputting label encoded information by switching an encoding table with the prediction value and encoding the label information of an encoded block; and a means for outputting the multiplexed alpha map encoded information and the label encoded information, with regard to a pixel block including both the pixels of the object region and the background region. COPYRIGHT: (C)2008,JPO&INPIT
摘要翻译：要解决的问题：提供一种用于有效地编码对象的简档信息的设备。解决方案：图像编码器包括用于对每个块的α映射信号的正方形区域进行编码的装置，用于分配仅表示对象区域的像素的标签，仅包括背景区域的像素或对象区域中的两个像素和每个块的背景区域; 参考编码帧中的标签信息，通过预测参考块的位置处的标签信息来确定预测值的装置; 标号编码装置，用于通过切换具有预测值的编码表并编码编码块的标签信息来输出标签编码信息; 以及关于包括对象区域和背景区域的像素的像素块的输出多路复用的阿尔法地图编码信息和标签编码信息的装置。版权所有（C）2008，JPO＆INPIT

100. 发明专利

JP2008102946A Contour extraction method for image, object extraction method from image and image transmission system using the object extraction method 审中-公开
标题翻译：使用对象提取方法从图像和图像传输系统的图像，对象提取方法的轮廓提取方法
公开(公告)号：JP2008102946A
公开(公告)日：2008-05-01
申请号：JP2007286611
申请日：2007-11-02
申请人： Toshiba Corp , 株式会社東芝
发明人： IDA TAKASHI , SANBONSUGI YOKO , HORI OSAMU
IPC分类号： G06T7/60 , G06T3/00 , G06T7/20 , H04N19/00 , H04N19/134 , H04N19/136 , H04N19/14 , H04N19/192 , H04N19/196 , H04N19/20 , H04N19/21 , H04N19/25 , H04N19/46 , H04N19/59 , H04N19/70
摘要： PROBLEM TO BE SOLVED: To speed up processing of an object area extraction system using self-similar mapping. SOLUTION: This contour extraction method comprises: a first step for setting a search reference block to an contour portion of shape data; a second step for finding a similar block having a similar pattern of image data and having a block size larger than the search reference block from the same image in each the search reference block; and a third step for replacing data obtained by binarizing the image inside each the search reference block with the shape data of the corresponding similar block. In contour extraction processing for bringing an contour of the shape data into line with a contour of an object by repeating the third step prescribed times, the contour extraction processing is repeatedly performed, and the image data, the shape data and the search reference block are reduced in the beginning of the repetition to perform contour position correction processing. COPYRIGHT: (C)2008,JPO&INPIT
摘要翻译：要解决的问题：加快使用自相似映射的对象区域提取系统的处理。解决方案：该轮廓提取方法包括：将搜索参考块设置到形状数据的轮廓部分的第一步骤; 第二步骤，用于从每个搜索参考块中的相同图像中找到具有相似图像数据类型并具有大于搜索参考块的块大小的类似块; 以及第三步骤，用于替换通过使用对应的相似块的形状数据对搜索参考块内的图像进行二值化获得的数据。在轮廓提取处理中，通过重复第三步规定时间使形状数据的轮廓与对象的轮廓一致，轮廓提取处理被重复执行，图像数据，形状数据和搜索参考块是在重复开始时减少进行轮廓位置校正处理。版权所有（C）2008，JPO＆INPIT

你已经成功收藏专利！

检索式保存成功!

IPRDB

热门服务

关于我们

友情链接

联系方式