专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

1. 发明专利

JP2013066034A Image discrimination device and image discrimination method 有权
标题翻译：图像识别装置和图像识别方法
公开(公告)号：JP2013066034A
公开(公告)日：2013-04-11
申请号：JP2011203093
申请日：2011-09-16
申请人： Mitsubishi Electric Corp , 三菱電機株式会社
发明人： KUDO DAIKI , KONISHI YOSHIKI , NISHIKAWA HIROBUMI
IPC分类号： H04N1/41 , H04N19/00 , H04N19/44 , H04N19/59 , H04N19/65 , H04N19/89
摘要： PROBLEM TO BE SOLVED: To automatically discriminate an image having errors.SOLUTION: An image discrimination device has: an image decoding part 1 decoding an encoded image to obtain a decoded image; an image processing part 2 performing on the decoded image obtained by the image decoding part 1 image processing for shifting a boundary between block-like image disturbance generated in the decoded image in the case that there are errors in the encoded image and a block used in encoding to obtain a processed image; an image encoding part 3 encoding the decoded image obtained by the image decoding part 1 and the processed image obtained by the image processing part 2 to obtain a re-encoded image and an encoded processed image; a code amount comparator 4 comparing a code amount of the re-encoded image obtained by the image encoding part 3 with that of the encoded processed image; and an image discriminating part 5 determining whether or not there are errors in the encoded image on the basis of the comparison results done by the code amount comparator 4.
摘要翻译：要解决的问题：自动识别具有错误的图像。图像识别装置具有：图像解码部分1对编码图像进行解码以获得解码图像; 图像处理部分2对通过图像解码部分1获得的解码图像执行图像处理，用于在编码图像中存在错误的情况下移位在解码图像中产生的块状图像干扰之间的边界，编码以获得经处理的图像; 编码由图像解码部分1获得的解码图像的图像编码部分3和由图像处理部分2获得的经处理的图像，以获得重新编码的图像和编码的处理图像; 比较由图像编码部分3获得的再编码图像的编码量与编码处理图像的编码量的代码量比较器4; 以及图像鉴别部分5，基于由代码量比较器4进行的比较结果，确定编码图像中是否存在错误。（C）2013，JPO和INPIT

2. 发明专利

JP2012129608A Moving image encoder, moving image decoder, moving image encoding method, and moving image decoding method 审中-公开
标题翻译：移动图像编码器，移动图像解码器，移动图像编码方法和移动图像解码方法
公开(公告)号：JP2012129608A
公开(公告)日：2012-07-05
申请号：JP2010276946
申请日：2010-12-13
申请人： Mitsubishi Electric Corp , 三菱電機株式会社
发明人： KUDO DAIKI , NISHIKAWA HIROBUMI
IPC分类号： H04N19/50 , H04N5/92 , H04N19/46 , H04N19/503 , H04N19/51 , H04N19/59 , H04N19/593 , H04N19/70
摘要： PROBLEM TO BE SOLVED: To provide a moving image encoder, a moving image decoder, a moving image encoding method, and a moving image decoding method, capable of enhancing random reproduction performance in a moving image decoder while retaining a high compression rate of an image.SOLUTION: The moving image encoder is so configured that timings of encoding process, in which intra encoding is used between encoding units 2 to 5, are different and an encoding process by intra encoding is conducted on any one of reduced images Pto Pgenerated at any time, thus enables enhancing of a random reproduction performance of a moving image decoder while retaining a high compression rate of an image.
摘要翻译：要解决的问题：提供一种运动图像编码器，运动图像解码器，运动图像编码方法和运动图像解码方法，其能够在保持高压缩率的同时增强运动图像解码器中的随机再现性能的图像。解决方案：运动图像编码器被配置为使得在编码单元2至5之间使用帧内编码的编码处理的定时不同，并且通过帧内编码的编码处理对缩小图像P < SB POS =“POST”> A 到任何时间产生的P D ，从而能够增强运动图像解码器的随机再现性能，同时保持高压缩图像的速度版权所有（C）2012，JPO＆INPIT

3. 发明专利

JP2012120243A Moving image prediction device, and moving image predicting method 有权
标题翻译：移动图像预测装置和移动图像预测方法
公开(公告)号：JP2012120243A
公开(公告)日：2012-06-21
申请号：JP2012028852
申请日：2012-02-13
申请人： Mitsubishi Electric Corp , 三菱電機株式会社
发明人： SEKIGUCHI SHUNICHI , ASAI KOTARO , MURAKAMI ATSUMICHI , NISHIKAWA HIROBUMI , KURODA SHINICHI , ISU YOSHIMI , HASEGAWA YURI
IPC分类号： H03M7/36 , H04N7/24 , H04N19/102 , H04N19/103 , H04N19/105 , H04N19/107 , H04N19/126 , H04N19/13 , H04N19/134 , H04N19/137 , H04N19/139 , H04N19/14 , H04N19/142 , H04N19/146 , H04N19/17 , H04N19/172 , H04N19/176 , H04N19/179 , H04N19/189 , H04N19/196 , H04N19/20 , H04N19/21 , H04N19/23 , H04N19/423 , H04N19/46 , H04N19/50 , H04N19/503 , H04N19/51 , H04N19/513 , H04N19/527 , H04N19/57 , H04N19/573 , H04N19/577 , H04N19/61 , H04N19/625 , H04N19/70 , H04N19/87 , H04N19/91
CPC分类号： H04N7/24 , H04N19/105 , H04N19/127 , H04N19/137 , H04N19/142 , H04N19/179 , H04N19/186 , H04N19/196 , H04N19/23 , H04N19/423 , H04N19/44 , H04N19/46 , H04N19/463 , H04N19/503 , H04N19/537 , H04N19/573 , H04N19/58 , H04N19/61 , H04N19/87
摘要： PROBLEM TO BE SOLVED: To achieve an encoding system capable of predicting a moving picture with high accuracy in accordance with contents of a scene, a degree of importance, and motion characteristics and the like, of the moving picture.SOLUTION: An encoding system comprises: a plurality of memories a, b and c storing image data to be referred; a motion compensator 5 capable of responding to an arbitrary transform parameter representing motion of an image region to be predicted and generating a predicted picture by using arbitrary data stored in the plurality of the memories a, b and c based on the transform parameter; and a memory update unit 15 for enabling update of memory contents of one or more of the plurality of the memories a, b and c at arbitrary time intervals.
摘要翻译：要解决的问题：为了实现能够根据场景的内容，运动图像的重要性和运动特性等高精度地预测运动图像的编码系统。解码：编码系统包括：存储要参考的图像数据的多个存储器a，b和c; 运动补偿器5，其能够响应于表示要预测的图像区域的运动的任意变换参数，并且基于变换参数，通过使用存储在多个存储器a，b和c中的任意数据生成预测图像; 以及存储器更新单元15，用于以任意时间间隔更新多个存储器a，b和c中的一个或多个的存储器内容。版权所有（C）2012，JPO＆INPIT

4. 发明专利

JP2012085348A Image decoder and image decoding method 有权
标题翻译：图像解码器和图像解码方法
公开(公告)号：JP2012085348A
公开(公告)日：2012-04-26
申请号：JP2011281596
申请日：2011-12-22
申请人： Mitsubishi Electric Corp , 三菱電機株式会社
发明人： KURODA SHINICHI , SEKIGUCHI SHUNICHI , ASAI KOTARO , NISHIKAWA HIROBUMI , MORIYA YOSHIMI , HASEGAWA YURI
IPC分类号： H04N19/20 , H04N7/12 , H04N7/62 , H04N19/00 , H04N19/21 , H04N19/25 , H04N19/46 , H04N19/463 , H04N19/503 , H04N19/70
CPC分类号： H04N19/00533 , H04N7/52 , H04N19/132 , H04N19/20 , H04N19/46 , H04N19/587 , H04N19/61 , H04N19/70 , H04N21/23412 , H04N21/234318 , H04N21/23614 , H04N21/4307 , H04N21/4348 , H04N21/44012 , H04N21/6547 , H04N21/8547
摘要： PROBLEM TO BE SOLVED: To provide an image decoder and an image decoding method which ensures more smooth and accurate restoration processing of an image by simple processing with a small amount of encoding transmission, and can achieve a system capable of obtaining a reproduced image by preventing generation of unnecessary volume of information and synthesizing a plurality of objects with a simple configuration.SOLUTION: The image decoder comprises an image frame decoding unit which decodes an image frame, a time code analysis unit which decodes a time code, and a header analysis unit which decodes the frame rate information defining the absolute time of an image frame that is decoded by the image frame decoding unit in combination with the time code decoded by the time code analysis unit.
摘要翻译：解决的问题：提供一种图像解码器和图像解码方法，其通过使用少量编码传输的简单处理来确保图像的更平滑和精确的恢复处理，并且可以实现能够获得再现的系统通过防止产生不必要的信息量并以简单的配置合成多个对象来形成图像。解决方案：图像解码器包括对图像帧进行解码的图像帧解码单元，对时间码进行解码的时间码分析单元，以及对定义图像帧的绝对时间的帧速率信息进行解码的头部分析单元其由图像帧解码单元与由时间码分析单元解码的时间码组合解码。版权所有（C）2012，JPO＆INPIT

5. 发明专利

JP2011160196A Video data decoding reproducing device 审中-公开
标题翻译：视频数据解码重现设备
公开(公告)号：JP2011160196A
公开(公告)日：2011-08-18
申请号：JP2010020258
申请日：2010-02-01
申请人： Mitsubishi Electric Corp , 三菱電機株式会社
发明人： NISHIKAWA HIROBUMI
IPC分类号： H04N19/50 , H04N19/12 , H04N19/136 , H04N19/139 , H04N19/146 , H04N19/196 , H04N19/423 , H04N19/426 , H04N19/44 , H04N19/503 , H04N19/513 , H04N19/61 , H04N19/625 , H04N19/91
摘要： PROBLEM TO BE SOLVED: To obtain a video data decoding reproducing device in which an image buffer required in reverse reproduction is minimized and a processing load in the reverse reproduction is made very small.
SOLUTION: Among pieces of compressed image data, an encoding parameter except texture encoded information is stored in an encoding parameter buffer 113. When a decoding operation mode 116 shows a simple reverse reproduction operation, the encoding parameter buffer 113 generates motion information required for performing motion compensation processing in the reverse direction. A motion compensator 109 performs motion compensation prediction of an image in the reverse direction from decoded images accumulated in an image buffer 111 using motion information from the encoding parameter buffer 113.
COPYRIGHT: (C)2011,JPO&INPIT
摘要翻译：要解决的问题：为了获得其中反向再现中所需的图像缓冲器最小化并且反向再现中的处理负载非常小的视频数据解码再现装置。解决方案：在压缩图像数据中，除了纹理编码信息之外的编码参数被存储在编码参数缓冲器113中。当解码操作模式116示出简单的反向再现操作时，编码参数缓冲器113产生所需的运动信息用于在相反方向执行运动补偿处理。运动补偿器109使用来自编码参数缓冲器113的运动信息，对累积在图像缓冲器111中的解码图像进行相反方向的图像的运动补偿预测。（C）2011，JPO＆INPIT

6. 发明专利

JP2011045122A Moving image coding apparatus and moving image decoding apparatus 有权
标题翻译：移动图像编码装置和移动图像解码装置
公开(公告)号：JP2011045122A
公开(公告)日：2011-03-03
申请号：JP2010231550
申请日：2010-10-14
申请人： Mitsubishi Electric Corp , 三菱電機株式会社
发明人： ISU YOSHIMI , SEKIGUCHI SHUNICHI , ASAI KOTARO , NISHIKAWA HIROBUMI , KURODA SHINICHI , HASEGAWA YURI
IPC分类号： H04N19/60 , H03M7/40 , H04N19/102 , H04N19/136 , H04N19/176 , H04N19/625 , H04N19/91 , H04N19/93
摘要： PROBLEM TO BE SOLVED: To provide a moving image decoding apparatus for rightly decoding a coded bit stream generated according to a coding method for coding coefficient data obtained by scanning a quantization conversion coefficient using a variable length coding table.
SOLUTION: In the moving image decoding apparatus, a VLC table group selector 107b6 selects one variable length coding table group predetermined by prescribed frequency domain divided beforehand from a group of variable length coding table groups in accordance with whether a high frequency component of a quantization conversion coefficient is included. A VLC table selector 107c10 selects from the selected variable length coding table group a variable length coding table corresponding to the variable length coding table selected in accordance with a non-zero coefficient continued to zero when the quantization conversion coefficient is scanned during coding, and performs variable length decoding on coefficient data obtained by scanning the quantization conversion coefficient using the selected variable length decoding table.
COPYRIGHT: (C)2011,JPO&INPIT
摘要翻译：解决的问题：提供一种运动图像解码装置，用于对根据通过使用可变长度编码表扫描量化转换系数而获得的编码系数数据的编码方法而生成的编码比特流进行正确解码。解决方案：在运动图像解码装置中，VLC表组组选择器107b6根据从一组可变长度编码表组预先划分的预定频域预定的一个可变长度编码表组，包括量化转换系数。 VLC表选择器107c10在编码期间当量化转换系数被扫描时，从选择的可变长度编码表组中选择与根据连续为零的非零系数选择的可变长度编码表相对应的可变长度编码表，并执行通过使用所选择的可变长度解码表扫描量化变换系数而获得的系数数据的可变长度解码。版权所有（C）2011，JPO＆INPIT

7. 发明专利

JP2009171622A Method of describing hint information 审中-公开
公开(公告)号：JP2009171622A
公开(公告)日：2009-07-30
申请号：JP2009111989
申请日：2009-05-01
申请人： Mitsubishi Electric Corp , 三菱電機株式会社
发明人： MORIYA YOSHIMI , NISHIKAWA HIROBUMI , SEKIGUCHI SHUNICHI , YAMADA ETSUHISA , OTOI KENJI , KURODA SHINICHI , OGAWA FUMINOBU , ASAI KOTARO
IPC分类号： H04N5/91 , G06F17/30 , G11B27/00 , H04N7/173 , H04N7/24
CPC分类号： G06F17/30908 , G06F17/30802 , G06F17/30811 , G06F17/30817 , G06F17/30858 , G06K9/00711 , G11B27/28 , G11B27/34 , H04N21/234336 , H04N21/234363 , H04N21/235 , H04N21/435 , H04N21/4668 , H04N21/8405 , H04N21/8456 , H04N21/8543
摘要： PROBLEM TO BE SOLVED: To obtain a method of describing hint information for reducing processing cost (throughput and amount of memory required) for analysis and regeneration of metadata.
SOLUTION: Multimedia content containing moving images and sound is divided into a plurality of scenes to generate metadata for each divided scene. The metadata, which include scene structure information metadata describing the hierarchical structure of the content, are generated in addition to scene interval information and titles. Also, the name or identifier of a descriptor included in the metadata is described as hint information when operating metadata comprising one or a plurality of descriptors for describing semantic content, construction, and characteristics of the content.
COPYRIGHT: (C)2009,JPO&INPIT

8. 发明专利

JP2009071877A Moving image prediction device and moving image predicting method 有权
标题翻译：移动图像预测装置和移动图像预测方法
公开(公告)号：JP2009071877A
公开(公告)日：2009-04-02
申请号：JP2008332876
申请日：2008-12-26
申请人： Mitsubishi Electric Corp , 三菱電機株式会社
发明人： SEKIGUCHI SHUNICHI , ASAI KOTARO , MURAKAMI ATSUMICHI , NISHIKAWA HIROBUMI , KURODA SHINICHI , ISU YOSHIMI , HASEGAWA YURI
IPC分类号： H03M7/36 , H04N7/24 , H04N19/102 , H04N19/103 , H04N19/105 , H04N19/107 , H04N19/126 , H04N19/13 , H04N19/134 , H04N19/137 , H04N19/139 , H04N19/14 , H04N19/142 , H04N19/146 , H04N19/17 , H04N19/172 , H04N19/176 , H04N19/179 , H04N19/189 , H04N19/196 , H04N19/20 , H04N19/21 , H04N19/23 , H04N19/423 , H04N19/46 , H04N19/50 , H04N19/503 , H04N19/51 , H04N19/513 , H04N19/527 , H04N19/57 , H04N19/573 , H04N19/577 , H04N19/61 , H04N19/625 , H04N19/70 , H04N19/87 , H04N19/91
CPC分类号： H04N7/24 , H04N19/105 , H04N19/127 , H04N19/137 , H04N19/142 , H04N19/179 , H04N19/186 , H04N19/196 , H04N19/23 , H04N19/423 , H04N19/44 , H04N19/46 , H04N19/463 , H04N19/503 , H04N19/537 , H04N19/573 , H04N19/58 , H04N19/61 , H04N19/87
摘要： PROBLEM TO BE SOLVED: To achieve an encoding system for predicting a moving picture with high accuracy in accordance with the content of a scene, a degree of importance, and motion characteristics, and the like, of a moving picture. SOLUTION: The encoding system is provided with memories (a), (b) and (c) storing image data to be referred; a motion compensator 5 compatible with an arbitrary transform parameter representing the motion of an image region to be predicted and generating a predicted picture by using arbitrary data stored in the memories (a), (b) and (c) based upon the transform parameter; and a memory update unit 15 for updating the content of one or more memories at arbitrary time intervals. COPYRIGHT: (C)2009,JPO&INPIT
摘要翻译：要解决的问题：根据场景的内容，运动图像的重要性和运动特性等来实现用于高精度地预测运动图像的编码系统。解码方案：编码系统具有存储要参考的图像数据的存储器（a），（b）和（c）基于变换参数，通过使用存储在存储器（a），（b）和（c）中的任意数据来生成预测图像，与运动补偿器5兼容，该任意变换参数表示要预测的图像区域的运动; 以及用于以任意时间间隔更新一个或多个存储器的内容的存储器更新单元15。版权所有（C）2009，JPO＆INPIT

9. 发明专利

JP2008269628A Image retrieval system and image retrieval method 有权
标题翻译：图像检索系统和图像检索方法
公开(公告)号：JP2008269628A
公开(公告)日：2008-11-06
申请号：JP2008130870
申请日：2008-05-19
申请人： Mitsubishi Electric Corp , 三菱電機株式会社
发明人： SEKIGUCHI SHUNICHI , ISU YOSHIMI , NISHIKAWA HIROBUMI , YAMADA ETSUHISA , ASAI KOTARO
IPC分类号： G06F17/30 , G06K9/00 , G06T1/00 , G06T7/00 , H04N5/76 , H04N5/91
CPC分类号： G06F17/30247 , G06F17/30256 , G06F17/30802 , G06F17/30808 , G06F17/30811 , G06F17/30817 , G06F17/30858 , G06K9/00711
摘要： PROBLEM TO BE SOLVED: To provide image retrieval processing that can be carried out efficiently. SOLUTION: A feature descriptor is generated in a segment unit in a feature descriptor set generation part 4, based on an image feature amount in image data extracted by a feature extraction part 3, and the image data and the feature descriptor are accumulated as a set in an image information accumulation part 5. A retrieval processing part 10 retrieves an attribute list accumulated in an attribute list accumulation part 7, and outputs and displays attribute information matched to a retrieval condition, on a display part 11, when the retrieval condition of the attribute list is input from a user interface part 9. Then, the retrieval processing part 10 retrieves the image data accumulated in the image information accumulation part 5, when a retrieval condition of similar retrieval is input from the user interface part 9, selects the image data having the feature descriptor matched to the retrieval condition, and outputs it to the display part 11 to be displayed. COPYRIGHT: (C)2009,JPO&INPIT
摘要翻译：要解决的问题：提供可以有效执行的图像检索处理。解决方案：基于由特征提取部分3提取的图像数据中的图像特征量，在特征描述符集生成部分4的段单元中生成特征描述符，并且累积图像数据和特征描述符作为图像信息累积部分5中的一组。检索处理部分10检索在属性列表累积部分7中累积的属性列表，并且在检索时输出并显示与检索条件匹配的属性信息，在显示部分11上从用户界面部分9输入属性列表的条件。然后，当从用户界面部分9输入类似检索的检索条件时，检索处理部分10检索累积在图像信息累积部分5中的图像数据，选择具有与检索条件相匹配的特征描述符的图像数据，并将其输出到显示部分11以进行显示。版权所有（C）2009，JPO＆INPIT

10. 发明专利

JP2007195234A Image decoder 有权
标题翻译：图像解码器
公开(公告)号：JP2007195234A
公开(公告)日：2007-08-02
申请号：JP2007060791
申请日：2007-03-09
申请人： Mitsubishi Electric Corp , 三菱電機株式会社
发明人： MORIYA YOSHIMI , SEKIGUCHI SHUNICHI , ASAI KOTARO , NISHIKAWA HIROBUMI , KURODA SHINICHI , HASEGAWA YURI
IPC分类号： H04N19/00 , H04N19/12 , H04N19/13 , H04N19/159 , H04N19/172 , H04N19/174 , H04N19/186 , H04N19/20 , H04N19/40 , H04N19/44 , H04N19/46 , H04N19/48 , H04N19/503 , H04N19/51 , H04N19/593 , H04N19/61 , H04N19/625 , H04N19/70 , H04N19/91 , H04N19/93
摘要： PROBLEM TO BE SOLVED: To decode an H. 263 coded bit stream through a simple configuration.
SOLUTION: When a first coded bit stream is received, image coded information used for decoding the first coded bit stream according to the syntax of second coded bit stream is determined, based on image coded information included in the header information of H. 263 coding system, and the first coded bit stream is decoded according to the syntax of the second coded bit stream, based on that image coded information. When a second coded bit stream is received, the second coded bit stream is decoded, based on the image coded information decoded by the decoding means.
COPYRIGHT: (C)2007,JPO&INPIT
摘要翻译：要解决的问题：通过简单的配置解码H. 263编码比特流。解决方案：当接收到第一编码比特流时，基于包含在头部信息中的图像编码信息确定根据第二编码比特流的语法来解码第一编码比特流的图像编码信息。 263编码系统，并且基于该图像编码信息，根据第二编码比特流的语法对第一编码比特流进行解码。当接收到第二编码比特流时，基于由解码装置解码的图像编码信息对第二编码比特流进行解码。版权所有（C）2007，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