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

JP2011234307A Image processing apparatus 审中-公开
标题翻译：图像处理设备
公开(公告)号：JP2011234307A
公开(公告)日：2011-11-17
申请号：JP2010105617
申请日：2010-04-30
申请人： Toshiba Corp , 株式会社東芝
发明人： MATSUNO TAKAYA , MORI HIROSHI , MORIMOTO MASAMI , SUZUKI SHINGO
IPC分类号： H04N7/01 , G09G3/20 , G09G5/00
摘要： PROBLEM TO BE SOLVED: To provide an image processing apparatus for performing optimal frame interpolating processing according to an input signal at the time of reproduction display so as not to form an unnatural image.SOLUTION: There is provided an image processing apparatus which generates by alpha-blending processing an interpolation image frame interpolated for performing interpolating processing of input moving image data. The apparatus is configured to include an alpha-blending section 40 which calculates a feature amount between a target image frame and a reference image frame, determines whether or not the feature amount satisfies a predetermined condition, and does not execute alpha-blending processing but duplicates a frame and outputs it as an interpolation image frame when the feature amount does not satisfy the predetermined condition.
摘要翻译：要解决的问题：提供一种用于在再现显示时根据输入信号执行最佳帧内插处理的图像处理装置，以便不形成不自然的图像。解决方案：提供一种图像处理装置，其通过阿尔法混合处理生成插值用于进行输入运动图像数据的内插处理的内插图像帧。该装置被配置为包括计算目标图像帧和参考图像帧之间的特征量的阿尔法混合部分40，确定特征量是否满足预定条件，并且不执行阿尔法混合处理，但是重复当特征量不满足预定条件时，将其作为内插图像帧输出。版权所有（C）2012，JPO＆INPIT

2. 发明专利

JP2011077948A Image processing apparatus 审中-公开
标题翻译：图像处理设备
公开(公告)号：JP2011077948A
公开(公告)日：2011-04-14
申请号：JP2009228909
申请日：2009-09-30
申请人： Toshiba Corp , 株式会社東芝
发明人： SUZUKI SHINGO
IPC分类号： H04N7/01 , H04N19/105 , H04N19/134 , H04N19/137 , H04N19/147 , H04N19/176 , H04N19/196 , H04N19/50 , H04N19/503 , H04N19/51 , H04N19/513 , H04N19/567 , H04N19/577 , H04N19/59
摘要： PROBLEM TO BE SOLVED: To provide an image processing apparatus which can improve quality of a moving image by generating an interpolating frame which has subjectively small degradation.
SOLUTION: When an interpolation frame is generated, a prediction vector candidate setting part 121 and a motion vector detection part 122 predict motion vectors of respective blocks constituting the interpolation frame, on the basis of two temporally-continuous frames. At such a point in time, dispersion of motion vectors is determined and processing is switched in accordance with a dispersion condition. If the dispersion of motion vectors is small, processing is shifted to a motion vector smoothing part 124 to smooth the motion vector on the basis of motion vectors of the two frames or neighboring blocks. On the basis of the smoothed motion vector, an interpolation frame is generated, If the dispersion of motion vectors is large, interpolation processing by the motion vectors is not performed but images spatially at the same positions are diverted as they are.
COPYRIGHT: (C)2011,JPO&INPIT
摘要翻译：要解决的问题：提供一种图像处理装置，其可以通过生成具有主观上小的劣化的内插帧来提高运动图像的质量。解决方案：当生成内插帧时，预测矢量候选设置部分121和运动矢量检测部分122基于两个时间上连续的帧来预测构成内插帧的各个块的运动矢量。在这样一个时间点，确定运动矢量的色散并根据色散条件切换处理。如果运动矢量的色散小，则将处理转移到运动矢量平滑部分124，以基于两帧或相邻块的运动矢量平滑运动矢量。基于平滑的运动矢量，生成内插帧，如果运动矢量的色散大，则不进行运动矢量的插值处理，而在相同位置的空间上的图像原样转移。版权所有（C）2011，JPO＆INPIT

3. 发明专利

JP2011071596A Image processor 审中-公开
标题翻译：图像处理器
公开(公告)号：JP2011071596A
公开(公告)日：2011-04-07
申请号：JP2009218697
申请日：2009-09-24
申请人： Toshiba Corp , 株式会社東芝
发明人： SUZUKI SHINGO , MORI HIROSHI
IPC分类号： H04N19/00 , G06T5/00 , G06T5/20 , H04N1/409 , H04N19/117 , H04N19/14 , H04N19/167 , H04N19/196 , H04N19/44 , H04N19/80 , H04N19/86
摘要： PROBLEM TO BE SOLVED: To provide an image processor which can effectively remove block distortion even if a position where the block distortion occurs is unknown. SOLUTION: An edge detection part 121 detects edge lines with respect to pixel lines in a vertical direction and a horizontal direction on a moving image frame, and a block distortion detection part 123 specifies an edge line caused by block distortion out of the edge lines, and a block distortion removal part 124 applies deblocking processing to the specified edge line. COPYRIGHT: (C)2011,JPO&INPIT
摘要翻译：要解决的问题：提供即使发生块失真的位置是未知的，也可以有效地消除块失真的图像处理器。边缘检测部分121检测关于运动图像帧上的垂直方向和水平方向上的像素线的边缘线，并且块失真检测部分123将由块失真引起的边缘线指定为边缘线和块失真去除部分124对指定的边缘线应用去块处理。版权所有（C）2011，JPO＆INPIT

4. 发明专利

JP2010239268A Image processing apparatus 有权
标题翻译：图像处理设备
公开(公告)号：JP2010239268A
公开(公告)日：2010-10-21
申请号：JP2009082988
申请日：2009-03-30
申请人： Toshiba Corp , 株式会社東芝
发明人： SUZUKI SHINGO
IPC分类号： H04N7/01 , H04N5/14
摘要： PROBLEM TO BE SOLVED: To provide an image processing apparatus, capable of improving quality of a moving image by generating an interpolating frame which has subjectively small degradation, without increasing the processing load.
SOLUTION: When an interpolation frame is generated, a prediction vector candidate setting section 121 and a motion vector detection section 122 predict motion vectors of respective blocks constituting the interpolation frame, based on two temporally-continuous frames; and a local motion vector smoothing section 124 smoothes the predicted motion vectors, based on the two frames and the predicted vector of an adjacent block to obtain a local motion vector. A global motion vector detecting section 123 obtains the inclination of motion vectors (global motion vectors) of the interpolation frame, based on a histogram of the predicted vectors. A motion vector reliance determining section 125 generates an interpolation frame, based on the local motion vector and the global motion vector.
COPYRIGHT: (C)2011,JPO&INPIT
摘要翻译：要解决的问题：提供一种图像处理装置，其能够通过生成具有主观上小的劣化的内插帧来提高运动图像的质量，而不增加处理负荷。解决方案：当生成内插帧时，预测矢量候选设置部分121和运动矢量检测部分122基于两个时间上连续的帧来预测构成内插帧的各个块的运动矢量; 并且局部运动矢量平滑部分124基于两个帧和相邻块的预测向量来平滑预测运动矢量以获得局部运动矢量。全局运动矢量检测部分123基于预测矢量的直方图来获得插值帧的运动矢量（全局运动矢量）的倾斜度。运动矢量依赖判定部125基于局部运动矢量和全局运动矢量生成内插帧。版权所有（C）2011，JPO＆INPIT

5. 发明专利

JP2009076979A Moving image processor 有权
标题翻译：移动图像处理器
公开(公告)号：JP2009076979A
公开(公告)日：2009-04-09
申请号：JP2007241509
申请日：2007-09-18
申请人： Toshiba Corp , 株式会社東芝
发明人： MORI HIROSHI , SUZUKI SHINGO
IPC分类号： H04N19/00 , H04N5/91 , H04N19/102 , H04N19/134 , H04N19/136 , H04N19/157 , H04N19/196 , H04N19/423 , H04N19/44 , H04N19/65 , H04N19/67 , H04N19/70 , H04N19/87
摘要： PROBLEM TO BE SOLVED: To perform proper shot division at all times by discriminating a normal zone surely from an error zone even if a moving image frame cannot be decoded partially owing to the impact of coding error.
SOLUTION: A PTS comparison section 103 determines whether the PTS difference value between an existing frame and a just preceding frame falls within a threshold THe or not. A determination is made that there is no missing frame if the PTS difference value falls within the threshold THe, and a section 104 for comparing the amount of feature determines the similarity of amount of feature of image between frames and then stores the amount of feature of image and the PTS of the existing frame as information representative of the changing point of image in a memory if the similarity is equal to or more than the threshold. A determination is made that there is a missing frame if the PTS difference value exceeds the threshold THe, and a section 106 for generating the amount of feature of error generates a specific amount of feature of image. The specific amount of feature of image is stored as information representative of the changing point of image in a memory along with a value obtained by adding the threshold THe to the PTS of a preceding frame.
COPYRIGHT: (C)2009,JPO&INPIT
摘要翻译：要解决的问题：即使运动图像帧由于编码错误的影响而不能部分解码，因此通过从误差区域确定正常区域来始终进行适当的分割。解决方案：PTS比较部分103确定现有帧和前一帧之间的PTS差值是否落在阈值THe内。如果PTS差值在阈值THe内，则不存在缺失帧，并且用于比较特征量的部分104确定帧之间图像的特征量的相似度，然后存储特征量图像和现有帧的PTS作为表示存储器中图像的变化点的信息，如果相似度等于或大于阈值。如果PTS差值超过阈值THe，则确定存在缺失帧，并且用于生成误差特征量的部分106产生特定量的图像特征。图像的特征量作为代表图像变化点的信息存储在存储器中，以及通过将阈值THe加到前一帧的PTS而获得的值。版权所有（C）2009，JPO＆INPIT

6. 发明专利

JP2007184846A Moving image coding apparatus 审中-公开
标题翻译：移动图像编码设备
公开(公告)号：JP2007184846A
公开(公告)日：2007-07-19
申请号：JP2006002686
申请日：2006-01-10
申请人： Toshiba Corp , 株式会社東芝
发明人： SUZUKI SHINGO , KIKUCHI YOSHIHIRO , KAWASHIMA YUJI
IPC分类号： H04N19/50 , H04N19/107 , H04N19/11 , H04N19/134 , H04N19/136 , H04N19/137 , H04N19/139 , H04N19/147 , H04N19/176 , H04N19/19 , H04N19/196 , H04N19/503 , H04N19/593 , H04N19/61 , H04N19/625 , H04N19/80 , H04N19/86 , H04N19/91
摘要： PROBLEM TO BE SOLVED: To provide a moving image coding apparatus capable of reducing a processing amount while maintaining high prediction accuracy.
SOLUTION: An inter predicting part 2 executes inter prediction processing. Then, an intra prediction processing unit control part 5 decides a pixel unit of intra prediction executed by an intra predicting part 3 on the basis of a parameter indicating a block state selected by the inter prediction processing.
COPYRIGHT: (C)2007,JPO&INPIT
摘要翻译：解决的问题：提供能够在保持高预测精度的同时降低处理量的运动图像编码装置。解决方案：帧间预测部分2执行帧间预测处理。然后，帧内预测处理部控制部5基于表示通过帧间预测处理选择的块状态的参数，决定帧内预测部3执行的帧内预测的像素单位。版权所有（C）2007，JPO＆INPIT

7. 发明专利

JP2014213071A 電子機器および通信制御方法有权
标题翻译：电子装置与通信控制方法
公开(公告)号：JP2014213071A
公开(公告)日：2014-11-17
申请号：JP2013094360
申请日：2013-04-26
申请人：株式会社東芝 , Toshiba Corp
发明人： MATSUNO TAKAYA , SUZUKI SHINGO , KIKUKAWA YUSAKU , YAMAMOTO KOJI , TAKEDA KENTARO , SUDO TAKASHI , AMADA KO , KANISHIMA YASUHIRO , SUGIURA CHIKASHI
IPC分类号： A61B5/00
CPC分类号： A61B5/0002 , A61B5/002 , A61B5/0022 , A61B5/1112 , A61B5/6802 , A61B5/72 , A61B5/742 , A61B2560/0209 , A61B2562/06 , G06F19/3418 , G06F19/345 , G16H50/20
摘要：【課題】生体情報を効率的に伝送できる電子機器および通信制御方法を提供する。【解決手段】ウェラブルデバイス11(電子機器)は、生体センサ110および通信制御プログラム112(抽出手段、状態検出手段、送信制御手段)を具備する。生体センサ110は第1生体データを生成する。抽出手段は、前記第1生体データから第1特徴量を抽出する。状態検出手段は、無線通信モジュール104とサーバとの間の通信状態を検出する。送信制御手段は、前記検出された通信状態に基づいて、前記サーバに、前記第1生体データと前記第1特徴量の少なくとも一方を送信する。【選択図】図2
摘要翻译：可解决的问题：提供能够高效地传输生物信息的电子设备和通信控制方法。解决方案：一种可穿戴装置11（电子设备）包括生物传感器110和通信控制程序112（提取装置，状态检测装置，和传输控制装置）。生物传感器110生成第一生物数据。提取装置从第一生物数据中提取第一特征量。状态检测装置检测无线电通信模块104和服务器之间的通信状态。发送控制装置根据检测到的通信状态向服务器发送第一生物数据和第一特征量中的至少一个。

8. 发明专利

JP2009169348A Video display device 审中-公开
标题翻译：视频显示设备
公开(公告)号：JP2009169348A
公开(公告)日：2009-07-30
申请号：JP2008010349
申请日：2008-01-21
申请人： Toshiba Corp , 株式会社東芝
发明人： SUZUKI SHINGO
IPC分类号： G09G3/34 , G03B21/00 , G09G3/20 , G09G3/36 , H01L33/00 , H01L33/58 , H01L33/60 , H04N5/74 , H05B37/02
CPC分类号： Y02B20/40
摘要： PROBLEM TO BE SOLVED: To provide a video display device which controls element drive current so that a luminous flux is provided with generally high efficiency.
SOLUTION: The video display device is characterized by including: a module constituted of a plurality of LED elements; and a power control means for independently controlling each LED element, wherein the power control means drops supplied power in an order of a low luminous flux to screen among the LED elements of the module in multi-step power saving mode.
COPYRIGHT: (C)2009,JPO&INPIT
摘要翻译：要解决的问题：提供一种控制元件驱动电流以便以高效率提供光通量的视频显示装置。视频显示装置的特征在于包括：由多个LED元件构成的模块; 以及用于独立地控制每个LED元件的功率控制装置，其中，所述功率控制装置以多级节电模式，在所述模块的所述LED元件之间以低光通量的顺序降低所提供的功率。版权所有（C）2009，JPO＆INPIT

9. 发明专利

JP2008017225A Communication terminal device 有权
标题翻译：通信终端设备
公开(公告)号：JP2008017225A
公开(公告)日：2008-01-24
申请号：JP2006187124
申请日：2006-07-06
申请人： Toshiba Corp , 株式会社東芝
发明人： SUZUKI SHINGO , KIKUCHI YOSHIHIRO , KAWASHIMA YUJI
IPC分类号： H04N19/50 , H03M7/36 , H04N19/119 , H04N19/127 , H04N19/136 , H04N19/137 , H04N19/176 , H04N19/186 , H04N19/196 , H04N19/503 , H04N19/593 , H04N19/61 , H04N19/625
CPC分类号： H04N19/136 , H04N19/109 , H04N19/132 , H04N19/172
摘要： PROBLEM TO BE SOLVED: To provide a communication terminal device capable of reducing the load of discrete cosine conversion processing and quantization processing and suppressing deterioration in image quality by controlling the discrete cosine conversion processing and quantization processing according to a selected encoding mode. SOLUTION: When a moving image input through an image input section 11 is encoded, a control unit 121 confirms which of a 4×4 intra-prediction encoding mode and a 16×16 intra-prediction encoding mode for a luminance signal, an intra-prediction encoding mode for a color difference signal, and an inter-prediction encoding mode is applied for encoding processing performed by a processing section 122, and further switches processing contents of discrete cosine conversion processing and quantization processing while taking a prediction error into consideration according to the selected encoding mode. COPYRIGHT: (C)2008,JPO&INPIT
摘要翻译：解决的问题：提供一种能够通过根据选择的编码模式控制离散余弦变换处理和量化处理来减少离散余弦变换处理和量化处理的负荷并抑制图像质量劣化的通信终端装置。解决方案：当通过图像输入部分11输入的运动图像被编码时，控制单元121确认用于亮度信号的4×4帧内预测编码模式和16×16帧内预测编码模式中的哪一个，用于色差信号的帧内预测编码模式和帧间预测编码模式被应用于由处理部分122执行的编码处理，并且进一步切换处理处理内容的离散余弦变换处理和量化处理同时采取预测误差根据选择的编码模式进行考虑。版权所有（C）2008，JPO＆INPIT

10. 发明专利

JP2007019761A Moving image coding apparatus 有权
标题翻译：移动图像编码设备
公开(公告)号：JP2007019761A
公开(公告)日：2007-01-25
申请号：JP2005198070
申请日：2005-07-06
申请人： Toshiba Corp , 株式会社東芝
发明人： SUZUKI SHINGO , KIKUCHI YOSHIHIRO , KAWASHIMA YUJI
IPC分类号： H04N19/50 , H03M7/36 , H04N19/105 , H04N19/127 , H04N19/139 , H04N19/176 , H04N19/503 , H04N19/51 , H04N19/513 , H04N19/523 , H04N19/533 , H04N19/57 , H04N19/61 , H04N19/625 , H04N19/82 , H04N19/91
CPC分类号： H04N19/176 , H04N19/523 , H04N19/573
摘要： PROBLEM TO BE SOLVED: To reduce a processing amount while maintaining high accuracy of motion compensation prediction. SOLUTION: When performing motion compensation prediction in a motion vector detection part 2 for input image data VD, three reference frames RVD in the past stored in a reference image storage part 6 are referred to first, an integer search is performed to the image data VD for each macroblock and thus a motion prediction vector is detected. Then, on the basis of the detected results of the motion prediction vector, a reference frame optimum for performing the motion compensation prediction is decided for each macroblock. Then, on the basis of the decided results, whether or not the reference frames considered as optimum by all blocks (9 blocks) match is determined. Then, when all candidate blocks select the same reference frame as the results of determination, only the reference frame considered as optimum is turned to the object of a non-integer search, and the motion prediction vector is detected by non-integer pixel accuracy. COPYRIGHT: (C)2007,JPO&INPIT
摘要翻译：要解决的问题：为了减少处理量，同时保持运动补偿预测的高精度。首先，参照在输入图像数据VD的运动矢量检测部2中进行运动补偿预测时，首先参照过去存储在基准图像存储部6中的三个参考帧RVD，对检测每个宏块的图像数据VD，从而检测运动预测矢量。然后，基于运动预测矢量的检测结果，对每个宏块确定最佳的用于执行运动补偿预测的参考帧。然后，根据所确定的结果，确定所有块（9个块）被认为是最佳的参考帧是否匹配。然后，当所有候选块选择与确定结果相同的参考帧时，仅将被认为是最优的参考帧转换为非整数搜索的对象，并且以非整数像素精度检测运动预测向量。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