会员体验
专利管家(专利管理)
工作空间(专利管理)
风险监控(情报监控)
数据分析(专利分析)
侵权分析(诉讼无效)
联系我们
交流群
官方交流:
QQ群: 891211   
微信请扫码    >>>
现在联系顾问~
热词
    • 1. 发明专利
    • Motion detecting device, motion detecting method and motion detecting program
    • 运动检测装置,运动检测方法和运动检测程序
    • JP2012198877A
    • 2012-10-18
    • JP2012024015
    • 2012-02-07
    • Morpho Inc株式会社モルフォ
    • SAWAI SHOGO
    • G06T7/20H04N19/00H04N19/105H04N19/137H04N19/176H04N19/186H04N19/189H04N19/51
    • PROBLEM TO BE SOLVED: To provide a motion detecting device capable of more accurately evaluating reliability of matching with a fewer operation amount.SOLUTION: A motion detecting device comprises a difference calculation part 111, a reliability calculation part 112 and a motion determination part 113. The difference calculation part 111 calculates a difference in a pixel value between an attention block of a reference image and a set of regions, which is as large as the attention block, in a comparison image. The reliability calculation part 112 sets a block with the smallest difference in the comparison image as a target block, and calculates reliability of matching in the attention block through the use of a first difference showing a difference between the target block and the attention block and a second difference showing a difference between surrounding blocks, which are blocks existing around the target block, and the attention block. The motion determination part 113 determines a motion of the attention block based on the reliability.
    • 要解决的问题:提供一种能够以更少的操作量更准确地评估匹配的可靠性的运动检测装置。 解决方案:运动检测装置包括差分计算部分111,可靠性计算部分112和运动确定部分113.差分计算部分111计算参考图像的注意块和像素值之间的像素值的差异 在比较图像中,与注意块一样大的区域集合。 可靠性计算部112将比较图像中具有最小差异的块设置为目标块,并且通过使用示出目标块和关注块之间的差的第一差异来计算注意块中的匹配的可靠性,以及 示出作为目标块周围的块的周围块与注意块之间的差的第二差。 动作确定部113基于可靠性确定注意力块的运动。 版权所有(C)2013,JPO&INPIT
    • 3. 发明专利
    • Image processor, image processing method, image processing program and storage medium
    • 图像处理器,图像处理方法,图像处理程序和存储介质
    • JP2013017155A
    • 2013-01-24
    • JP2012041715
    • 2012-02-28
    • Morpho Inc株式会社モルフォ
    • KAMOSHITA YUKIONYUKAI JUNHABUKA KANESUKEHIRAGA TADAMOTO
    • H04N5/232G06T7/20
    • H04N5/2329
    • PROBLEM TO BE SOLVED: To provide an image processor for stably and precisely estimating a rolling shutter strain component and to provide an image processing method and an image processing program.SOLUTION: The image processor 1 includes: an image input section 10; a motion vector acquisition section 11; and a strain component estimating section 12. The image input section 10 inputs a plurality of frame images. The motion vector acquisition section 11 acquires a parallel translation amount between the two frame images. The strain component estimating section 12 approximates that a motion matrix showing motion between the frames consists only of a parallel translation component and the rolling shutter strain component, calculates a temporary distortion coefficient included in the rolling shutter distortion component at every frame image by using the parallel translation amount and estimates a strain coefficient by using the plurality of calculated temporary strain coefficients.
    • 要解决的问题:提供一种用于稳定和精确地估计滚动快门应变分量的图像处理器,并提供图像处理方法和图像处理程序。 解决方案:图像处理器1包括:图像输入部分10; 运动矢量获取部11; 和应变分量估计部分12.图像输入部分10输入多个帧图像。 运动矢量获取部11获取两帧图像之间的平行平移量。 应变分量估计部12近似于表示帧之间的运动的运动矩阵仅由平行平移分量和滚动快门应变分量组成,通过使用并行计算在每帧图像处计算包括在滚动快门失真分量中的暂时失真系数 平移量,并通过使用多个计算的临时应变系数来估计应变系数。 版权所有(C)2013,JPO&INPIT
    • 4. 发明专利
    • Image processor, image processing method, and image processing program
    • 图像处理器,图像处理方法和图像处理程序
    • JP2010258648A
    • 2010-11-11
    • JP2009104896
    • 2009-04-23
    • Morpho IncNtt Docomo Inc株式会社エヌ・ティ・ティ・ドコモ株式会社モルフォ
    • FUJIBAYASHI AKIRABUN CHUNSENHIRAGA TADAMOTOSHIINO HIROKI
    • H04N5/20G06T5/10G09G5/36
    • G06T5/002G06T2207/20012H04N7/0132
    • PROBLEM TO BE SOLVED: To reduce the amount of operation for distortion reduction processing while reducing distortion contained in a video. SOLUTION: An image processor 10 is equipped with: a process pixel group determination means 102 which determines a process object pixel group based on a predetermined rule in a plurality of frame images of an inputted video signal, and determines a processed pixel group based on a predetermined standard in the processed frame image having been stored; a frequency component processing means 103 which extracts at least a time-frequency component out of the process object pixel group and the processed pixel group, and executes a predetermined coefficient process with a high time-frequency component in the time-frequency component, for acquiring a process object pixel group after execution; a process completed frame generating means 104 which generates a process completed frame image from the process object pixel group after execution; and an image output means 106 which outputs the process completed frame image and stores it in a processed frame storing means 105. The frequency component processing means 103 avoids execution of a predetermined coefficient process on the component of low sensitivity in visual characteristic in the high time-frequency component. COPYRIGHT: (C)2011,JPO&INPIT
    • 要解决的问题:减少在失真减少处理中的操作量,同时减少包含在视频中的失真。 解决方案:图像处理器10配备有:基于输入视频信号的多个帧图像中的预定规则来确定处理对象像素组的处理像素组确定装置102,并且确定处理对象像素组 基于已经存储的已处理帧图像中的预定标准; 频率分量处理装置103,其从处理对象像素组和处理像素组中提取至少一个时间分量,并且在时间 - 频率分量中执行具有高时间分量的预定系数处理,以获取 执行后的进程对象像素组; 处理完成帧生成装置104,其在执行之后从处理对象像素组生成处理完成帧图像; 以及图像输出装置106,其输出处理完成的帧图像并将其存储在处理的帧存储装置105中。频率分量处理装置103避免在高时间的视觉特性中对低灵敏度的分量执行预定系数处理 频率分量。 版权所有(C)2011,JPO&INPIT
    • 5. 发明专利
    • Image processor
    • 图像处理器
    • JP2007226643A
    • 2007-09-06
    • JP2006048581
    • 2006-02-24
    • Morpho Inc株式会社モルフォ
    • HIRAGA TADAMOTOHABUKA KANESUKE
    • G06T7/20G06T1/00
    • G06T5/003G06T7/207G06T2207/10016
    • PROBLEM TO BE SOLVED: To calculate high-accuracy movement data by comparatively simple processing such that a load is not imposed on a computer. SOLUTION: Multiple resolution images of a reference image and a comparison image are created. To the images of low resolution, whole range matching is first applied to detect two-dimensional displacement between the images. To the images of comparatively high resolution, displacement in each block is detected by block matching. By correcting the movement data by the higher-resolution image by use of the movement data found in the preceding stage as an initial value sequentially from minimum resolution to high resolution, accuracy of the image is improved. COPYRIGHT: (C)2007,JPO&INPIT
    • 要解决的问题:通过相对简单的处理来计算高精度移动数据,使得不对计算机施加负载。

      解决方案:创建参考图像和比较图像的多分辨率图像。 对于低分辨率的图像,首先应用全范围匹配来检测图像之间的二维位移。 对于较高分辨率的图像,通过块匹配来检测每个块中的位移。 通过使用从前一级中发现的移动数据通过较高分辨率图像将运动数据从最小分辨率依次提高为高分辨率,从而提高了图像的精度。 版权所有(C)2007,JPO&INPIT

    • 6. 发明专利
    • 画像処理装置及び画像処理方法
    • 图像处理装置和图像处理方法
    • JP2015033072A
    • 2015-02-16
    • JP2013163110
    • 2013-08-06
    • 株式会社モルフォMorpho Inc
    • KOBAYASHI MICHIHIROMIURA TAKESHIHIRAGA TADAMOTO
    • H04N5/225G03B15/00G03B19/07H04N5/232H04N5/335H04N5/367
    • 【課題】歩留まりの低下を防止しながら、レンズ及びイメージセンサからなるカメラを複数配列することにより高解像度の画像を取得する。【解決手段】画像処理装置1は、レンズ及びイメージセンサからなるカメラモジュールを複数含むカメラアレイ2と、カメラアレイ2の各カメラモジュールから出力される複数の画像を合成して合成画像を出力する出力部3とを備える。カメラアレイ2は、複数のカメラモジュールからなるモジュールブロック21を複数含む。一のモジュールブロックに含まれる複数のカメラモジュールのレンズは略同方向の光軸を有するので、それらのカメラモジュール22から出力される複数の画像に基づき、いわゆる超解像の画像を生成できる。各モジュールブロック21の光軸方向が互いに異なるので、モジュールブロック21ごとの超解像画像をスティッチング処理により合成して広範囲な画像を出力できる。【選択図】図1
    • 要解决的问题:通过排列由透镜和图像传感器形成的多个相机来获取高分辨率图像,同时防止产量降低。解决方案:图像处理装置1包括:相机阵列2,包括多个 每个由透镜和图像传感器形成的相机模块; 以及输出部分3,用于通过组合从相机阵列2的相机模块输出的多个图像来输出合成图像。相机阵列2包括由多个相机模块形成的多个模块21。 由于包括在一个模块中的多个相机模块的透镜具有基本上在相同方向上的光轴,所以可以基于从这些相机模块22输出的多个图像来产生所谓的超分辨率图像。由于 模块21的光轴方向彼此不同,通过缝合处理来合成各模块21的超分辨率图像,能够输出宽幅图像。
    • 7. 发明专利
    • Image display device, image display method and program
    • 图像显示装置,图像显示方法和程序
    • JP2014160461A
    • 2014-09-04
    • JP2014011384
    • 2014-01-24
    • Morpho Inc株式会社モルフォ
    • HIRAGA TADAMOTONAKANO TOMOYUKI
    • G06F3/048G06F3/041G06F3/0488G09G3/20G09G5/00G09G5/36G09G5/38
    • G06F3/0416G06F3/0412G06F3/04817G06F3/0482G06F3/04842G06F3/04845G06F3/0488G06F3/04883G06F2203/04104G06F2203/04806G06F2203/04808
    • PROBLEM TO BE SOLVED: To simplify the setting of an image display mode or the selection of an image in a touch screen type image display device.SOLUTION: All thumb nails belonging to a designated range by a pinch operation are specified, and when all the thumb nails belonging to the designated range are not displayed in a selective mode (S2004: NO), all the thumb nails belonging to the designated range are displayed in the selective mode (S2005), and a status is put in a selective state (S2006), and when all the thumb nails belonging to the designated range are displayed in the selective mode (S2004: YES), all the thumb nails belonging to the designated range are displayed in non-selective mode (S2007), and the status is put in a non-selective state (S2008). When a transition to a pinch-out operation or a pinch-in operation is continuously made (S2010: NO), a selective region or non-selective region is enlarged/reduced in accordance with the status (S2014, S2015).
    • 要解决的问题:为了简化图像显示模式的设置或触摸屏型图像显示装置中的图像的选择。解释:指定了通过捏合操作属于指定范围的所有拇指钉,并且当所有 属于指定范围的拇指钉不以选择模式显示(S2004:否),属于指定范围的所有拇指钉都以选择模式显示(S2005),状态为选择状态(S2006 ),并且当以选择模式显示属于指定范围的所有拇指钉(S2004:是)时,属于指定范围的所有拇指钉以非选择模式显示(S2007),并且状态被放置 处于非选择状态(S2008)。 当连续进行到缩小操作或夹持操作的转变(S2010:否)时,根据状态放大/缩小选择区域或非选择区域(S2014,S2015)。
    • 8. 发明专利
    • Image synthesizing apparatus, image synthesizing method, image synthesizing program, and recording medium
    • 图像合成装置,图像合成方法,图像合成程序和记录介质
    • JP2013243780A
    • 2013-12-05
    • JP2013175399
    • 2013-08-27
    • Morpho Inc株式会社モルフォ
    • HIRAI YOSHITOMIURA TAKESHI
    • H04N5/235G06T3/00H04N5/225
    • PROBLEM TO BE SOLVED: To provide an image synthesizing apparatus, an image synthesizing method, and an image synthesizing program which are capable of generating an appropriate synthetic image even if a subject moves, and to provide a recording medium with the image synthesizing program stored thereon.SOLUTION: The apparatus generates a synthetic image using a first image and a second image different in exposure condition. The apparatus comprises an image input unit 10, a likelihood calculation unit 13, an exposure estimation unit 14, and a synthesizing unit 16. The image input unit 10 receives the first image and the second image. Before the estimation of an exposure transformation function for matching the exposure conditions of the first image and the second image, the likelihood calculation unit 13 calculates the likelihood of a moving subject on the basis of the difference between the first image and the second image. The exposure estimation unit 14 estimates the exposure transformation function on the basis of the likelihood of a moving subject. The synthesis unit 16 synthesizes the first image and the second image using the exposure transformation function.
    • 要解决的问题:提供一种图像合成装置,图像合成方法和图像合成程序,即使对象移动也能够产生合适的合成图像,并且提供其上存储有图像合成程序的记录介质 解决方案:该装置使用不同于曝光条件的第一图像和第二图像来生成合成图像。 该装置包括图像输入单元10,似然率计算单元13,曝光估计单元14和合成单元16.图像输入单元10接收第一图像和第二图像。 在估计用于匹配第一图像和第二图像的曝光条件的曝光变换功能之前,似然度计算单元13基于第一图像和第二图像之间的差来计算移动被摄体的可能性。 曝光估计单元14基于移动的被摄体的可能性来估计曝光变换功能。 合成单元16使用曝光变换功能合成第一图像和第二图像。
    • 9. 发明专利
    • Feature point detection device, feature point detection method, feature point detection program and recording medium
    • 特征点检测装置,特征点检测方法,特征点检测程序和记录介质
    • JP2013218530A
    • 2013-10-24
    • JP2012088571
    • 2012-04-09
    • Morpho Inc株式会社モルフォ
    • INOMATA TEPPEI
    • G06T7/00G06T1/00
    • PROBLEM TO BE SOLVED: To provide a feature point detection device, a feature point detection method, a feature point detection program and a recording medium which can detect a feature point included in an object adequately and fast.SOLUTION: A feature point detection device 1 comprises a candidate point selection unit 12. The candidate point selection unit 12 acquires a feature amount of parts of an object as a standard, compares it with a feature amount of a target image 30 and outputs positional information of the feature point of face parts of the target image 30. The feature amount of the parts of the object as the standard is a set of codes encoding big or small relations of a pixel value of each pixel pair in ternary in a plurality of pixel pairs selected out of pixels of face parts drawn in a teacher image. The candidate point selection unit 12 selects a plurality of pixel pairs from the pixel of the target image, encodes big or small relations of the pixel value of each pixel pair in ternary and uses the set of the obtained codes as the feature amount of the target image.
    • 要解决的问题:提供一种特征点检测装置,特征点检测方法,特征点检测程序和能够充分且快速地检测对象中的特征点的记录介质。解决方案:特征点检测装置1 包括候选点选择单元12.候选点选择单元12获取作为标准的对象的部分的特征量,将其与目标图像30的特征量进行比较,并且输出目标图像30的特征点的位置信息 目标图像30.作为标准的对象的部分的特征量是编码从面部的像素中选择的多个像素对中的三元中的每个像素对的像素值的大或小关系的代码的集合 画在教师形象中。 候选点选择单元12从目标图像的像素中选择多个像素对,对三元组中的每个像素对的像素值的大小关系进行编码,并使用所获得的代码的集合作为目标的特征量 图片。
    • 10. 发明专利
    • Image processing device, image processing method, image processing program, and recording medium
    • 图像处理装置,图像处理方法,图像处理程序和记录介质
    • JP2013081091A
    • 2013-05-02
    • JP2011220269
    • 2011-10-04
    • Morpho Inc株式会社モルフォ
    • HABUKA KANESUKEABE TAKESHI
    • H04N13/02H04N5/225H04N5/232
    • G06T5/003G06T5/50G06T7/248G06T2207/10012G06T2207/10021G06T2207/20021G06T2207/20201H04N5/23254H04N5/23274
    • PROBLEM TO BE SOLVED: To perform compatible correction processing between images acquired from a plurality of imaging elements.SOLUTION: An image processing device comprises: an installation information acquisition unit 40 which acquires installation information indicating an installation state with respect to an imaging unit 10 of an imaging element per imaging element; a motion information acquisition unit 30 which acquires motion information of an imaging device between a first frame image and a second frame image; and an output region acquisition unit 50 which acquires positional information of a second output region corresponding to a first output region in the first frame image which is a region in the second frame image per imaging element. The output region acquisition unit 50 acquires a second output region of a reference imaging element among a plurality of imaging elements on the basis of the positional information and motion information of the first output region determined according to the installation information. This acquisition unit also acquires a second output region of another imaging element different from the reference imaging element among the plurality of imaging elements on the basis of the installation information of the reference imaging element, the installation information of the another imaging element, and the motion information.
    • 要解决的问题:执行从多个成像元件获取的图像之间的兼容校正处理。 解决方案:图像处理装置包括:安装信息获取单元40,其获取关于每个成像元件的成像元件的成像单元10指示安装状态的安装信息; 运动信息获取单元30,其获取第一帧图像和第二帧图像之间的成像装置的运动信息; 以及输出区域获取单元50,其获取与作为每个成像元件的第二帧图像中的区域的第一帧图像中的与第一输出区域对应的第二输出区域的位置信息。 输出区域获取单元50基于根据安装信息确定的第一输出区域的位置信息和运动信息,获取多个成像元件中的参考成像元件的第二输出区域。 该获取单元还基于参考成像元件的安装信息,另一成像元件的安装信息和运动来获取多个成像元件中与参考成像元件不同的另一成像元件的第二输出区域 信息。 版权所有(C)2013,JPO&INPIT