公开(公告)号：US20110235897A1

公开(公告)日：2011-09-29

申请号：US13052659

申请日：2011-03-21

申请人： Masaharu Watanabe ,  Fumiaki Tomita ,  Kouta Fujimura ,  Yasuyuki Ikeda ,  Kengo Ichimura

发明人： Masaharu Watanabe ,  Fumiaki Tomita ,  Kouta Fujimura ,  Yasuyuki Ikeda ,  Kengo Ichimura

IPC分类号： G06K9/00

CPC分类号： G06K9/00214 ,  G06T7/75 ,  G06T2207/10016

摘要： The device includes: an input unit (4) for receiving image data obtained by capturing images of an object by imaging units (C1 to C3); and an arithmetic unit (1) that performs: 1) finding a three-dimensional reconstruction point set that contain three-dimensional position information of segments obtained by dividing a boundary of the object in the image data, and a feature set that contain three-dimensional information regarding vertices of the segments, for each of multiple pairs of two different images; 2) calculating a total three-dimensional reconstruction point set and a total feature set by totaling the three-dimensional reconstruction point sets and the feature sets of the multiple pairs; and 3) matching a model feature set regarding model data of the object with the total feature set, thereby determining, among the total three-dimensional reconstruction point set, points corresponding to model points of the object.

摘要翻译： 该装置包括：输入单元（4），用于接收通过成像单元（C1至C3）捕获对象的图像而获得的图像数据; 以及运算单元（1），其执行以下操作：1）找到包含通过划分所述图像数据中的对象的边界而获得的段的三维位置信息的三维重构点集，以及包含三维重建点的特征集， 对于两对不同图像的多对中的每一个，关于段的顶点的三维信息; 2）通过总计三维重建点集合和多对特征集来计算总三维重建点集和总特征集; 以及3）将关于对象的模型数据的模型特征集合与总特征集进行匹配，从而在总的三维重建点集合中确定与对象的模型点相对应的点。

公开(公告)号：US20110235898A1

公开(公告)日：2011-09-29

申请号：US13053819

申请日：2011-03-22

申请人： Masaharu Watanabe ,  Fumiaki Tomita

发明人： Masaharu Watanabe ,  Fumiaki Tomita

IPC分类号： G06K9/32

CPC分类号： G06T7/33 ,  G06T2207/10012

摘要： The matching process includes: finding first and second three-dimensional reconstruction point sets that contain three-dimensional position coordinates of segments, and first and second feature set that contain three-dimensional information regarding vertices of the segments, from image data of an object (S2 to S4); and matching the first feature set with the second feature set, thereby determining a corresponding point of the first and second three-dimensional reconstruction point sets (S5 to S10), wherein Step S5 carries out initial matching with respect to the segments; Step S6 selects a feature point from the first feature set; Step S7 specifies an adjustment region containing the selected feature point; Step S8 carries out fine adjustment of matching with respect to the segments within the adjustment region; and Steps S6 to S8 are performed at each increase of the adjustment region and at each selection of the feature point.

摘要翻译： 匹配过程包括：从对象的图像数据（第一和第二特征集合）中找到包含段的三维位置坐标的第一和第二三维重构点集合以及包含关于段顶点的三维信息的第一和第二特征集 S2至S4）; 并且将所述第一特征集合与所述第二特征集匹配，从而确定所述第一和第二三维重构点集合的对应点（S5至S10），其中步骤S5针对所述片段执行初始匹配; 步骤S6从第一特征集中选择特征点; 步骤S7指定包含所选特征点的调整区域; 步骤S8对调整区域内的段进行匹配的微调整; 并且在调整区域的每次增加和特征点的每次选择时执行步骤S6至S8。

公开(公告)号：US5845006A

公开(公告)日：1998-12-01

申请号：US683708

申请日：1996-07-18

申请人： Yasushi Sumi ,  Fumiaki Tomita ,  Yutaka Ishiyama

发明人： Yasushi Sumi ,  Fumiaki Tomita ,  Yutaka Ishiyama

IPC分类号： G06T1/00 ,  G06T7/00 ,  G06T7/20 ,  G06T7/60 ,  H04N13/00 ,  G06K9/00 ,  G06K9/34

CPC分类号： G06T7/0042 ,  G06T7/0075 ,  G06T7/2033 ,  G06T2207/10012 ,  H04N13/0055 ,  H04N13/0059 ,  H04N13/0239 ,  H04N13/0296 ,  H04N2013/0081

摘要： A method of processing image information is disclosed, for detecting the movement of an object with a simplifed system structure, and high speed treatment. The method is comprised of the steps of entry of a stereoscopic image of an object, extraction of edges of the object, and dividing it into segments based on the local feature of the image. On each segment, tracing points (sampling points) are selected, the inter-frame correlation of selected points is obtained, the three-dimensional coordinates of the points are measured, and thus the position and attitude of the object are detected, wherein the inter-frame correlation is performed by evaluating the similarity between frames based on correlation between the tangential direction of tracing points and the luminance value of the surface.

摘要翻译： 公开了一种处理图像信息的方法，用于以简化的系统结构和高速处理来检测物体的移动。 该方法包括以下步骤：输入物体的立体图像，提取物体的边缘，并根据图像的局部特征将其分割成段。 在每个段上，选择跟踪点（采样点），获得所选点的帧间相关性，测量点的三维坐标，从而检测对象的位置和姿态，其中， 通过基于跟踪点的切线方向和表面的亮度值之间的相关性来评估帧之间的相似度来执行帧相关。

公开(公告)号：US5202928A

公开(公告)日：1993-04-13

申请号：US601657

申请日：1989-07-21

申请人： Fumiaki Tomita ,  Kazuhide Sugimoto ,  Hironobu Takahashi

发明人： Fumiaki Tomita ,  Kazuhide Sugimoto ,  Hironobu Takahashi

IPC分类号： G06T1/00 ,  G06K9/68 ,  G06T7/00 ,  H04N13/00

CPC分类号： G06T7/0075 ,  G06K9/00201 ,  G06K9/469 ,  G06K2209/40 ,  G06T2207/10012 ,  H04N2013/0081 ,  H04N2013/0092

摘要： An image processing method for matching edges detected in images of the same objects in a three-dimensional scene which are taken simultaneously by two or more image pick-up devices at different positions, in which the boundary segments on the edges are classified into apparently real segments, apparently imaginary segments or undefined segments, a planar surface is found for each boundary which has all apparently real segments and no apparently imaginary segment on it by using the information of the points of occlusion, the surface is regarded as the real surface in the case where there is only one surface which has no segment under it, so that surface information and the distance information at arbitrary points on the real surface are obtained.

摘要翻译： 一种图像处理方法，用于将在两个或更多个图像拾取装置同时拍摄的三维场景中的相同对象的图像中检测到的边缘进行匹配，其中边缘上的边界段被分类为明显的真实 段，明显虚构的段或未定义的段，通过使用闭塞点的信息，为每个边界找到具有所有表面上真实段并且在其上没有明显虚拟段的平面，该表面被认为是 只有一个表面在其下面没有段的情况，从而获得表面信息和实际表面任意点处的距离信息。

公开(公告)号：US07064776B2

公开(公告)日：2006-06-20

申请号：US10078882

申请日：2002-02-19

申请人： Yasushi Sumi ,  Fumiaki Tomita ,  Yutaka Ishiyama ,  Fumio Kubo

发明人： Yasushi Sumi ,  Fumiaki Tomita ,  Yutaka Ishiyama ,  Fumio Kubo

IPC分类号： H04N5/232

CPC分类号： G01S3/7865

摘要： An object movement tracking apparatus and its method tracks an object with an unknown initial position and which object is continually moving, the apparatus for recovering a tracking process even when the tracking has failed and includes an observing device such as a video camera for observing a scene at a video frame rate. Time-sequentially acquired image data from the observing device are stored in an image memory device and the data processed in parallel. During data processing, an object recognition module selects a frame memory based on predetermined information. An estimated value of position/posture of the object are calculated from the image data and transmitted. A movement tracking module receives transmitted data from the object recognition module, replacing a current frame with a next frame and waits until image data in the next frame are inputted. When image data inputting is finished, the up dated replacement in position/posture of the object and its evaluated value are calculated by comparing image data and the object model. The system mode is changed and updated based on the obtained evaluated value.

摘要翻译： 物体运动跟踪装置及其方法跟踪具有未知初始位置和哪个物体不断移动的物体，即使跟踪失败也用于恢复跟踪处理的装置，并且包括用于观察场景的诸如摄像机的观察装置 以视频帧速率。 来自观察装置的时间顺序获取的图像数据被存储在图像存储装置中，并且并行处理数据。 在数据处理期间，对象识别模块基于预定信息来选择帧存储器。 根据图像数据计算物体的位置/姿势的估计值并发送。 移动跟踪模块从对象识别模块接收发送的数据，用下一帧代替当前帧，并等待直到下一帧中的图像数据被输入。 当图像数据输入完成时，通过比较图像数据和对象模型来计算对象的位置/姿势的更新日期替换及其评估值。 基于获得的评估值改变和更新系统模式。

公开(公告)号：US6028954A

公开(公告)日：2000-02-22

申请号：US196866

申请日：1994-02-14

申请人： Fumiaki Tomita ,  Hironobu Takahashi

发明人： Fumiaki Tomita ,  Hironobu Takahashi

IPC分类号： G06T7/00 ,  H04N13/00 ,  G06K9/32

CPC分类号： G06T7/002 ,  G06T7/0022 ,  G06T7/004 ,  H04N13/0239 ,  G06T2207/10012 ,  H04N13/0246 ,  H04N2013/0081

摘要： An image processing method and an apparatus therefor calculating the rotations and the magnifications of a plurality of image pick-up devices, which are located at different positions and rotations and rotates, and pick-up images of the same object in a three-dimensional space, from the following equation by matching the image picked-up by the two arbitrary image pick-up devices, extracting the corresponding points, and detecting the coordinates values thereof on each image, ##EQU1## where, {expression}.sub.y : y component of the expression,{expression}.sub.z : z component of the expression,x,y,z: coordinate values of a point in a three-dimensional space,X.sub.L ', Y.sub.L ', X.sub.R ', Y.sub.R ': coordinate values of x, y on both images,.theta..sub.L, .phi..sub.L, .alpha..sub.L, .phi..sub.R, .alpha..sub.R,: differences in the rotation from the x-, y-, and z-axes of both image pick-up devices from the state to become a reference,f.sub.L,f.sub.R : magnifications of both image pick-up devices.

摘要翻译： 一种用于计算位于不同位置和旋转并旋转的多个摄像装置的旋转和放大率的图像处理方法及其装置，并且在三维空间中拾取同一物体的图像 通过匹配由两个任意图像拾取装置拾取的图像，提取相应的点，并且检测每个图像上的坐标值，从下面的等式中，表达式的{表达式} y：y分量， {表达式} z：z表达式的分量x，y，z：三维空间中的点的坐标值，XL'，YL'，XR'，YR'：两个图像上的x，y的坐标值 ，θL，phi L，αL，phi R，αR，：从两个图像拾取装置的x轴，y轴和z轴到从成为参考的状态的旋转差，fL， fR：两个图像拾取装置的放大率。

公开(公告)号：US06950550B1

公开(公告)日：2005-09-27

申请号：US09669384

申请日：2000-09-26

申请人： Yasushi Sumi ,  Fumiaki Tomita ,  Yutaka Ishiyama

发明人： Yasushi Sumi ,  Fumiaki Tomita ,  Yutaka Ishiyama

IPC分类号： G01B11/00 ,  G01C3/06 ,  G06K9/00 ,  G06K9/32 ,  G06T1/00 ,  G06T7/00 ,  G06T7/20 ,  G06T7/60 ,  H04N5/225 ,  H04N13/00

CPC分类号： G06K9/32 ,  G06T7/285 ,  G06T7/75 ,  H04N13/243

摘要： After stereo image is inputted (S1), the tracing point corresponding to the silhouette of a curved object observed by the stereo image is selected on the basis of the information on the present position of the object (S2), using a three-dimensional geometric model. And the corresponding points on the silhouette corresponding to respective tracing points are sampled from the stereo image (S3), and their three-dimensional coordinates are measured (S4). Then, from the group of the three-dimensional coordinates of tracing points and corresponding points, the position/posture of the object concerned and the measurement error are determined (S5), to discriminate the error (S6). In case the error is not small enough, processes starting from (S2) are repeated, regarding the detected position/posture as a renewed present position, while, if the error is small enough, processes are repeated starting from (S1), regarding the detected position/posture as a renewed present position, thereby providing the technique for high-speed tracing of three-dimensional rigid body motion of an arbitrarily-curved and smooth faced object utilizing an object model, applying a stereo camera system as sensors.

摘要翻译： 在输入立体声图像之后（S1），根据关于对象的当前位置的信息（S 2）选择与立体图像观察到的对象的轮廓对应的跟踪点， 立体几何模型。 并且从立体图像（S 3）中对与各个跟踪点对应的轮廓上的对应点进行采样，并测量它们的三维坐标（S 4）。 然后，从跟踪点和对应点的三维坐标组中，确定对象的位置/姿势和测量误差（S 5），以区分误差（S 6）。 在误差不够小的情况下，对于作为更新的当前位置的检测到的位置/姿势，重复从（S 2）开始的处理，而如果误差足够小，则从（S 1）开始重复处理， 关于作为更新的当前位置的检测到的位置/姿势，从而提供使用对象模型来高速跟踪任意弯曲和平滑面对象的三维刚体运动的技术，应用立体摄像机系统作为传感器。

公开(公告)号：US5887083A

公开(公告)日：1999-03-23

申请号：US683710

申请日：1996-07-18

申请人： Yasushi Sumi ,  Fumiaki Tomita ,  Yutaka Ishiyama

发明人： Yasushi Sumi ,  Fumiaki Tomita ,  Yutaka Ishiyama

IPC分类号： H04N7/18 ,  G06K9/64 ,  G06T1/00 ,  G06T7/00 ,  G06T7/60 ,  H04N13/00 ,  G06K9/00 ,  G06K9/48 ,  G06T17/10

CPC分类号： G06T7/0046 ,  G06K9/6204 ,  G06T7/0075 ,  G06T2207/10012 ,  H04N13/0055 ,  H04N13/0059 ,  H04N13/0239 ,  H04N2013/0081 ,  H04N2013/0092

摘要： A method of processing image information is disclosed, for recognizing the position and attitude of an object with a high speed treatment and improved recognition accuracy. The method comprised of the steps of entry of a stereoscopic image of an object, extraction of edges of the object, and dividing it into segments based on the local feature of the image, adding the apex information to the segments to produce the recognition data, verifying the recognition data with reference to an object model based on local geometric features to detect a corresponding candidate; finely adjusting each of corresponding candidates based on an entire geometric features; and detecting the position and attitude of the object on the basis of recognition including the initial verification and the fine adjustment.

摘要翻译： 公开了一种处理图像信息的方法，用于以高速处理识别对象的位置和姿态并提高识别精度。 该方法包括步骤：输入对象的立体图像，提取对象的边缘，并根据图像的局部特征将其划分成多个段，将顶点信息添加到段以产生识别数据， 参照基于局部几何特征的对象模型来验证识别数据以检测相应的候选者; 基于整个几何特征精细调整每个相应的候选者; 并且基于包括初始验证和微调的识别来检测对象的位置和姿态。