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

JP2010231398A Device, method and program for determination of three-dimensional object 有权
标题翻译：用于确定三维对象的设备，方法和程序
公开(公告)号：JP2010231398A
公开(公告)日：2010-10-14
申请号：JP2009076918
申请日：2009-03-26
申请人： Toshiba Corp , 株式会社東芝
发明人： KOSAKAYA TATSUO , KUBOTA SUSUMU
IPC分类号： G06T1/00 , G06T7/00
CPC分类号： G06K9/00255 , G06K9/00899 , G06T7/75 , G06T2207/30201
摘要： PROBLEM TO BE SOLVED: To precisely determine whether an object included in an acquired image is a three-dimensional object. SOLUTION: A three-dimensional object determining device includes: a detection unit 2 which detects a plurality of feature points of an object included in acquired image data; a pattern normalizing unit 5 which generates a normalization pattern provided by normalizing the feature points by a three-dimensional model; an estimation unit 6 which estimates the illuminating direction of light applied on the object in the image data from the three-dimensional model and the normalization pattern; and a determination unit 7 which determines whether the object in the image data is a three-dimensional object by using the illuminating direction. COPYRIGHT: (C)2011,JPO&INPIT
摘要翻译：要解决的问题：精确地确定所获取的图像中包括的对象是否是三维对象。解决方案：三维物体确定装置包括：检测单元2，其检测所获取的图像数据中包括的对象的多个特征点; 图案归一化单元5，其生成通过三维模型归一化特征点而提供的归一化图案; 估计单元6，估计来自三维模型和标准化模式的图像数据中施加在物体上的光的照明方向; 以及确定单元7，其通过使用照明方向来确定图像数据中的对象是否是三维对象。版权所有（C）2011，JPO＆INPIT

2. 发明专利

JP2010182013A Linear transformation matrix calculation device, and method thereof and program thereof 有权
标题翻译：线性变换矩阵计算装置及其方法及其程序
公开(公告)号：JP2010182013A
公开(公告)日：2010-08-19
申请号：JP2009023805
申请日：2009-02-04
申请人： Toshiba Corp , 株式会社東芝
发明人： KUBOTA SUSUMU , KAWAHARA TOMOKAZU
IPC分类号： G06T7/00
CPC分类号： G06K9/624
摘要： PROBLEM TO BE SOLVED: To provide a linear transformation matrix calculation device for obtaining a linear transformation matrix wherein similarity between partial spaces belonging to different categories is increased and wherein similarity between partial spaces belonging to the same category is reduced. SOLUTION: The linear transformation matrix calculation device linearly transforms a plurality of dictionary partial spaces each belonging to each category by a linear transformation matrix, selects a plurality of sets of combinations each between two dictionary partial spaces from the plurality of linearly transformed dictionary partial spaces, calculates a loss function by use of each similarity between the dictionary partial spaces of each selected set, calculates a differential parameter when differentiating the loss function with respect to the linear transformation matrix, calculates a new linear transformation matrix by the differential parameter and the linear transformation matrix by a steepest descent method, and updates the new linear transformation matrix as the linear transformation matrix used in a linear transformation part. COPYRIGHT: (C)2010,JPO&INPIT
摘要翻译：要解决的问题：提供一种用于获得线性变换矩阵的线性变换矩阵计算装置，其中属于不同类别的部分空间之间的相似性增加，并且其中属于同一类别的部分空间之间的相似性减小。解决方案：线性变换矩阵计算装置通过线性变换矩阵对属于每个类别的多个词典部分空间进行线性变换，从多个线性变换词典中选择两组词典部分空间之间的多组合通过使用每个选定集合的字典部分空间之间的每个相似度来计算损失函数，在相对于线性变换矩阵区分损失函数时计算差分参数，通过差分参数计算新的线性变换矩阵，以及通过最速下降法的线性变换矩阵，并且将新的线性变换矩阵更新为在线性变换部分中使用的线性变换矩阵。版权所有（C）2010，JPO＆INPIT

3. 发明专利

JP2008176504A Object detector and method therefor 审中-公开
标题翻译：对象检测器及其方法
公开(公告)号：JP2008176504A
公开(公告)日：2008-07-31
申请号：JP2007008473
申请日：2007-01-17
申请人： Toshiba Corp , 株式会社東芝
发明人： MARUYAMA MASAYUKI , KUBOTA SUSUMU , NAKAI HIROAKI
IPC分类号： G06T7/20 , H04N7/18
摘要： PROBLEM TO BE SOLVED: To provide an object detector capable of stably detecting and tracking an object at relatively low cost.
SOLUTION: The object detector 10 comprises a camera part 12, a change area detection part 14, an intersecting area detection part 16, an object tracking part 18, an object determination part 20, and a background update part 22. The detector regards an object moved from an initial position as a target by performing background subtraction method and visual hull method and mutually exchanging output results of matching by pattern learning and matching using an object image at a previous time as a template.
COPYRIGHT: (C)2008,JPO&INPIT
摘要翻译：要解决的问题：提供能够以相对低的成本稳定地检测和跟踪物体的物体检测器。解决方案：物体检测器10包括照相机部分12，变化区域检测部分14，交叉区域检测部分16，物体跟踪部分18，物体确定部分20和背景更新部分22。将通过执行背景减法和视觉船体方法从初始位置移动的对象视为通过模式学习和匹配的输出结果，以前一时间的对象图像作为模板相互交换输出结果。版权所有（C）2008，JPO＆INPIT

4. 发明专利

JP2007249309A Obstacle tracking system and method 审中-公开
标题翻译： OBSTACLE跟踪系统和方法
公开(公告)号：JP2007249309A
公开(公告)日：2007-09-27
申请号：JP2006068402
申请日：2006-03-13
申请人： Toshiba Corp , 株式会社東芝
发明人： NAKANO TAKESHI , KUBOTA SUSUMU
IPC分类号： G06T7/00 , G01B11/00 , G06T1/00 , G08G1/16 , H04N7/18
CPC分类号： G06T7/20 , G06K9/00805 , G06K9/32
摘要： PROBLEM TO BE SOLVED: To provide an obstacle tracking system that detects the accurate position of an obstacle in consideration of the ambiguity of detected obstacle candidate areas. SOLUTION: The obstacle tracking system comprises an image input part 1 mounted on a mobile body to capture time series images of an obstacle, an obstacle detection part 2 for detecting current obstacle candidate areas from the time series images, a hypothesis generation part 3 for holding a state hypothesis group about the obstacle at the preceding time and generating an observational hypothesis group combining observational hypotheses in the positions of the obstacle candidate areas and an observational hypothesis based on no obstacle detection, a likelihood calculation part 5 for calculating likelihoods of combinations from the state hypothesis group and observational hypothesis group, selecting the combinations that have the highest likelihood from the observational hypothesis group and state hypothesis group and updating the selected state hypothesis group as a current state hypothesis group, and a hypothesis selection part 6 for selecting the state hypothesis that has the highest likelihood in the current state hypothesis group as a detected state of the obstacle. COPYRIGHT: (C)2007,JPO&INPIT
摘要翻译：要解决的问题：提供一种障碍物跟踪系统，其考虑到检测到的障碍物候补区域的模糊性来检测障碍物的准确位置。解决方案：障碍物跟踪系统包括安装在移动体上的图像输入部分1，以捕获障碍物的时间序列图像，用于从时间序列图像中检测当前障碍物候补区域的障碍物检测部分2，假设生成部分 3，用于在前一时刻持有关于障碍物的状态假设组，并且生成将观察假设组合在障碍物候补区域的位置和基于无障碍物检测的观测假设的观测假设组，似然计算部分5，用于计算来自状态假设组和观察假设组的组合，从观察假设组和状态假设组中选择具有最高可能性的组合，并将所选择的状态假设组更新为当前状态假设组，以及用于选择的假设选择部分6 国家假设有高在当前状态假设组中作为障碍物的检测状态的可能性。版权所有（C）2007，JPO＆INPIT

5. 发明专利

JP2004191354A Calibration method, calibration support device, calibration device and production method of camera system 有权
公开(公告)号：JP2004191354A
公开(公告)日：2004-07-08
申请号：JP2003324615
申请日：2003-09-17
申请人： Toshiba Corp , 株式会社東芝
发明人： MAEDA KENICHI , TANIGUCHI TAKAHIRO , KUBOTA SUSUMU , NAKAI HIROAKI
IPC分类号： G01B11/00 , G01B11/245 , G06T1/00 , G06T7/00 , H04N13/00
CPC分类号： H04N13/0246 , G06T7/80 , G06T7/85
摘要： PROBLEM TO BE SOLVED: To ensure highly accurate calibration even in a relatively small space.
SOLUTION: A mirror 101 and a mirror 102 are arranged in parallel with each other so that the reflective faces of the mirrors are opposed to each other, and a mark 104 is disposed therebetween. A camera 105 photographs the mark 104 and an inverted image thereof from the vicinity of the upper end central portion of the mirror 102 for calibration. Namely, the inverted image of the mark reflected by the mirrors 101 and 102 simulates multiple marks as if scattered over a broad range up to the distance. Such simulated marks are photographed so that the same calibration can be conducted as in the case where the state where the multiple marks are actually arranged is photographed.
COPYRIGHT: (C)2004,JPO&NCIPI

6. 发明专利

JP2012123787A Touch panel device, control device and control method 审中-公开
标题翻译：触控面板装置，控制装置和控制方法
公开(公告)号：JP2012123787A
公开(公告)日：2012-06-28
申请号：JP2011241056
申请日：2011-11-02
申请人： Toshiba Corp , 株式会社東芝
发明人： ITO SATOSHI , KUBOTA SUSUMU
IPC分类号： G06F3/045 , G06F3/041
CPC分类号： G06F3/045 , G06F3/0416
摘要： PROBLEM TO BE SOLVED: To obtain at least a positional relation between two points touched on a touch panel at the time of the touch.SOLUTION: According to one embodiment of the present invention, a control device 110 controls a touch panel 100 in which a first resistance film and a second resistance film are laminated with a space therebetween so that the resistance films come into contact at a touched position. The control device 110 includes an application unit 101, a measurement unit 102, and calculation units 103, 104, 105 and 106. The application unit 101 applies voltage between a first terminal and a second terminal of the first resistance film at a first timing. The measurement unit 102 measures voltages of a third terminal and a fourth terminal of the second resistance film at the first timing. The calculation units 103, 104, 105 and 106 obtains a positional relation between two points touched on the touch panel 100 at the time of the touch on the basis of the voltages of the third terminal and the fourth terminal measured at the first timing.
摘要翻译：要解决的问题：为了在触摸时获得在触摸面板上触摸的两个点之间的至少一个位置关系。解决方案：根据本发明的一个实施例，控制装置110控制触摸面板100，其中第一电阻膜和第二电阻膜之间具有间隔层压，使得电阻膜在触动的位置。控制装置110包括应用单元101，测量单元102和计算单元103,104,105和106.应用单元101在第一定时处在第一电阻膜的第一端子和第二端子之间施加电压。测量单元102在第一定时测量第二电阻膜的第三端子和第四端子的电压。计算单元103,104,105和106基于在第一定时测量的第三终端和第四终端的电压来获得在触摸时触摸的触摸面板100上的两个点之间的位置关系。版权所有（C）2012，JPO＆INPIT

7. 发明专利

JP2010039788A Image processing apparatus and method thereof, and image processing program 审中-公开
标题翻译：图像处理装置及其方法和图像处理程序
公开(公告)号：JP2010039788A
公开(公告)日：2010-02-18
申请号：JP2008202291
申请日：2008-08-05
申请人： Toshiba Corp , 株式会社東芝
发明人： ITO SATOSHI , KUBOTA SUSUMU , IKE TSUKASA , KOSAKAYA TATSUO , TAKEGUCHI SATOYUKI
IPC分类号： G06T7/20 , G06T7/00 , H04N7/18
CPC分类号： G06K9/6231
摘要： PROBLEM TO BE SOLVED: To provide an image processing apparatus which is robust and achieves a high-speed tracking, and to provide a method thereof and an image processing program.
SOLUTION: The image processing apparatus 100 is provided with: a discrimination part 152 which uses N-pieces of preliminarily generated feature extraction parts 151 to extract N-pieces of feature quantities from an input image and calculates a reliability showing target object likeliness from the N extracted feature quantities; an object detection part 120 which detects the target object included in the input image on the basis of the reliability; a feature selection part 140 which selects M-pieces of feature extraction parts from the N feature extraction parts 151 so that the degree of separation between the reliability of the target object and a reliability of its background is higher than that obtained by using the N-pieces of feature extraction parts 151; and an object tracking part 140 which uses the M-pieces of feature extraction parts selected by the feature selection part 130 to extract M-pieces of feature quantities from the input image and uses the M-pieces of extracted feature quantities to track the target object.
COPYRIGHT: (C)2010,JPO&INPIT
摘要翻译：要解决的问题：提供一种坚固的并实现高速跟踪的图像处理装置，并提供其方法和图像处理程序。解决方案：图像处理装置100设置有：判别部152，其使用N个预先生成的特征提取部分151从输入图像中提取N个特征量，并计算显示目标对象可能性的可靠性从N提取特征量; 对象检测部分120，其基于可靠性来检测包括在输入图像中的目标对象; 特征选择部分140，其从N个特征提取部分151中选择M个特征提取部分，使得目标对象的可靠性与其背景的可靠性之间的分离程度高于通过使用N- 特征提取部分151; 以及对象追踪部140，其使用由特征选择部130选择的M个特征提取部，从输入图像中提取M个特征量，并使用提取的M个特征量来跟踪目标对象。版权所有（C）2010，JPO＆INPIT

8. 发明专利

JP2009198445A Device and method for object detection 有权
标题翻译：用于物体检测的装置和方法
公开(公告)号：JP2009198445A
公开(公告)日：2009-09-03
申请号：JP2008043109
申请日：2008-02-25
申请人： Toshiba Corp , 株式会社東芝
发明人： NAKANO TAKESHI , KUBOTA SUSUMU , OKAMOTO KYOICHI
IPC分类号： G01B11/00 , G01P3/36 , G06T1/00 , H04N5/225 , H04N5/232
CPC分类号： G06K9/00785 , G06T7/248 , G06T2207/10021 , G06T2207/30236
摘要： PROBLEM TO BE SOLVED: To detect an object robustly when it has influence of change of lighting or weather, or to detect a motion information, such as position and speed or the like of the object with sufficient accuracy, by the method of calculating three-dimensional coordinate of the object of pedestrian, bicycle or the like and detecting the object from information on height, and asking for parallax image using correlation operation for image acquired from stereo camera attached in pillar or the like which exists in road. SOLUTION: A processing area established according to position and height of the object and arrangement of stereo camera is established plurally for acquired image. The object of pedestrian, bicycle or the like is detected by applying the method using a principle of a plane projection stereo method for every established processing area. The object of whole image is detected by unifying the object detection result for every processing area. The motion information, such as position, speed or the like is detected by tracking the detected object sequentially. COPYRIGHT: (C)2009,JPO&INPIT
摘要翻译：要解决的问题：当其受到照明或天气变化的影响时，可以鲁棒地检测物体，或者通过以下方法检测物体的位置和速度等运动信息，即，计算行人，自行车等的物体的三维坐标，并根据高度信息检测物体，并且使用相关操作对由道路中存在的立柱等附接的立体摄像机获得的图像进行相关操作。解决方案：根据物体的位置和高度建立的处理区域和立体摄像机的布置被多次建立用于获取的图像。通过对于每个建立的处理区域应用使用平面投影立体声方法的原理的方法来检测行人，自行车等的对象。通过统一每个处理区域的对象检测结果来检测整个图像的对象。通过依次跟踪检测到的物体来检测位置，速度等运动信息。版权所有（C）2009，JPO＆INPIT

9. 发明专利

JP2004289709A Image pickup device, and image pickup method 审中-公开
标题翻译：图像拾取器件和图像拾取方法
公开(公告)号：JP2004289709A
公开(公告)日：2004-10-14
申请号：JP2003081840
申请日：2003-03-25
申请人： Toshiba Corp , 株式会社東芝
发明人： TANIGUCHI TAKAHIRO , ONOGUCHI KAZUNORI , NAKAI HIROAKI , HATTORI HIROSHI , KUBOTA SUSUMU , OKADA RYUZO , NAKAYAMA HIROMITSU , NAKANO TAKESHI , FURUKAWA KENJI
IPC分类号： H04N5/335 , H04N5/341 , H04N5/351 , H04N5/374
摘要： PROBLEM TO BE SOLVED: To suppress picture quality deterioration caused by image correction for dealing with shaking by accurately detecting the shaking of an image pickup device.
SOLUTION: The image pickup device is provided with: an image pickup part 100 for capturing both a moving image of a high frame rate and a moving image of an ordinary frame rate; a local region setting part 101 for setting a local region that serves as a reference for finding the extent of the shaking; a motion detecting part 102 for detecting a motion in the local region by using the moving image of the high frame rate; a shaking calculating part 103 for calculating the quantity of shaking between the images of the ordinary frame rate from the found motion vector; a shaking correcting part 104 for correcting the images of the ordinary frame rate on the basis of the extent of shaking; and an image output part 105 for outputting the corrected image.
COPYRIGHT: (C)2005,JPO&NCIPI
摘要翻译：要解决的问题：通过精确地检测图像拾取装置的抖动来抑制由用于处理抖动的图像校正引起的图像质量劣化。解决方案：图像拾取装置设置有：用于捕获高帧率的运动图像和普通帧率的运动图像的图像拾取部分100; 局部区域设定部101，用于设定作为用于求出振动程度的基准的局部区域; 运动检测部102，用于通过使用高帧率的运动图像来检测局部区域中的运动; 摇动计算部分103，用于根据找到的运动矢量计算普通帧率的图像之间的抖动量; 振动校正部104，用于根据振动的程度校正普通帧率的图像; 以及用于输出校正图像的图像输出部105。版权所有（C）2005，JPO＆NCIPI

10. 发明专利

JP2014123184A Recognition device, method, and program 审中-公开
标题翻译：识别设备，方法和程序
公开(公告)号：JP2014123184A
公开(公告)日：2014-07-03
申请号：JP2012277935
申请日：2012-12-20
申请人： Toshiba Corp , 株式会社東芝
发明人： NAKAI TOMOHIRO , KUBOTA SUSUMU , ITO SATOSHI
IPC分类号： G06T7/00
CPC分类号： G06K9/627
摘要： PROBLEM TO BE SOLVED: To provide a recognition device, method, and program capable of improving recognition accuracy and robustness.SOLUTION: The recognition device includes an acquisition unit, a storage unit, a first calculation unit, a second calculation unit, a determination unit, and an output unit. In the storage unit, a plurality of learning patterns each of which belongs to one of a plurality of categories are stored. The acquisition unit acquires a recognition object pattern to be recognized. The first calculation unit calculates, for each category, a distance histogram showing a distribution of the numbers of learning patterns belonging to the category to distances between the recognition object pattern and the learning patterns belonging to the category. A second calculation unit analyzes respective distance histograms of the plurality of categories to calculate reliabilities of the categories. The determination unit uses the reliabilities to determine a category of the recognition object pattern from among the plurality of categories. The output unit outputs the category of the recognition object pattern.
摘要翻译：要解决的问题：提供一种能够提高识别精度和鲁棒性的识别装置，方法和程序。解决方案：识别装置包括获取单元，存储单元，第一计算单元，第二计算单元，确定单元，以及输出单元。在存储单元中，存储多个属于多个类别之一的学习模式。获取单元获取要识别的识别对象图案。第一计算单元对于每个类别，计算出显示属于该类别的学习模式的数量的分布的距离直方图，该距离直方图包括识别对象模式和属于该类别的学习模式之间的距离。第二计算单元分析多个类别的各个距离直方图以计算类别的可靠性。确定单元使用可靠性来确定多个类别中的识别对象模式的类别。输出单元输出识别对象图案的类别。

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