公开(公告)号：US20080025568A1

公开(公告)日：2008-01-31

申请号：US11780109

申请日：2007-07-19

申请人： Feng Han ,  Ying Shan ,  Ryan Cekander ,  Harpreet Sawhney ,  Rakesh Kumar

发明人： Feng Han ,  Ying Shan ,  Ryan Cekander ,  Harpreet Sawhney ,  Rakesh Kumar

IPC分类号： G06K9/00

CPC分类号： G06K9/4642

摘要： The present invention provides an improved system and method for object detection with histogram of oriented gradient (HOG) based support vector machine (SVM). Specifically, the system provides a computational framework to stably detect still or not moving objects over a wide range of viewpoints. The framework includes providing a sensor input of images which are received by the “focus of attention” mechanism to identify the regions in the image that potentially contain the target objects. These regions are further computed to generate hypothesized objects, specifically generating selected regions containing the target object hypothesis with respect to their positions. Thereafter, these selected regions are verified by an extended HOG-based SVM classifier to generate the detected objects.

摘要翻译： 本发明提供一种用于基于定向梯度（HOG）的支持向量机（SVM）直方图的物体检测的改进的系统和方法。 具体地，该系统提供了一个计算框架，用于在宽范围的视点中稳定地检测静止或不移动的物体。 框架包括提供通过“注意力”机制接收的图像的传感器输入，以识别可能包含目标对象的图像中的区域。 进一步计算这些区域以产生假设对象，特别地生成包含关于其位置的目标对象假设的选定区域。 此后，通过扩展的基于HOG的SVM分类器验证这些选择的区域以生成检测到的对象。

公开(公告)号：US07853072B2

公开(公告)日：2010-12-14

申请号：US11780109

申请日：2007-07-19

申请人： Feng Han ,  Ying Shan ,  Ryan Cekander ,  Harpreet S. Sawhney ,  Rakesh Kumar

发明人： Feng Han ,  Ying Shan ,  Ryan Cekander ,  Harpreet S. Sawhney ,  Rakesh Kumar

IPC分类号： G06K9/62

CPC分类号： G06K9/4642

摘要： The present invention provides an improved system and method for object detection with histogram of oriented gradient (HOG) based support vector machine (SVM). Specifically, the system provides a computational framework to stably detect still or not moving objects over a wide range of viewpoints. The framework includes providing a sensor input of images which are received by the “focus of attention” mechanism to identify the regions in the image that potentially contain the target objects. These regions are further computed to generate hypothesized objects, specifically generating selected regions containing the target object hypothesis with respect to their positions. Thereafter, these selected regions are verified by an extended HOG-based SVM classifier to generate the detected objects.

摘要翻译： 本发明提供一种用于基于定向梯度（HOG）的支持向量机（SVM）直方图的物体检测的改进的系统和方法。 具体地，该系统提供了一个计算框架，用于在宽范围的视点中稳定地检测静止或不移动的物体。 框架包括提供通过“注意力”机制接收的图像的传感器输入，以识别可能包含目标对象的图像中的区域。 进一步计算这些区域以产生假设对象，特别地生成包含关于其位置的目标对象假设的选定区域。 此后，通过扩展的基于HOG的SVM分类器验证这些选择的区域以生成检测到的对象。

公开(公告)号：US20080089579A1

公开(公告)日：2008-04-17

申请号：US11762400

申请日：2007-06-13

申请人： Feng Han ,  Ying Shan ,  Harpreet Sawhney ,  Rakesh Kumar

发明人： Feng Han ,  Ying Shan ,  Harpreet Sawhney ,  Rakesh Kumar

IPC分类号： G06K9/62

CPC分类号： G06K9/6256

摘要： The present invention provides a computer implemented process for detecting multi-view multi-pose objects. The process comprises training of a classifier for each intra-class exemplar, training of a strong classifier and combining the individual exemplar-based classifiers with a single objective function. This function is optimized using the two nested AdaBoost loops. The first loop is the outer loop that selects discriminative candidate exemplars. The second loop, the inner loop selects the discriminative candidate features on the selected exemplars to compute all weak classifiers for a specific position such as a view/pose. Then all the computed weak classifiers are automatically combined into a final classifier (strong classifier) which is the object to be detected.

摘要翻译： 本发明提供了一种用于检测多视点多姿态对象的计算机实现过程。 该过程包括针对每个类内样本的分类器的训练，强分类器的训练和将单个基于样本的分类器与单个目标函数组合。 使用两个嵌套的AdaBoost循环来优化此功能。 第一个循环是选择区分候选样本的外循环。 第二个循环，内循环选择所选样本上的鉴别候选特征，以计算特定位置（例如视图/姿态）的所有弱分类器。 然后将所有计算的弱分类器自动组合成最终分类器（强分类器），该分类器是要检测的对象。

公开(公告)号：US20120002869A1

公开(公告)日：2012-01-05

申请号：US13134885

申请日：2011-06-20

申请人： Feng Han ,  Ying Shan ,  Harpreet Singh Sawhney ,  Rakesh Kumar

发明人： Feng Han ,  Ying Shan ,  Harpreet Singh Sawhney ,  Rakesh Kumar

IPC分类号： G06K9/62

CPC分类号： G06K9/6256

摘要： The present invention provides a computer implemented process for detecting multi-view multi-pose objects. The process comprises training of a classifier for each intra-class exemplar, training of a strong classifier and combining the individual exemplar-based classifiers with a single objective function. This function is optimized using the two nested AdaBoost loops. The first loop is the outer loop that selects discriminative candidate exemplars. The second loop, the inner loop selects the discriminative candidate features on the selected exemplars to compute all weak classifiers for a specific position such as a view/pose. Then all the computed weak classifiers are automatically combined into a final classifier (strong classifier) which is the object to be detected.

摘要翻译： 本发明提供了一种用于检测多视点多姿态对象的计算机实现过程。 该过程包括针对每个类内样本的分类器的训练，强分类器的训练和将单个基于样本的分类器与单个目标函数组合。 使用两个嵌套的AdaBoost循环来优化此功能。 第一个循环是选择区分候选样本的外循环。 第二个循环，内循环选择所选样本上的鉴别候选特征，以计算特定位置（例如视图/姿态）的所有弱分类器。 然后将所有计算的弱分类器自动组合成最终分类器（强分类器），该分类器是要检测的对象。

公开(公告)号：US08391592B2

公开(公告)日：2013-03-05

申请号：US13134885

申请日：2011-06-20

申请人： Feng Han ,  Ying Shan ,  Harpreet Singh Sawhney ,  Rakesh Kumar

发明人： Feng Han ,  Ying Shan ,  Harpreet Singh Sawhney ,  Rakesh Kumar

IPC分类号： G06K9/62

CPC分类号： G06K9/6256

摘要： The present invention provides a computer implemented process for detecting multi-view multi-pose objects. The process comprises training of a classifier for each intra-class exemplar, training of a strong classifier and combining the individual exemplar-based classifiers with a single objective function. This function is optimized using the two nested AdaBoost loops. The first loop is the outer loop that selects discriminative candidate exemplars. The second loop, the inner loop selects the discriminative candidate features on the selected exemplars to compute all weak classifiers for a specific position such as a view/pose. Then all the computed weak classifiers are automatically combined into a final classifier (strong classifier) which is the object to be detected.

摘要翻译： 本发明提供了一种用于检测多视点多姿态对象的计算机实现过程。 该过程包括针对每个类内样本的分类器的训练，强分类器的训练和将单个基于样本的分类器与单个目标函数组合。 使用两个嵌套的AdaBoost循环来优化此功能。 第一个循环是选择区分候选样本的外循环。 第二个循环，内循环选择所选样本上的鉴别候选特征，以计算特定位置（例如视图/姿态）的所有弱分类器。 然后将所有计算的弱分类器自动组合成最终分类器（强分类器），该分类器是要检测的对象。

公开(公告)号：US07965886B2

公开(公告)日：2011-06-21

申请号：US11762400

申请日：2007-06-13

申请人： Feng Han ,  Ying Shan ,  Harpreet Singh Sawhney ,  Rakesh Kumar

发明人： Feng Han ,  Ying Shan ,  Harpreet Singh Sawhney ,  Rakesh Kumar

IPC分类号： G06K9/62

CPC分类号： G06K9/6256

摘要： The present invention provides a computer implemented process for detecting multi-view multi-pose objects. The process comprises training of a classifier for each intra-class exemplar, training of a strong classifier and combining the individual exemplar-based classifiers with a single objective function. This function is optimized using the two nested AdaBoost loops. The first loop is the outer loop that selects discriminative candidate exemplars. The second loop, the inner loop selects the discriminative candidate features on the selected exemplars to compute all weak classifiers for a specific position such as a view/pose. Then all the computed weak classifiers are automatically combined into a final classifier (strong classifier) which is the object to be detected.

摘要翻译： 本发明提供了一种用于检测多视点多姿态对象的计算机实现过程。 该过程包括针对每个类内样本的分类器的训练，强分类器的训练和将单个基于样本的分类器与单个目标函数组合。 使用两个嵌套的AdaBoost循环来优化此功能。 第一个循环是选择区分候选样本的外循环。 第二个循环，内循环选择所选样本上的鉴别候选特征，以计算特定位置（例如视图/姿态）的所有弱分类器。 然后将所有计算的弱分类器自动组合成最终分类器（强分类器），该分类器是要检测的对象。

公开(公告)号：US08340349B2

公开(公告)日：2012-12-25

申请号：US11763559

申请日：2007-06-15

申请人： Garbis Salgian ,  Supun Samarasekera ,  Jiangjian Xiao ,  James Russell Bergen ,  Rakesh Kumar ,  Feng Han

发明人： Garbis Salgian ,  Supun Samarasekera ,  Jiangjian Xiao ,  James Russell Bergen ,  Rakesh Kumar ,  Feng Han

IPC分类号： G06K9/00

CPC分类号： G06K9/3241 ,  G06K9/209 ,  G06T7/285 ,  G06T2207/10021 ,  G06T2207/30212 ,  G06T2207/30252

摘要： A method for detecting a moving target is disclosed that receives a plurality of images from at least one camera; receives a measurement of scale from one of a measurement device and a second camera; calculates the pose of the at least one camera over time based on the plurality of images and the measurement of scale; selects a reference image and an inspection image from the plurality of images of the at least one camera; and detects a moving target from the reference image and the inspection image based on the orientation of corresponding portions in the reference image and the inspection image relative to a location of an epipolar direction common to the reference image and the inspection image; and displays any detected moving target on a display. The measurement of scale can derived from a second camera or, for example, a wheel odometer. The method can also detect moving targets by combining the above epipolar method with a method based on changes in depth between the inspection image and the reference image and based on changes in flow between the inspection image and the reference image.

摘要翻译： 公开了一种用于检测移动目标的方法，其从至少一个相机接收多个图像; 从测量装置和第二相机中的一个接收刻度的测量; 基于多个图像和尺度的测量来计算随时间的至少一个相机的姿态; 从所述至少一个照相机的多个图像中选择参考图像和检查图像; 并且基于所述参考图像和所述检查图像中的对应部分相对于所述参考图像和所述检查图像共同的核极方向的位置的取向，从所述参考图像和所述检查图像中检测移动目标; 并在显示器上显示任何检测到的移动目标。 刻度的测量可以从第二相机或例如车轮里程表得到。 该方法还可以通过将上述对极方法与基于检查图像和参考图像之间的深度变化的方法并基于检查图像与参考图像之间的流动变化组合来检测移动目标。

公开(公告)号：US20100202657A1

公开(公告)日：2010-08-12

申请号：US12604298

申请日：2009-10-22

申请人： Garbis Salgian ,  John Benjamin Southall ,  Sang-Hack Jung ,  Vlad Branzoi ,  Jiangjian Xiao ,  Feng Han ,  Supun Samarasekera ,  Rakesh Kumar ,  Jayan Eledath

发明人： Garbis Salgian ,  John Benjamin Southall ,  Sang-Hack Jung ,  Vlad Branzoi ,  Jiangjian Xiao ,  Feng Han ,  Supun Samarasekera ,  Rakesh Kumar ,  Jayan Eledath

IPC分类号： G06K9/48 ,  G06K9/00

CPC分类号： G06K9/00369 ,  G06K9/00805

摘要： The present invention relates to a system and method for detecting one or more targets belonging to a first class (e.g., moving and/or stationary people), from a moving platform in a 3D-rich environment. The framework described here is implemented using a number of monocular or stereo cameras distributed around the vehicle to provide 360 degrees coverage. Furthermore, the framework described here utilizes numerous filters to reduce the number of false positive identifications of the targets.

摘要翻译： 本发明涉及一种用于在富3D环境中从移动平台检测属于第一类（例如移动和/或固定人物）的一个或多个目标的系统和方法。 这里描述的框架使用分布在车辆周围的多个单目或立体摄像机来实现，以提供360度覆盖。 此外，这里描述的框架利用许多过滤器来减少目标的假阳性识别的数量。

公开(公告)号：US08355539B2

公开(公告)日：2013-01-15

申请号：US12146897

申请日：2008-06-26

申请人： Yi Tan ,  Feng Han ,  Jayan Eledath ,  Rakesh Kumar ,  Faroog Abdel-kareem Ibrahim

发明人： Yi Tan ,  Feng Han ,  Jayan Eledath ,  Rakesh Kumar ,  Faroog Abdel-kareem Ibrahim

IPC分类号： G06K9/00

CPC分类号： G06K9/00798

摘要： A method for determining whether a target vehicle in front of a host vehicle intends to change lanes using radar data and image data is disclosed, comprising the steps of processing the image data to detect the boundaries of the lane of the host vehicle; estimating a ground plane by determining a projected vanishing point of the detected lane boundaries; using a camera projection matrix to map the target vehicle from the radar data to image coordinates; and determining lane change intentions of the target vehicle based on a moving trajectory and an appearance change of the target vehicle. Determining lane change intentions based on a moving trajectory of the target vehicle is based on vehicle motion trajectory relative to the center of the lane such that the relative distance of the target vehicle from the center of the lane follows a predetermined trend. Determining lane change intentions based on an appearance change of the target vehicle is based on a template that tracks changes to the appearance of the rear part of the target vehicle due to rotation.

摘要翻译： 公开了一种用于确定本车辆前方的目标车辆是否使用雷达数据和图像数据来改变车道的方法，包括以下步骤：处理图像数据以检测主车辆的车道边界; 通过确定检测到的车道边界的预计消失点来估计接地平面; 使用相机投影矩阵将目标车辆从雷达数据映射到图像坐标; 以及基于所述目标车辆的移动轨迹和外观变化来确定所述目标车辆的车道改变意图。 基于目标车辆的移动轨迹确定车道改变意图是基于相对于车道中心的车辆运动轨迹，使得目标车辆与车道中心的相对距离遵循预定趋势。 基于目标车辆的外观变化确定车道改变意图是基于跟踪由于旋转而导致的目标车辆的后部的外观的变化的模板。

公开(公告)号：US20080089556A1

公开(公告)日：2008-04-17

申请号：US11763559

申请日：2007-06-15

申请人： Garbis Salgian ,  Supun Samarasekera ,  Jiangjian Xiao ,  James Bergen ,  Rakesh Kumar ,  Feng Han

发明人： Garbis Salgian ,  Supun Samarasekera ,  Jiangjian Xiao ,  James Bergen ,  Rakesh Kumar ,  Feng Han

IPC分类号： G06K9/00

CPC分类号： G06K9/3241 ,  G06K9/209 ,  G06T7/285 ,  G06T2207/10021 ,  G06T2207/30212 ,  G06T2207/30252

摘要： A method for detecting a moving target is disclosed that receives a plurality of images from at least one camera; receives a measurement of scale from one of a measurement device and a second camera; calculates the pose of the at least one camera over time based on the plurality of images and the measurement of scale; selects a reference image and an inspection image from the plurality of images of the at least one camera; and detects a moving target from the reference image and the inspection image based on the orientation of corresponding portions in the reference image and the inspection image relative to a location of an epipolar direction common to the reference image and the inspection image; and displays any detected moving target on a display. The measurement of scale can derived from a second camera or, for example, a wheel odometer. The method can also detect moving targets by combining the above epipolar method with a method based on changes in depth between the inspection image and the reference image and based on changes in flow between the inspection image and the reference image.

摘要翻译： 公开了一种用于检测移动目标的方法，其从至少一个相机接收多个图像; 从测量装置和第二相机中的一个接收刻度的测量; 基于多个图像和尺度的测量来计算随时间的至少一个相机的姿态; 从所述至少一个照相机的多个图像中选择参考图像和检查图像; 并且基于所述参考图像和所述检查图像中的对应部分相对于所述参考图像和所述检查图像共同的核极方向的位置的取向，从所述参考图像和所述检查图像中检测移动目标; 并在显示器上显示任何检测到的移动目标。 刻度的测量可以从第二相机或例如车轮里程表得到。 该方法还可以通过将上述对极方法与基于检查图像和参考图像之间的深度变化的方法并基于检查图像与参考图像之间的流动变化组合来检测移动目标。