公开(公告)号：US06608910B1

公开(公告)日：2003-08-19

申请号：US09389874

申请日：1999-09-02

申请人： Narayan Srinivasa ,  Yuri Owechko

发明人： Narayan Srinivasa ,  Yuri Owechko

IPC分类号： G06K900

CPC分类号： G06K9/00362 ,  G06K9/3241

摘要： An computer vision method and system for recognizing and tracking occupants in a fixed space under variable illumination. The system utilizes a camera to capture an initial image of the unoccupied fixed space and subsequently captures images of the occupied fixed space. The edge maps of the current estimate of the unoccupied fixed space including illumination variations and the occupied fixed space are computed. The edge map of the current estimate of the unoccupied fixed space is then subtracted from the edge map of the occupied fixed space to yield a residual edge map, which is then processed to extract the image of the occupant. At least one equivalent rectangle is then computed from the two-dimensional moments of the image of the occupant. The equivalent rectangles are then used to determine the occupant type and position and to track changes in real-time. This method and system is generally designed for use with automobile safety systems such as “smart” airbags. However, it may be tailored to many applications such as computer gaming, adaptive automotive controls, and “smart” homes, among others.

摘要翻译： 一种用于在可变照明下识别和跟踪固定空间中的乘客的计算机视觉方法和系统。 该系统利用相机捕获未占用的固定空间的初始图像，并随后捕获被占用的固定空间的图像。 计算包括照明变化和占用固定空间在内的未占用固定空间的当前估计的边缘图。 然后从占用的固定空间的边缘图中减去未占用的固定空间的当前估计的边缘图，以产生剩余边缘图，然后对其进行处理以提取乘员的图像。 然后从乘员的图像的二维矩中计算至少一个等效的矩形。 然后使用等效矩形来确定乘员类型和位置并实时跟踪变化。 该方法和系统通常设计用于诸如“智能”安全气囊之类的汽车安全系统。 然而，它可以针对诸如计算机游戏，适应性汽车控制和“智能”家庭等众多应用。

公开(公告)号：US5737083A

公开(公告)日：1998-04-07

申请号：US797705

申请日：1997-02-11

申请人： Yuri Owechko ,  K. Wojtek Przytula

发明人： Yuri Owechko ,  K. Wojtek Przytula

IPC分类号： B60N2/00 ,  B60R21/01 ,  B60R21/015 ,  G01S17/02 ,  G01S17/46 ,  G01S17/87 ,  G01S17/89 ,  G01B11/24

CPC分类号： B60R21/01542 ,  B60N2/002 ,  B60R21/01538 ,  G01S17/48 ,  G01S17/89 ,  G01S17/026 ,  G01S17/87 ,  G01S7/4814

摘要： An optical sensor system for measuring the approximate three-dimensional profile and position of an object. A reduced to practice embodiment of the optical sensor system has been used to measure the three-dimensional profile and position of an object within a range of 40 inches to 2 inches with high accuracy and high update rates (>1 KHz). The sensor system generates a set of optical beams using a projection lens and multiple light emitting diodes (LED) located in a back focal plane of the projection lens. The position of each LED determines the angle of a beam transmitted thereby. By turning on each LED sequentially in time, a set of beams is generated at various angles that illuminates the object to produce a set of spots on the object. The range from each LED to each of the spatially-separated illuminated spots on the object is determined by imaging the spots onto a two-dimensional transverse-effect photodiode. Signals derived from a pair of photocurrent outputs from the photodiode are processed to determine the positions of the spots on the two-dimensional transverse-effect photodiode. Computations are performed that implement optical triangulation to determine the range and approximate three-dimensional profile to the object.

摘要翻译： 一种用于测量物体的近似三维轮廓和位置的光学传感器系统。 已经使用光学传感器系统的简化实践实施例以高精度和高更新率（> 1KHz）在40英寸至2英寸的范围内测量物体的三维轮廓和位置。 传感器系统使用投影透镜和位于投影透镜的后焦平面中的多个发光二极管（LED）产生一组光束。 每个LED的位置确定由此传输的光束的角度。 通过按时间顺序地接通每个LED，产生一组以照射物体的各种角度的光束，以在物体上产生一组斑点。 通过将斑点成像到二维横向光电二极管上，确定物体上每个LED到每个空间分离的照明点的范围。 对来自光电二极管的一对光电流输出的信号进行处理，以确定二维横向光电二极管上斑点的位置。 执行计算，其实现光学三角测量以确定对象的范围和近似三维轮廓。

公开(公告)号：US20130287290A1

公开(公告)日：2013-10-31

申请号：US13459643

申请日：2012-04-30

申请人： Yuri Owechko

发明人： Yuri Owechko

IPC分类号： G06K9/00

CPC分类号： G06T7/344 ,  G06T17/05

摘要： An accurate, flexible and scalable technique for multi-modal image registration is described, a technique that does not need to rely on direct feature matching and does not need to rely on precise geometric models. The methods and/or systems described in this disclosure enable the registration (fusion) of multi-modal images of a scene with a three dimensional (3D) representation of the same scene using, among other information, viewpoint data from a sensor that generated a target image, as well as 3D-GeoArcs. The registration techniques of the present disclosure may be comprised of three main steps, as shown in FIG. 1. The first main step includes forming a 3D reference model of a scene. The second main step includes estimating the 3D geospatial viewpoint of a sensor that generated a target image using 3D-GeoArcs. The third main step includes projecting the target image's data into a composite 3D scene representation.

摘要翻译： 描述了一种用于多模式图像配准的精确，灵活和可扩展的技术，这种技术不需要依赖于直接特征匹配，并且不需要依赖于精确的几何模型。 在本公开中描述的方法和/或系统使得能够使用来自传感器的视点数据（其中生成一个或多个）的视点数据，将场景的多模态图像与同一场景的三维（3D）表示进行注册（融合） 目标图像，以及3D-GeoArcs。 本公开的注册技术可以由三个主要步骤组成，如图1所示。 第一个主要步骤包括形成场景的3D参考模型。 第二个主要步骤包括使用3D-GeoArcs估计生成目标图像的传感器的3D地理空间视点。 第三个主要步骤包括将目标图像的数据投影到复合3D场景表示中。

公开(公告)号：US08195591B1

公开(公告)日：2012-06-05

申请号：US12788229

申请日：2010-05-26

申请人： Yuri Owechko

发明人： Yuri Owechko

IPC分类号： G06F17/00 ,  G06N5/00

CPC分类号： G06K9/00369 ,  G06K9/6229 ,  G06K9/6292

摘要： Described is a signal processing system. The system comprises a signal processing module having signal processing parameters and being configured to receive a plurality of signals. The signal processing module uses the signal processing parameters to output a processed signal, as either a fused signal or a plurality of separate signals. A classification module is included to recognize information encoded in the processed signal to classify the information encoded in the process signal, with the classification having a confidence level. An optimization module is configured, in a feedback loop, to utilize the information encoded in the processed signal to adjust the signal processing parameters to optimize the confidence level of the classification, thereby optimizing an output of the signal processing module.

摘要翻译： 描述了一种信号处理系统。 该系统包括具有信号处理参数并被配置为接收多个信号的信号处理模块。 信号处理模块使用信号处理参数来输出经处理的信号，作为融合信号或多个单独的信号。 包括分类模块以识别经处理的信号中编码的信息，以对分类具有置信水平的过程信号中编码的信息进行分类。 在反馈环路中，优化模块被配置为利用经处理的信号中编码的信息来调整信号处理参数以优化分类的置信水平，从而优化信号处理模块的输出。

公开(公告)号：US07715591B2

公开(公告)日：2010-05-11

申请号：US10132875

申请日：2002-04-24

申请人： Yuri Owechko ,  Narayan Srinivasa ,  Swarup Medasani ,  Riccardo Boscolo

发明人： Yuri Owechko ,  Narayan Srinivasa ,  Swarup Medasani ,  Riccardo Boscolo

IPC分类号： G06K9/00

CPC分类号： G06K9/6292 ,  G06K9/00201 ,  G06K9/00369 ,  G06K9/00832 ,  G06K9/209 ,  G06K9/522 ,  G06T7/97 ,  G06T2207/10021 ,  G06T2207/20064 ,  G06T2207/30268 ,  H04N2013/0081 ,  H04N2013/0092

摘要： A vision-based system for automatically detecting the type of object within a specified area, such as the type of occupant within a vehicle is presented. The type of occupant can then be used to determine whether an airbag deployment system should be enabled or not. The system extracts different features, including wavelet features and/or a disparity map from images captured by image sensors. These features are then processed by classification algorithms to produce class confidences for various occupant types. The occupant class confidences are fused and processed to determine occupant type. In a preferred embodiment, image features from image edges, wavelet features, and disparity are used. Various classification algorithms may be implemented to classify the object. Use of the disparity map and/or wavelet features provides greater computational efficiency.

摘要翻译： 提出了一种基于视觉的系统，用于自动检测指定区域内物体的类型，例如车辆内乘客的类型。 然后可以使用乘客的类型来确定是否应启用安全气囊展开系统。 该系统从图像传感器捕获的图像中提取不同的特征，包括小波特征和/或视差图。 然后通过分类算法对这些特征进行处理，以便为各种乘客类型生成类别信息。 乘员班信心被融合和处理以确定乘客类型。 在优选实施例中，使用来自图像边缘的图像特征，小波特征和视差。 可以实现各种分类算法来对对象进行分类。 使用视差图和/或小波特征提供更大的计算效率。

公开(公告)号：US20070263936A1

公开(公告)日：2007-11-15

申请号：US11800265

申请日：2007-05-03

申请人： Yuri Owechko

发明人： Yuri Owechko

IPC分类号： G06K9/62

CPC分类号： G06K9/6292 ,  G06K9/00369 ,  G06K9/6229

摘要： Described is a signal processing system. The system comprises a signal processing module having signal processing parameters and being configured to receive a plurality of signals. The signal processing module uses the signal processing parameters to output a processed signal, as either a fused signal or a plurality of separate signals. A classification module is included to recognize information encoded in the processed signal to classify the information encoded in the process signal, with the classification having a confidence level. An optimization module is configured, in a feedback loop, to utilize the information encoded in the processed signal to adjust the signal processing parameters to optimize the confidence level of the classification, thereby optimizing an output of the signal processing module.

摘要翻译： 描述了一种信号处理系统。 该系统包括具有信号处理参数并被配置为接收多个信号的信号处理模块。 信号处理模块使用信号处理参数来输出经处理的信号，作为融合信号或多个单独的信号。 包括分类模块以识别经处理的信号中编码的信息，以对分类具有置信水平的过程信号中编码的信息进行分类。 在反馈环路中，优化模块被配置为利用经处理的信号中编码的信息来调整信号处理参数以优化分类的置信水平，从而优化信号处理模块的输出。

公开(公告)号：US20070183670A1

公开(公告)日：2007-08-09

申请号：US11433159

申请日：2006-05-12

申请人： Yuri Owechko ,  Swarup Medasani

发明人： Yuri Owechko ,  Swarup Medasani

IPC分类号： G06K9/62 ,  G06K9/46

CPC分类号： G06K9/6292 ,  G06K9/00369 ,  G06K9/6229

摘要： An object recognition system is described that incorporates swarming classifiers. The swarming classifiers comprise a plurality of software agents configured to operate as a cooperative swarm to classify an object group in a domain. Each node N represents an object in the group having K object attributes. Each agent is assigned an initial velocity vector to explore a KN-dimensional solution space for solutions matching the agent's graph. Further, each agent is configured to search the solution space for an optimum solution. The agents keep track of their coordinates in the KN-dimensional solution space that are associated with an observed best solution (pbest) and a global best solution (gbest). The gbest is used to store the best solution among all agents which corresponds to a best graph among all agents. Each velocity vector thereafter changes towards pbest and gbest, allowing the cooperative swarm to classify of the object group.

摘要翻译： 描述了包含群组分类器的对象识别系统。 群集分类器包括被配置为作为协作群进行操作以将域中的对象组分类的多个软件代理。 每个节点N表示具有K个对象属性的组中的对象。 为每个代理分配一个初始速度向量，以探索与代理图相匹配的解决方案的KN维解决方案空间。 此外，每个代理被配置为搜索解空间以获得最佳解决方案。 代理人跟踪与观察到的最佳解决方案（pbest）和全局最佳解决方案（gbest）相关联的KN维解决方案空间中的坐标。 gbest用于在所有代理之间存储对应于最佳图形的所有代理中的最佳解决方案。 其后每个速度矢量向pbest和gbest变化，允许协作群对目标群进行分类。

公开(公告)号：US07133699B1

公开(公告)日：2006-11-07

申请号：US10421167

申请日：2003-04-22

申请人： Yuri Owechko ,  David B. Shu

发明人： Yuri Owechko ,  David B. Shu

IPC分类号： H04M1/00

CPC分类号： H04B7/0854 ,  G01S3/74

摘要： A method of separating mixed wireless signals is provided. The method includes receiving, at an antenna comprising a first quantity of antenna elements, mixed signals comprising a mixture of source signals communicated from a second quantity of wireless signal sources, and separating the mixed signals to estimate the source signals. The second quantity is greater than the first quantity, and the source signals communicated from at least one of the wireless signal sources are received at the antenna as complex signals.

摘要翻译： 提供了一种分离混合无线信号的方法。 该方法包括在包括第一数量的天线元件的天线处接收包括从第二数量的无线信号源传送的源信号的混合的混合信号，以及分离混合信号以估计源信号。 第二数量大于第一数量，并且从天线中的至少一个无线信号源传送的源信号作为复信号被接收。

公开(公告)号：US06642887B2

公开(公告)日：2003-11-04

申请号：US09799821

申请日：2001-02-28

申请人： Yuri Owechko

发明人： Yuri Owechko

IPC分类号： H01Q322

CPC分类号： H01Q1/246 ,  H01Q3/2605 ,  H04B7/024

摘要： An adaptive antenna beamformer is presented, said beamformer including a first signal source 100 and a second signal source 102 and an ambient noise or other interfering signals source 106, which are exposed to an antenna array 104. The mixed signals 108 are provided to a blind source separation processor 110. The blind separation processor 110, in this case an Independent Component Analysis element, is comprised of a group of processes that are configured to separate mixtures of signals blindly. The blind separation processor 110, provides three outputs, a first signal output 112, a second signal output 114, and a third signal output 116. The signal outputs each correspond to their respective signal input.

摘要翻译： 呈现自适应天线波束形成器，所述波束形成器包括暴露于天线阵列104的第一信号源100和第二信号源102以及环境噪声或其它干扰信号源106.混合信号108被提供给盲 源分离处理器110.盲分离处理器110（在这种情况下为独立分量分析元件）包括被配置为盲目地分离信号混合的一组处理。 盲分离处理器110提供三个输出，第一信号输出112，第二信号输出114和第三信号输出116.每个信号输出对应于它们各自的信号输入。

公开(公告)号：US06564198B1

公开(公告)日：2003-05-13

申请号：US09504641

申请日：2000-02-16

申请人： Srinivasa Narayan ,  Yuri Owechko

发明人： Srinivasa Narayan ,  Yuri Owechko

IPC分类号： G06F1700

CPC分类号： G06K9/626 ,  G06K9/00369 ,  G06K9/00832 ,  G06N3/0427 ,  G06N7/046

摘要： An method and apparatus for extracting an interpretable, meaningful, and concise rule set from neural networks is presented. The method involves adjustment of gain parameter, &lgr; and the threshold, Tj for the sigmoid activation function of the interactive-or operator used in the extraction/development of a rule set from an artificial neural network. A multi-stage procedure involving coarse and fine adjustment is used in order to constrain the range of the antecedents of the extracted rules to the range of values of the inputs to the artificial neural network. Furthermore, the consequents of the extracted rules are provided based on degree of membership such that they are easily understandable by human beings. The method disclosed may be applied to any pattern recognition task, and is particularly useful in applications such as vehicle occupant sensing and recognition, object recognition, gesture recognition, and facial pattern recognition, among others.

摘要翻译： 提出了一种用于从神经网络中提取可解释，有意义和简洁的规则集的方法和装置。 该方法包括调整增益参数，lambd和阈值Tj，用于从人造神经网络提取/开发规则集中使用的交互式或运算符的S形激活函数。 使用涉及粗调和微调的多级程序，以便将提取的规则的前提的范围限制为人造神经网络的输入值的范围。 此外，提取的规则的结果是基于成员的程度提供，使得它们容易被人理解。 所公开的方法可以应用于任何模式识别任务，并且在诸如车辆乘员感测和识别，对象识别，手势识别和面部模式识别等应用中特别有用。