公开(公告)号：US20160267304A1

公开(公告)日：2016-09-15

申请号：US14643721

申请日：2015-03-10

Applicant: Xerox Corporation

Inventor： Wencheng Wu ,  Robert P. Loce ,  Raja Bala ,  Peter Paul ,  Thomas F. Wade ,  Dennis L. Venable ,  Steven R. Moore ,  Peter J. Knausdorf

IPC: G06K7/10 ,  G06T5/00 ,  G06K7/14

CPC classification number: G06K7/10732 ,  G06K7/10574 ,  G06K7/10722 ,  G06K7/10821 ,  G06K7/1443 ,  G06K7/146 ,  G06K7/1465 ,  G06T5/001 ,  G06T2207/10152

Abstract: Eliminate or reduce the impact of glare in printed information tag recognition applications using single- and multi-pose external illumination coupled with intelligent processing. A shelf imager can acquire shelf images for printed information tag localization and recognition. An external illuminator can provide at least one illumination condition/pose for shelf image acquisition in addition to lighting associated with the enclosed environment. A glare region of interest (ROI) detector can analyze all or a portion of the acquired shelf images for glare to determine whether additional images need to be acquired using different illumination conditions provided by the single- or multi-pose external illuminator or whether full or portion of acquired images need to be analyzed by a printed information tag locator and recognizer. A printed information tag locator and recognizer can analyze all or a portion of the acquired images to localize and recognize data printed on the printed information tags.

Abstract translation: 使用单一和多姿态外部照明加上智能处理，消除或减少眩光对印刷信息标签识别应用的影响。 架子成像器可以获取用于印刷信息标签定位和识别的搁架图像。 除了与封闭环境相关联的照明之外，外部照明器还可以为搁架图像采集提供至少一个照明条件/姿势。 眩光感兴趣区域（ROI）检测器可以分析所获取的搁架图像的全部或一部分用于眩光，以确定是否需要使用由单一或多姿态外部照明器提供的不同照明条件获取附加图像， 所获取的图像的一部分需要通过打印的信息标签定位器和识别器进行分析。 打印的信息标签定位器和识别器可以分析所获取的图像的全部或一部分，以便本地化并识别打印在打印信息标签上的数据。

公开(公告)号：US20160173771A1

公开(公告)日：2016-06-16

申请号：US14566835

申请日：2014-12-11

Applicant: Xerox Corporation

Inventor： Edgar A. Bernal ,  Xuejin Wen ,  Qun Li ,  Raja Bala

IPC: H04N5/232

CPC classification number: H04N5/23232 ,  H04N5/2254 ,  H04N5/2259

Abstract: A method and system for reconstructing an image of a scene comprises configuring a digital light modulator according to a spatially varying pattern. Light energy associated with the scene and incident on the spatially varying pattern is collected and optically focused on the photodetectors. Data indicative of the intensity of the focused light energy from each of said at least two photodetectors is collected. Data from the photodetectors is then combined to reconstruct an image of the scene.

Abstract translation: 用于重建场景的图像的方法和系统包括根据空间变化的图案配置数字光调制器。 收集与场景相关的光能和空间变化图案上的入射光，并光学地聚焦在光电探测器上。 收集指示来自所述至少两个光电检测器中的每一个的聚焦光能的强度的数据。 然后将来自光电检测器的数据组合以重建场景的图像。

公开(公告)号：US09317780B2

公开(公告)日：2016-04-19

申请号：US14326635

申请日：2014-07-09

Applicant: Xerox Corporation

Inventor： Raja Bala ,  Zhigang Fan ,  Aaron Burry ,  Vishal Monga ,  Xuan Mo

IPC: G06K9/00 ,  G06K9/62

CPC classification number: G06K9/6249 ,  G06K9/00771

Abstract: Methods and systems for detecting anomalies in transportation related video footage. In an offline training phase, receiving video footage of a traffic location can be received. Also, in an offline training phase, event encodings can be extracted from the video footage and collected or compiled into a training dictionary. One or more input video sequences captured at the traffic location or a similar traffic location can be received in an online detection phase. Then, an event encoding corresponding to the input video sequence can be extracted. The event encoding can be reconstructed with a low rank sparsity prior model applied with respect to the training dictionary. The reconstruction error between actual and reconstructed event encodings can then be computed in order to determine if an event thereof is anomalous by comparing the reconstruction error with a threshold.

Abstract translation: 检测交通相关影像异常的方法和系统。 在离线训练阶段，可以接收到接收到交通位置的录像带。 此外，在离线训练阶段，可以从视频素材中提取事件编码，并将其收集或编译成训练词典。 在在线检测阶段可以接收在业务位置或类似业务位置处捕获的一个或多个输入视频序列。 然后，可以提取与输入视频序列相对应的事件编码。 可以使用相对于训练词典应用的低秩稀疏性先验模型来重构事件编码。 然后可以计算实际和重建事件编码之间的重建误差，以便通过将重建误差与阈值进行比较来确定其事件是否是异常的。

公开(公告)号：US20160037071A1

公开(公告)日：2016-02-04

申请号：US14885186

申请日：2015-10-16

Applicant: Xerox Corporation

Inventor： Phillip J. Emmett ,  Raja Bala ,  Michael R. Furst ,  Lina Fu

IPC: H04N5/232 ,  G06K9/00

CPC classification number: H04N5/23245 ,  G06K9/00355 ,  G06K9/00442 ,  G06K9/00496 ,  G06K9/036 ,  G06K9/228 ,  G06K9/32

Abstract: A system creates an electronic file corresponding to a printed artifact by launching a video capture module that causes a mobile electronic device to capture a video of a scene that includes the printed artifact. The system analyzes image frames in the video in real time as the video is captured to identify a suitable instance. In one example, the suitable instance is a frame or sequence of frames that contain an image of a page or side of the printed artifact and that do not exhibit a page-turn event. In response to identification of the suitable instance, the system will automatically cause a photo capture module of the device to capture a still image of the printed artifact. The still image has a resolution that is higher than that of the image frames in the video. The system will save the captured still images to a computer-readable file.

Abstract translation: 系统通过启动视频捕获模块来创建与打印的伪像相对应的电子文件，该视频捕获模块使得移动电子设备捕获包括打印伪像的场景的视频。 当捕获视频以识别合适的实例时，系统实时分析视频中的图像帧。 在一个示例中，合适的实例是帧或帧序列，其包含打印伪像的页面或侧面的图像，并且不呈现翻页事件。 响应于合适的实例的识别，系统将自动地使设备的照片捕获模块捕获打印的伪像的静止图像。 静止图像的分辨率高于视频中的图像帧。 系统将捕获的静态图像保存到计算机可读文件中。

公开(公告)号：US09247136B2

公开(公告)日：2016-01-26

申请号：US13972346

申请日：2013-08-21

Applicant: Xerox Corporation

Inventor： Phillip J. Emmett ,  Raja Bala ,  Michael R. Furst ,  Lina Fu

IPC: H04N5/232

CPC classification number: H04N5/23245 ,  G06K9/00355 ,  G06K9/00442 ,  G06K9/00496 ,  G06K9/036 ,  G06K9/228 ,  G06K9/32

Abstract: A system creates an electronic file corresponding to a printed artifact by launching a video capture module that causes an imaging sensor of a mobile electronic device to capture a video of a scene that includes the printed artifact. The system analyzes image frames in the video in real time as the video is captured to identify a suitable instance. The suitable instance is a frame or sequence of frames that contains at least a portion of the printed artifact and that also satisfies one or more image quality criteria or other criteria. Upon identification of each suitable instance, the system will automatically cause a photo capture module of the device to capture a still image of the printed artifact. The still image has a resolution that is higher than that of the image frames in the video. The system will save the captured still images to a computer-readable file.

Abstract translation: 系统通过启动视频捕获模块来创建与打印的伪影相对应的电子文件，所述视频捕获模块使得移动电子设备的成像传感器捕获包括打印伪像的场景的视频。 当捕获视频以识别合适的实例时，系统实时分析视频中的图像帧。 合适的实例是包含打印伪像的至少一部分并且还满足一个或多个图像质量标准或其他标准的帧或帧序列。 在识别每个合适的实例之后，系统将自动地使得设备的照片捕获模块捕获打印的伪像的静止图像。 静止图像的分辨率高于视频中的图像帧。 系统将捕获的静态图像保存到计算机可读文件中。

公开(公告)号：US20150213323A1

公开(公告)日：2015-07-30

申请号：US14534790

申请日：2014-11-06

Applicant: Xerox Corporation ,  The Penn State Research Foundation

Inventor： Raja Bala ,  Aaron M. Burry ,  Vishal Monga ,  Xuan Mo

IPC: G06K9/52 ,  G06K9/00 ,  G06K9/66

CPC classification number: G06K9/00785 ,  G06K9/00335

Abstract: Methods, systems, and processor-readable media for video anomaly detection based upon a sparsity model. A video input can be received and two or more diverse descriptors of an event can be computed from the video input. The descriptors can be combined to form an event matrix. A sparse reconstruction of the event matrix can be performed with respect to an over complete dictionary of training events represented by the diverse descriptors. A step can then be performed to determine if the event is anomalous by computing an outlier rejection measure on the sparse reconstruction.

Abstract translation: 基于稀疏模型的视频异常检测的方法，系统和处理器可读介质。 可以接收视频输入，并且可以从视频输入计算事件的两个或多个不同描述符。 描述符可以组合形成一个事件矩阵。 相对于由不同描述符表示的训练事件的完整字典，可以执行事件矩阵的稀疏重建。 然后可以执行一个步骤，以通过计算稀疏重建的异常值拒绝测量来确定事件是否是异常的。

公开(公告)号：US20150200922A1

公开(公告)日：2015-07-16

申请号：US14154519

申请日：2014-01-14

Applicant: Xerox Corporation

Inventor： Reiner Eschbach ,  Raja Bala ,  Saurabh Prabhat ,  Michael R. Furst

IPC: H04L29/06 ,  G06F21/60 ,  G06K15/00

CPC classification number: H04L63/08 ,  G06F17/211 ,  G06F17/2229 ,  G06F17/241 ,  G06F21/60 ,  G06F21/6227 ,  G06F2221/2113 ,  G06K15/4095 ,  H04L63/04

Abstract: A document presentation system routes a document having sensitive data to various users, wherein the various users have different levels of permission to access the sensitive data. When any user displays the document on a display of an electronic device, the display will show document so that sensitive data is replaced with an augmented reality (AR) marker. The AR marker may include a descriptor of the class of data to which the sensitive data belongs. The system will also display an AR overlay for each AR marker. For each user, the AR overlay for each AR marker will include none, some, or all of the sensitive data corresponding to the AR marker. The amount of the sensitive data that will be displayed will depend on the user's authorization level.

Abstract translation: 文档呈现系统将具有敏感数据的文档路由到各种用户，其中各种用户具有访问敏感数据的不同级别的许可。 当任何用户在电子设备的显示器上显示文档时，显示器将显示文档，以便敏感数据被替换为增强现实（AR）标记。 AR标记可以包括敏感数据所属的数据类别的描述符。 系统还将显示每个AR标记的AR重叠。 对于每个用户，每个AR标记的AR重叠将包括对应于AR标记的任何，一些或全部敏感数据。 将显示的敏感数据量将取决于用户的授权级别。

公开(公告)号：US20150110357A1

公开(公告)日：2015-04-23

申请号：US14326635

申请日：2014-07-09

Applicant: Xerox Corporation

Inventor： Raja Bala ,  Zhigang Fan ,  Aaron Burry ,  Vishal Monga ,  Xuan Mo

IPC: G06K9/00

CPC classification number: G06K9/6249 ,  G06K9/00771

Abstract: Methods and systems for detecting anomalies in transportation related video footage. In an offline training phase, receiving video footage of a traffic location can be received. Also, in an offline training phase, event encodings can be extracted from the video footage and collected or compiled into a training dictionary. One or more input video sequences captured at the traffic location or a similar traffic location can be received in an online detection phase. Then, an event encoding corresponding to the input video sequence can be extracted. The event encoding can be reconstructed with a low rank sparsity prior model applied with respect to the training dictionary. The reconstruction error between actual and reconstructed event encodings can then be computed in order to determine if an event thereof is anomalous by comparing the reconstruction error with a threshold.

Abstract translation: 检测交通相关影像异常的方法和系统。 在离线训练阶段，可以接收到接收到交通位置的录像带。 此外，在离线训练阶段，可以从视频素材中提取事件编码，并将其收集或编译成训练词典。 在在线检测阶段可以接收在业务位置或相似的业务位置处捕获的一个或多个输入视频序列。 然后，可以提取与输入视频序列相对应的事件编码。 可以使用相对于训练词典应用的低秩稀疏性先验模型来重构事件编码。 然后可以计算实际和重建事件编码之间的重建误差，以便通过将重建误差与阈值进行比较来确定其事件是否是异常的。

公开(公告)号：US20150048934A1

公开(公告)日：2015-02-19

申请号：US14451978

申请日：2014-08-05

Applicant: Xerox Corporation

Inventor： Robert P. Loce ,  Raja Bala ,  Michael R. Furst

IPC: G08G1/16

CPC classification number: G08G1/166

Abstract: Systems and methods are provided for real time dynamic triggering of a conspicuous signal for a vehicle on a path of travel. A sensor array detects environmental factors presenting a predetermined risk to the vehicle. A decision module assesses the environmental factors and the associated risks and determines if the conspicuousness signal is warranted and a type of signal to be made. An actuating module actuates the conspicuousness signal based on the determining of the decision module.

Abstract translation: 提供了系统和方法，用于在行进路径上的车辆的显着信号的实时动态触发。 传感器阵列检测向车辆呈现预定风险的环境因素。 决策模块评估环境因素和相关风险，并确定显着信号是否得到保证，并确定一种信号。 执行模块基于决定模块的确定来驱动显着信号。

公开(公告)号：US20140270353A1

公开(公告)日：2014-09-18

申请号：US13827222

申请日：2013-03-14

Applicant: XEROX CORPORATION

Inventor： Raja Bala ,  Zhigang Fan ,  Aaron Michael Burry ,  Jose Antonio Rodriguez-Serrano ,  Vishal Monga ,  Xuan Mo

IPC: G06K9/00 ,  G06K9/62

CPC classification number: G06K9/00771 ,  G06K9/6249

Abstract: Methods, systems, and processor-readable media for pruning a training dictionary for use in detecting anomalous events from surveillance video. Training samples can be received, which correspond to normal events. A dictionary can then be constructed, which includes two or more classes of normal events from the training samples. Sparse codes are then generated for selected training samples with respect to the dictionary derived from the two or more classes of normal events. The size of the dictionary can then be reduced by removing redundant dictionary columns from the dictionary via analysis of the sparse codes. The dictionary is then optimized to yield a low reconstruction error and a high-interclass discriminability.

Abstract translation: 用于修剪用于检测来自监视视频的异常事件的训练词典的方法，系统和处理器可读介质。 可以收到培训样本，对应于正常事件。 然后可以构建字典，其包括来自训练样本的两个或更多类的正常事件。 然后针对从两个或多个正常事件类派生的字典为选定的训练样本生成稀疏码。 然后可以通过对稀疏代码的分析从字典中删除冗余字典列来减少字典的大小。 然后对该字典进行优化，以产生低重构误差和高阶间的可辨别性。