公开(公告)号：US09865056B2

公开(公告)日：2018-01-09

申请号：US15015768

申请日：2016-02-04

Applicant: Xerox Corporation

Inventor： Edgar A. Bernal ,  Aaron Michael Burry ,  Orhan Bulan

IPC: G06K9/00 ,  G06T7/00 ,  G06Q10/00 ,  G08G1/015 ,  G08G1/017 ,  G08G1/04 ,  G08G1/065 ,  G08G1/095 ,  G07C11/00 ,  G08G1/02 ,  G08G1/042 ,  G08G1/0967

CPC classification number: G06T7/55 ,  G06K9/00785 ,  G06Q10/00 ,  G06Q10/04 ,  G06T2207/30236 ,  G07C11/00 ,  G08G1/015 ,  G08G1/0175 ,  G08G1/02 ,  G08G1/04 ,  G08G1/042 ,  G08G1/065 ,  G08G1/095 ,  G08G1/096716 ,  G08G1/096783

Abstract: A method, non-transitory computer readable medium, and apparatus for directing a vehicle in a side-by-side drive-thru are disclosed. For example, the method receives one or more video images of a side-by-side drive-thru comprising two or more lanes, detects a vehicle approaching an entrance of the side-by-side drive-thru, calculating an estimated order time for the vehicle and directs the vehicle to one of the two or more lanes based on the estimated order time for the vehicle or a previously estimated order time of each one of the a plurality of vehicles already in the first lane and the second lane of the drive-thru.

公开(公告)号：US09378556B2

公开(公告)日：2016-06-28

申请号：US14261510

申请日：2014-04-25

Applicant: Xerox Corporation

Inventor： Aaron Michael Burry

IPC: G06T7/20

CPC classification number: G06T7/2006 ,  G06T7/246 ,  G06T2207/10016 ,  G06T2207/20081 ,  G06T2207/30236

Abstract: Video-based object tracking accuracy is improved by a tentative identification of objects to be tracked. An identified blob that does not encompass a previously established set of tracking features (“tracker”) triggers initialization of an infant tracker. If that tracker remains the only tracker or becomes the “oldest” tracker associated with a blob identified in subsequent video frames, the “age” of the tracker is increased. If, in subsequent frames, the tracker is encompassed by a blob that is associated with an “older” tracker, the “age” of the tracker is decreased. Infant trackers that reach or exceed a threshold “age” are promoted to adult status. Adult trackers can be processed as being associated with valid objects. Trackers established for blobs identified due to mask segmentation tend not to cause false object detections. When segmentation is corrected, blob segments are combined and redundant trackers for the associated object are demoted and ignored.

Abstract translation: 通过暂时识别要跟踪的对象来改善基于视频的对象跟踪精度。 不包含先前建立的一组跟踪特征（“跟踪器”）的识别的Blob触发婴儿跟踪器的初始化。 如果该跟踪器仍然是唯一的跟踪器，或者成为与随后视频帧中识别的斑点相关联的“最旧”跟踪器，则跟踪器的“年龄”增加。 如果在随后的帧中，跟踪器被与“较旧”跟踪器相关联的斑点包围，则跟踪器的“年龄”减小。 达到或超过阈值“年龄”的婴儿跟踪器被提升为成年人身份。 成人跟踪器可以被处理为与有效对象相关联。 由于掩模分割而识别的斑点建立的跟踪器往往不会导致伪对象检测。 当分段被纠正时，组合段和相关对象的冗余跟踪器被降级和忽略。

公开(公告)号：US09224058B2

公开(公告)日：2015-12-29

申请号：US13857657

申请日：2013-04-05

Applicant: Xerox Corporation

Inventor： Peter Paul ,  Vladimir Kozitsky ,  Aaron Michael Burry ,  Martin Edward Hoover ,  Mark Cantelli

IPC: G06K9/00 ,  G06K9/20 ,  G06K9/32

CPC classification number: G06K9/2027 ,  G06K9/3258 ,  G06K2209/15

Abstract: Methods, systems, and processor-readable media for data augmentation utilized in an automatic license plate recognition engine. A machine-readable code can be associated with an automatic license plate recognition engine. The machine-readable code can be configured to define parameters that drive processing within the automatic license plate recognition engine to produce recognition results thereof and enhance a machine readability of a license plate recognized and analyzed via the automatic license plate recognition engine.

Abstract translation: 用于在自动车牌识别引擎中使用的用于数据增加的方法，系统和处理器可读介质。 机器可读代码可以与自动车牌识别引擎相关联。 机器可读代码可以被配置为定义驱动自动车牌识别引擎内的处理以产生其识别结果的参数，并提高通过自动车牌识别引擎识别和分析的牌照的机器可读性。

公开(公告)号：US09195908B2

公开(公告)日：2015-11-24

申请号：US13899705

申请日：2013-05-22

Applicant: Xerox Corporation

Inventor： Stuart Schweid ,  Aaron Michael Burry ,  Vladimir Kozitsky

IPC: G06K9/62

CPC classification number: G06K9/623

Abstract: Methods, systems and processor-readable media for determining, post training, which locations of a classifier window are most significant in discriminating between class and non-class objects. The important locations can be determined by calculating the mean and standard deviation of every pixel location in the classifier context for both the positive and negative samples of the classifier. Using a combination of t-scores and mean differences, the importance of all pixel locations in the classifier score can be rank ordered. A sufficient number of pixel locations can then be selected to achieve a detection rate close enough to the full classifier for a particular application.

Abstract translation: 用于确定后期训练的方法，系统和处理器可读介质，分类器窗口的哪些位置在区分类和非类对象之间最显着。 可以通过计算分类器的正和负样本的分类器上下文中每个像素位置的平均值和标准偏差来确定重要位置。 使用t分和平均差的组合，分类器得分中所有像素位置的重要性可以是排序的。 然后可以选择足够数量的像素位置，以实现足够接近特定应用的全分类器的检测率。

公开(公告)号：US09098749B2

公开(公告)日：2015-08-04

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。

公开(公告)号：US20140169633A1

公开(公告)日：2014-06-19

申请号：US13715041

申请日：2012-12-14

Applicant: Xerox Corporation

Inventor： Richard W. Seyfried ,  Aaron Michael Burry ,  Vladimir Kozitsky

IPC: G06K9/00

CPC classification number: G06K9/3258 ,  G06K9/00791 ,  G06K2209/15

Abstract: Disclosed are methods and systems for monitoring and reporting road violations of vehicles sharing roads with responding emergency vehicles. According to an exemplary method video is captured from a forward and/or rear facing camera mounted to an emergency vehicle, and the video is processed to identify any vehicles in violation within a prescribed distance from the emergency vehicle. A license plate id of a vehicle determined to be in violation is identified and communicated to the appropriate authorities.

Abstract translation: 公开了监测和报告与应急车辆共享道路的车辆的道路违规的方法和系统。 根据示例性方法，从安装到紧急车辆的前方和/或后方相机捕获视频，并且处理视频以识别违规车辆在距紧急车辆规定距离内的任何车辆。 确定违章的车辆的车牌号被确定并传达给有关当局。

公开(公告)号：US09213910B2

公开(公告)日：2015-12-15

申请号：US14159590

申请日：2014-01-21

Applicant: Xerox Corporation

Inventor： Aaron Michael Burry ,  Farnaz Abtahi

IPC: G06K9/32 ,  G06K9/62

CPC classification number: G06K9/325 ,  G06K9/00791 ,  G06K9/34 ,  G06K9/6256 ,  G06K2209/15

Abstract: Methods and systems for achieving accurate segmentation of characters with respect to a license plate image utilizing a reinforcement learning approach. A vehicle image can be captured by an image capturing unit and processed utilizing an ALPR (Automatic License Plate Recognition) unit. The reinforcement learning (RL) approach can be configured to initialize a segmentation agent with a starting location. A proper segmentation path (cuts) from top to bottom and from a darker to lighter area in a cropped license plate image can be identified by the segmentation agent during a training phase. Rewards can be provided based on a number of good and bad moves. The association between a current state and a sensory input with a preferred action can be learned by the segmentation agent at the end of the training phase.

Abstract translation: 使用强化学习方法实现与车牌图像相关的字符的精确分割的方法和系统。 车辆图像可以由图像捕获单元捕获并且使用ALPR（自动车牌识别）单元进行处理。 强化学习（RL）方法可以配置为初始化具有起始位置的分段代理。 可以在训练阶段通过分割代理来识别裁剪后的车牌图像中从顶部到底部以及从较暗到更轻的区域的适当分割路径（切割）。 奖励可以根据一些好的和坏的举动提供。 当前状态与具有优选动作的感觉输入之间的关联可以在训练阶段结束时被分段代理学习。

公开(公告)号：US09082037B2

公开(公告)日：2015-07-14

申请号：US13899742

申请日：2013-05-22

Applicant: Xerox Corporation

Inventor： Aaron Michael Burry ,  Vladimir Kozitsky

IPC: G06K9/00 ,  G06K9/32

CPC classification number: G06K9/325 ,  G06K2209/15

Abstract: Methods and systems for automatically determining the issuing state of a license plate. An image of a license plate acquired by an ALPR engine can be processed via one or more OCR engines such that each OCR engine among the OCR engines is tuned to a particular state. Confidence data output from the OCR engine(s) can be analyzed (among other factors) to estimate the issuing state associated with the license plate. Multiple observations related to the issuing state can be merged to derive an overall conclusion and assign an associated confidence value with respect to the confidence data and determine a likely issuing state associated with the license plate.

Abstract translation: 自动确定车牌发卡状态的方法和系统。 可以通过一个或多个OCR引擎来处理由ALPR引擎获取的车牌的图像，使得OCR引擎中的每个OCR引擎被调谐到特定状态。 可以分析从OCR引擎输出的置信度数据（以及其他因素）来估计与车牌相关的发卡状态。 可以合并与发布状态有关的多个观察结果，以获得总体结论，并且相对于置信度数据分配相关的置信度值，并确定与该牌照相关联的可能发布状态。

公开(公告)号：US09025865B2

公开(公告)日：2015-05-05

申请号：US13746412

申请日：2013-01-22

Applicant: Xerox Corporation

Inventor： Vladimir Kozitsky ,  Aaron Michael Burry ,  Peter Paul

IPC: G06K9/62 ,  A47C7/14 ,  B60N2/44 ,  B60N2/02 ,  B60N2/18

CPC classification number: A47C7/14 ,  B60N2/02 ,  B60N2/1803 ,  B60N2/1835 ,  B60N2/1864 ,  B60N2002/0236 ,  G06K9/3258 ,  G06K9/623 ,  G06K9/6267 ,  G06K2009/4666 ,  G06K2209/01 ,  G06K2209/15

Abstract: Methods and systems for reducing the required footprint of SNoW-based classifiers via optimization of classifier features. A compression technique involves two training cycles. The first cycle proceeds normally and the classifier weights from this cycle are used to rank the Successive Mean Quantization Transform (SMQT) features using several criteria. The top N (out of 512 features) are then chosen and the training cycle is repeated using only the top N features. It has been found that OCR accuracy is maintained using only 60 out of 512 features leading to an 88% reduction in RAM utilization at runtime. This coupled with a packing of the weights from doubles to single byte integers added a further 8× reduction in RAM footprint or a reduction of 68× over the baseline SNoW method.

Abstract translation: 通过优化分类器功能来减少基于SNoW的分类器所需占用面积的方法和系统。 压缩技术涉及两个训练周期。 第一个周期正常进行，并且使用来自该周期的分类器权重使用几个标准对连续平均量化变换（SMQT）特征进行排序。 然后选择顶部N（512个特征之​​一），并且仅使用前N个特征重复训练周期。 已经发现，使用仅512个特征中的60个来维持OCR精度，导致运行时RAM利用率降低88％。 这加上权重从双精度到单字节整数的加载，增加了RAM占用的8倍减少或基线SNoW方法减少了68倍。

公开(公告)号：US20140355836A1

公开(公告)日：2014-12-04

申请号：US13904559

申请日：2013-05-29

Applicant: Xerox Corporation

Inventor： Vladimir Kozitsky ,  Aaron Michael Burry

IPC: G06K9/32

CPC classification number: G06K9/3258 ,  G06K9/685 ,  G06K2209/15

Abstract: Methods and systems for improving automated license plate recognition performance. One or more images of a vehicle can be captured via an automated license plate recognition engine. Vehicle class information associated with the vehicle can be obtained using the automated license place recognition engine. Such vehicle class information can be analyzed with respect to the vehicle. Finally, data can be dynamically adjusted with respect to the vehicle based on a per image basis to enhance recognition of the vehicle via the automated license plate recognition engine.

Abstract translation: 提高自动牌照识别性能的方法和系统。 可以通过自动车牌识别引擎捕获车辆的一个或多个图像。 与车辆相关的车辆类别信息可以使用自动许可证位置识别引擎获得。 可以相对于车辆分析这样的车辆类别信息。 最后，基于每个图像可以相对于车辆动态地调整数据，以通过自动车牌识别引擎增强对车辆的识别。