公开(公告)号：US09111136B2

公开(公告)日：2015-08-18

申请号：US13677548

申请日：2012-11-15

Applicant: Xerox Corporation

Inventor： Peter Paul ,  Abu Islam

IPC: H04N7/18 ,  G06K9/00 ,  G08G1/017 ,  G07B15/06 ,  G07C5/00

CPC classification number: G06K9/00369 ,  G07B15/063 ,  G07C5/008 ,  G08G1/0175

Abstract: A multi-view imaging system for Vehicle Occupancy Detection (VOD) including a gantry mounted camera and illuminator to view the front seat of vehicles, and a roadside mounted camera and illuminator to view the rear seat of vehicles. The system controls the illuminator units to preserve/maximize bulb life, thus reducing the service cost of the system. In one embodiment, a target vehicle's license plate is read. If the vehicle is on a pre-approved list to use the HOV lane, then no further interrogation of the vehicle is performed. If the vehicle is not on the pre-approved list, then the front seats are interrogated by a camera and illuminator located on an overhead gantry as the vehicle continues down the highway. If the front seat analysis indicates that the passenger seat is not occupied, then the system interrogates the rear seats using a separate camera and illuminator located on the roadside.

Abstract translation: 一种用于车辆占用检测（VOD）的多视图成像系统，包括具有龙门架的照相机和照明器，用于观察车辆的前座椅，以及路边安装的照相机和照明器，以观察车辆的后座。 系统控制照明器单元以保持/最大化灯泡寿命，从而降低系统的服务成本。 在一个实施例中，读取目标车辆的车牌。 如果车辆在预先批准的列表上使用HOV车道，则不进行车辆的进一步询问。 如果车辆不在预先批准的列表上，则当车辆继续沿着高速公路行驶时，前排座椅被位于架空机架上的相机和照明器询问。 如果前排座椅分析表明乘客座椅未被占用，则系统使用位于路边的独立摄像机和照明器来询问后排座椅。

公开(公告)号：US20150070471A1

公开(公告)日：2015-03-12

申请号：US14022488

申请日：2013-09-10

Applicant: XEROX CORPORATION

Inventor： Robert P. Loce ,  Wencheng Wu ,  Thomas F. Wade ,  Daniel S. Hann ,  Peter Paul ,  Aaron M. Burry

IPC: H04N13/02 ,  G06Q20/38 ,  H04N7/18

CPC classification number: H04N13/239 ,  G06K9/00785 ,  G06Q20/145 ,  G06Q20/38 ,  G06T7/285 ,  G06T2207/30256 ,  H04N7/181

Abstract: Methods and devices acquire images using a stereo camera or camera network aimed at a first location. The first location comprises multiple parallel primary lanes merging into a reduced number of at least one secondary lane, and moving items within the primary lanes initiate transactions while in the primary lanes and complete the transactions while in the secondary lane. Such methods and devices calculate distances of the moving items from the camera to identify in which of the primary lanes each of the moving items was located before merging into the secondary lane. These methods and devices then order the transactions in a merge order corresponding to a sequence in which the moving items entered the secondary lane from the primary lanes. Also, the methods and devices output the transactions in the merge.

Abstract translation: 方法和设备使用瞄准在第一位置的立体相机或照相机网络获取图像。 第一位置包括合并到减少数量的至少一个次级车道中的多个并行主车道，并且主车道内的移动物品在主车道中启动交易，并且在次车道中完成交易。 这些方法和装置计算移动物品与照相机的距离，以识别在合并到次通道之前每个移动物品位于哪个主车道中。 这些方法和设备然后按照与从主车道进入辅助车道的移动物品的序列相对应的合并顺序来排序交易。 此外，方法和设备在合并中输出事务。

公开(公告)号：US08971579B2

公开(公告)日：2015-03-03

申请号：US13859047

申请日：2013-04-09

Applicant: Xerox Corporation

Inventor： Beilei Xu ,  Peter Paul

IPC: G06K9/00 ,  G06K9/36 ,  G06K9/52

CPC classification number: G06K9/00832 ,  G06K9/00785

Abstract: A system and method to capture an image of an oncoming target vehicle and localize the windshield of the target vehicle. Upon capturing an image, it is then analyzed to detect certain features of the target vehicle. Based on geometrical relationships of the detected features, the area of the image containing the windshield of the vehicle can then be identified and localized for downstream processing.

Abstract translation: 捕获迎面而来的目标车辆的图像并定位目标车辆的挡风玻璃的系统和方法。 在捕获图像时，然后对其进行分析以检测目标车辆的某些特征。 基于检测到的特征的几何关系，可以识别包含车辆挡风玻璃的图像的区域，以便进行下游处理。

公开(公告)号：US20140301606A1

公开(公告)日：2014-10-09

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: G06K9/18

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：US20140079315A1

公开(公告)日：2014-03-20

申请号：US13746412

申请日：2013-01-22

Applicant: XEROX CORPORATION

Inventor： Vladimir Kozitsky ,  Aaron Michael Burry ,  Peter Paul

IPC: G06K9/62

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：US20170243067A1

公开(公告)日：2017-08-24

申请号：US15049568

申请日：2016-02-22

Applicant: Xerox Corporation

Inventor： Orhan Bulan ,  Beilei Xu ,  Peter Paul

IPC: G06K9/00 ,  G06K9/62 ,  G06K9/46

CPC classification number: G06K9/00771 ,  G06K9/00785 ,  G06K9/00832 ,  G06K9/00838 ,  G06K9/4609 ,  G06K9/627 ,  G06K2209/23 ,  G06T2207/30248 ,  G06T2207/30268

Abstract: Methods, systems, and processor-readable media for detecting the side window of a vehicle. A spatial probability map can be calculated, which includes data indicative of likely side window locations of a vehicle in an image. A side window detector can be run with respect to the image of the vehicle to determine detection scores. The detection scores can be weighted based on the spatial probability map. A detected region of interest can be extracted from the image as extracted image patch. An image classification can then be performed with respect to the extracted patch to provide a classification that indicates whether or not a passenger is in the vehicle or no-passenger is in the vehicle.

公开(公告)号：US20170213463A1

公开(公告)日：2017-07-27

申请号：US15004832

申请日：2016-01-22

Applicant: Xerox Corporation

Inventor： Howard Mizes ,  Rakesh Suresh Kulkarni ,  Peter Paul

IPC: G08G1/14 ,  G06K9/00

Abstract: A method, non-transitory computer readable medium and apparatus for calculating a by spot occupancy of a parking lot are disclosed. For example, the method includes receiving an indication of a triggering event, sending a query to receive a first image and a second image in response to the triggering event, receiving the first image and the second image, analyzing the first image and the second image to determine a change in an occupancy status of a parking spot within the parking lot and calculating the by spot occupancy of the parking lot based on the change in the occupancy status of the parking spot.

公开(公告)号：US20170140253A1

公开(公告)日：2017-05-18

申请号：US15179403

申请日：2016-06-10

Applicant: Xerox Corporation

Inventor： Safwan Wshah ,  Beilei Xu ,  Orhan Bulan ,  Jayant Kumar ,  Peter Paul

IPC: G06K9/62 ,  G06K9/46 ,  G06N3/08

CPC classification number: G06N3/08 ,  G06K9/00771 ,  G06K9/00785 ,  G06K9/4628 ,  G06K9/629 ,  G06K2209/23 ,  G06N3/0454

Abstract: A method and system for domain adaptation based on multi-layer fusion in a convolutional neural network architecture for feature extraction and a two-step training and fine-tuning scheme. The architecture concatenates features extracted at different depths of the network to form a fully connected layer before the classification step. First, the network is trained with a large set of images from a source domain as a feature extractor. Second, for each new domain (including the source domain), the classification step is fine-tuned with images collected from the corresponding site. The features from different depths are concatenated with and fine-tuned with weights adjusted for a specific task. The architecture is used for classifying high occupancy vehicle images.

公开(公告)号：US09524486B2

公开(公告)日：2016-12-20

申请号：US14637830

申请日：2015-03-04

Applicant: Xerox Corporation

Inventor： Wencheng Wu ,  Dennis L. Venable ,  Thomas F. Wade ,  Peter Paul ,  Adrien P. Cote

IPC: G06F19/00 ,  G06Q10/08 ,  G06K9/46 ,  G06T11/60

CPC classification number: G06Q10/087 ,  G06T11/60

Abstract: A system to detect and maintain retail store promotional price tags (PPTs) includes a heuristic PPT description extractor module, a heuristic rule deriver module, a store shelf image acquisition system, a barcode locator and recognizer module, and a heuristic PPT classifier module. The heuristic PPT description extractor module extracts heuristic descriptions of PPTs. The heuristic rule deriver module derives a set of heuristic parameters for the PPTs. The barcode locator and recognizer module analyzes images acquired by the store shelf image acquisition system to localize and recognize barcodes. The heuristic PPT description classifier module extracts heuristic attributes from the images acquired by the store shelf image acquisition system using the set of PPT parameters supplied by the heuristic rule deriver module.

Abstract translation: 检测和维护零售店促销价格标签（PPT）的系统包括启发式PPT描述提取器模块，启发式规则提升模块，商店架图像采集系统，条形码定位器和识别器模块以及启发式PPT分类器模块。 启发式PPT描述提取器模块提取PPT的启发式描述。 启发式规则引导模块为PPT引入了一组启发式参数。 条形码定位器和识别器模块分析由商店货架图像采集系统获取的图像以定位和识别条形码。 启发式PPT描述分类器模块使用由启发式规则提升模块提供的一组PPT参数从存储架图像采集系统获取的图像中提取启发式属性。

公开(公告)号：US09460369B2

公开(公告)日：2016-10-04

申请号：US14514627

申请日：2014-10-15

Applicant: XEROX CORPORATION

Inventor： Steven R. Moore ,  Wencheng Wu ,  Dennis L. Venable ,  Thomas F. Wade ,  Peter Paul

IPC: G06K15/00 ,  G06K9/78 ,  H04N1/00 ,  H04N1/21 ,  G06K9/00 ,  H04N5/225 ,  G06Q30/02 ,  H04N101/00

CPC classification number: G06K9/78 ,  G06K9/00577 ,  G06K9/00664 ,  G06Q30/0276 ,  H04N1/00278 ,  H04N1/00562 ,  H04N1/2154 ,  H04N5/2251 ,  H04N2101/00 ,  H04N2201/0084

Abstract: The present disclosure relates to systems and methods for use in a retail store. An example system includes a mobile base, a printer, an image capture subsystem on the mobile base and coupled to the printer, the image capture system including at least one image capture device and at least one image processor, the image capture device configured to obtain images of items in the retail store, the image processor configured to derive item identification data from the images of items, and a control subsystem coupled to the printer and to the image capture subsystem, where the control subsystem is configured to receive information identifying items requiring signage, acquire item identification data from the image capture subsystem, determine, based on the information identifying items requiring signage and on the item identification data, items requiring signage, and to direct the printer to print signage for the items requiring signage.

Abstract translation: 本公开涉及用于零售商店的系统和方法。 示例性系统包括移动基座，打印机，移动基座上的图像捕捉子系统，并且耦合到打印机，所述图像捕获系统包括至少一个图像捕获装置和至少一个图像处理器，所述图像捕获装置被配置成获得 所述图像处理器被配置为从所述物品的图像导出物品识别数据，以及耦合到所述打印机和所述图像捕获子系统的控制子系统，其中所述控制子系统被配置为接收识别需要的物品的信息 标识，从图像捕获子系统获取物品识别数据，基于识别需要标示的物品的信息和物品识别数据确定需要标示的物品，并指示打印机打印需要标牌的物品的标牌。