公开(公告)号：WO2018231766A1

公开(公告)日：2018-12-20

申请号：PCT/US2018/037012

申请日：2018-06-12

申请人： VISTEON GLOBAL TECHNOLOGIES, INC.

发明人： BOWDEN, Upton, Beall ,  WHIKEHART, J., William ,  SCHUPFNER, Markus

IPC分类号： G10L25/51 ,  G10L25/30 ,  G06K9/00 ,  G06K9/46 ,  G06N3/04 ,  G06N3/08 ,  H04R1/40

CPC分类号： G10L25/51 ,  G06K9/00805 ,  G06K9/4628 ,  G06K9/6271 ,  G06N3/04 ,  G10L25/30 ,  H04R1/406 ,  H04R2499/13

摘要： Disclosed herein are systems, methods, and devices for optimally performing object identification employing a neural network (NN), for example a convolutional neural network (CNN). The aspects disclosed herein employ audio data captured by one or more microphones in to at least identify an object, or augment image capturing to perform the same. The audio data and the image data are each propagated to the NN, to perform object identification.

公开(公告)号：WO2018231480A1

公开(公告)日：2018-12-20

申请号：PCT/US2018/034259

申请日：2018-05-24

申请人： MICROSOFT TECHNOLOGY LICENSING, LLC

发明人： SCHLESINGER, Benny ,  WEINBERG, Shira

IPC分类号： G01S17/93 ,  G01S13/93 ,  G06K9/00 ,  G08G1/16 ,  G01S15/93 ,  G07C5/08 ,  G01C21/28

CPC分类号： G05D1/0214 ,  G01C21/28 ,  G06K9/00805 ,  G06K9/00979 ,  G06K9/22 ,  G06K9/6289 ,  G07C5/008 ,  G07C5/08 ,  G07C5/0841 ,  G08G1/0112 ,  G08G1/0133 ,  G08G1/0141 ,  G08G1/0967 ,  G08G1/096716 ,  G08G1/096758 ,  G08G1/096775 ,  G08G1/096791 ,  G08G1/164 ,  G08G1/165 ,  G08G1/166

摘要： Implementations described herein disclose a road hazard detection method including receiving a plurality of mobile device sensor signals from one or more sensors located on a mobile device within a vehicle, determining the road hazard encountered by the vehicle by analyzing the plurality of mobile device sensor signals, and reporting the existence of the road hazard to other users. In one implementation, the road hazard detection method also uses sensor signals from various sensors located on the vehicle and/or sensor signals from mobile devices of users on other vehicles to determine the road hazard.

公开(公告)号：WO2018203974A1

公开(公告)日：2018-11-08

申请号：PCT/US2018/022724

申请日：2018-03-15

申请人： QUALCOMM INCORPORATED

发明人： SWEET III, Charles Wheeler ,  SWART, Hugo

IPC分类号： G01S17/02 ,  G01S17/93

CPC分类号： G06K9/00805 ,  G01C19/02 ,  G01C21/005 ,  G01P3/66 ,  G01S7/40 ,  G01S7/497 ,  G01S7/52004 ,  G01S13/9303 ,  G01S15/93 ,  G01S17/023 ,  G01S17/42 ,  G01S17/933 ,  G01S19/51 ,  G05D1/0027 ,  G06Q10/04

摘要： Methods and apparatuses are disclosed for determining a characteristic of a device's object detection sensor oriented in a first direction. An example device may include one or more processors. The device may further include a memory coupled to the one or more processors, the memory including one or more instructions that when executed by the one or more processors cause the device to determine a direction of travel for the device, compare the direction of travel to the first direction to determine a magnitude of difference, and determine a characteristic of the object detection sensor based on the magnitude of difference.

公开(公告)号：WO2018162933A1

公开(公告)日：2018-09-13

申请号：PCT/GB2018/050622

申请日：2018-03-12

申请人： ARTIFICIAL INTELLIGENCE RESEARCH GROUP LIMITED ,  ARONSON, Bill

发明人： AVNI, Yossi ,  SUCHARD, Eytan

IPC分类号： G06K9/00

CPC分类号： G06K9/00805

摘要： According to an aspect of the present invention there is provided a method of identifying a potential obstacle to a moving vehicle, during a vehicle driving period. The method may comprise: receiving a first image of an environment in which the vehicle is located during the driving period from an image capture device, the environment comprising one or more objects physically located within the environment; receiving an optical signal from an optical distance measurement device, the optical signal having been reflected from one or more objects physically located within the environment, the optical signal comprising one or more optical signal data points; determining a distance value associated with one or more of the optical signal data points; selecting the optical signal data points associated with a distance value equal to or less than a predetermined threshold value; selecting a plurality of image pixels comprised within the received image associated with the selected optical signal data points; and identifying the one or more objects associated with the selected plurality of image pixels comprised within the received first image from an analysis of the selected plurality of image pixels. The first image of the environment may comprise a viewpoint of the environment with respect to a reference point. The first image may comprise a point-of-view image from the viewpoint of the vehicle, e.g. an image of the vehicle's field of view (FOV). The viewpoint of the environment may comprise at least a portion of the environment comprised within the vehicle's intended travel path. In this way the first image comprises objects that could potentially pose an obstacle to the vehicle based on its intended travel path.

公开(公告)号：WO2018161037A1

公开(公告)日：2018-09-07

申请号：PCT/US2018/020767

申请日：2018-03-02

申请人： KENNESAW STATE UNIVERSITY RESEARCH AND SERVICE FOUNDATION, INC.

发明人： YANG, Jidong

IPC分类号： G08G1/01 ,  G05D1/02 ,  G06K9/00 ,  G08G1/04 ,  G08G1/095 ,  G08G1/16

CPC分类号： G08G1/166 ,  G06K9/00785 ,  G06K9/00791 ,  G06K9/00805 ,  G08G1/0112 ,  G08G1/0116 ,  G08G1/0129 ,  G08G1/0133 ,  G08G1/0141 ,  G08G1/04 ,  G08G1/095 ,  G08G1/164

摘要： Systems and methods for quantitatively assessing collision risk of bodies, such as vehicles, and severity of a conflict between the bodies are disclosed. A method may comprise receiving image data associated with bodies. Based on the image data, an affinity (proneness) to collision of the bodies may be determined. Based on the determined affinity (proneness) to collision, a proximity (closeness) of collision of the bodies may be determined. Based on the determined proximity (closeness) of collision of the bodies, a collision risk of the bodies may be determined. The determined collision risk may be transmitted to a computing device, such as via a user interface. The determined collision risk may be used to control operations of a traffic control device or an autonomous vehicle.

公开(公告)号：WO2018070655A1

公开(公告)日：2018-04-19

申请号：PCT/KR2017/008862

申请日：2017-08-16

申请人： 주식회사 피엘케이 테크놀로지

发明人： 임상묵 ,  김진혁 ,  박광일

IPC分类号： B60W50/14 ,  B60W30/08 ,  B60W40/02 ,  B60W10/18 ,  G08B21/18

CPC分类号： G08G1/166 ,  B60Q5/006 ,  B60Q9/008 ,  B60T7/22 ,  B60T2201/022 ,  B60T2210/34 ,  G06K9/00805 ,  G06K9/4604 ,  G06K9/4642 ,  G06K9/6201 ,  G06K9/6212 ,  G06K9/6215 ,  G06K9/6269 ,  G08G1/167 ,  H04N7/181

摘要： 본 발명의 대형 차량의 이동 객체 충돌 경고 장치는, 대형 차량의 적어도 일측면 후방에 장착되며 상기 대형 차량의 전방을 지향하여 피촬상물을 촬상하는 후방 카메라 모듈; 상기 후방 카메라 모듈에서 촬상된 영상을 수신하는 영상 수신부; 상기 영상 수신부에서 수신된 영상을 인식하여 영상 내에 포함된 객체를 추출하고 추출된 객체가 자전거, 이륜차, 및 보행자를 포함하는 이동 객체인지를 결정하는 이동 객체 결정부; 상기 이동 객체 결정부에서 결정된 상기 이동 객체가 미리 설정된 위험 구역 내에 위치하는지를 판단하여 충돌 위험신호를 출력하는 이동 객체 충돌 가능성 결정부; 상기 대형 차량의 운전자에게 시각적 또는 청각적 경고신호를 출력하는 경고부; 및 상기 이동 객체 충돌 가능성 결정부에서 충돌 위험신호가 출력되면 상기 경고부를 동작시키도록 제어하는 제어부를 포함한다.

摘要翻译：

被安装用于通过引导重型车辆的前方对象成像的水的摄像后大型车辆后方相机模块中的至少一个侧表面上，根据本发明的一个大的车辆装置的运动物体碰撞警告; 图像接收单元，用于接收由后置相机模块捕获的图像; 图像接收移动物体确定单元通过提取包含在图像中的对象，并确定所提取的移动对象是否包括自行车，摩托车，行人和对象在接收到识别图像; 运动物体碰撞概率确定器，用于确定由运动物体确定器确定的运动物体是否在预定危险区域内并输出碰撞风险信号; 警报单元，用于向大型车辆的驾驶员输出视觉或听觉警告信号; 以及控制单元，用于当从运动物碰撞概率确定单元输出碰撞危险信号时控制警告单元进行操作。

公开(公告)号：WO2018049643A1

公开(公告)日：2018-03-22

申请号：PCT/CN2016/099187

申请日：2016-09-18

申请人： SZ DJI TECHNOLOGY CO., LTD.

发明人： ZHOU, You ,  TANG, Ketan ,  ZHAO, Cong ,  YAN, Jiaqi ,  DU, Jiexi

IPC分类号： G05D1/10 ,  G06T7/20

CPC分类号： G05D1/0246 ,  B64C39/024 ,  B64C2201/027 ,  B64C2201/127 ,  B64C2201/146 ,  G05D1/106 ,  G06K9/00637 ,  G06K9/00664 ,  G06K9/00805 ,  G06T7/20 ,  G06T7/215 ,  G06T7/269 ,  G06T7/70 ,  G06T7/73 ,  G06T2207/10016 ,  G06T2207/30261

摘要： Systems and methods are used to operate a movable object to avoid obstacles. A plurality of image frames are acquired. The plurality of images are captured within a predefined time window by an imaging device borne by the movable object moving along a navigation path. One or more objects adjacent the movable object are identified by measuring pixel movements within the plurality of image frames. Movements of the one or more objects relative to the movable object are estimated using dimensional variations of the one or more objects within the plurality of image frames. The navigation path of the movable object is adjusted in accordance with the estimated movements of the one or more objects.

摘要翻译：

系统和方法用于操作可移动物体以避开障碍物。 获取多个图像帧。 多个图像在预定义的时间窗口内被沿着导航路径移动的可移动物体承载的成像装置捕获。 通过测量多个图像帧内的像素移动来识​​别与可移动物体相邻的一个或多个物体。 使用多个图像帧内的一个或多个对象的尺寸变化来估计一个或多个对象相对于可移动对象的移动。 根据一个或多个物体的估计运动来调整可移动物体的导航路径。

公开(公告)号：WO2018047295A1

公开(公告)日：2018-03-15

申请号：PCT/JP2016/076623

申请日：2016-09-09

申请人： 三菱電機株式会社

发明人： 井上　悟 ,  原瀬　真一 ,  鈴木　涼太郎 ,  辻田　亘 ,  朝比奈　努 ,  澤　良次

IPC分类号： B60R21/00 ,  G01S15/93 ,  G08G1/16

CPC分类号： B60W30/06 ,  B60R21/00 ,  G01S15/93 ,  G01S15/931 ,  G01S2015/932 ,  G06K9/00805 ,  G08G1/16

摘要： 駐車支援装置（１００）は、車両（１）に設けられた左右一対のソナー（４Ｌ，４Ｒ）と、ソナー（４Ｌ，４Ｒ）が受信した反射波のレベル値を互いに異なる第１閾値（Ｌｔｈ１）及び第２閾値（Ｌｔｈ２）の各々と比較することにより、車両（１）に対する側方の判定対象距離範囲（ΔＤ）内に存在する障害物（Ｏ）が壁（Ｗ）であるか縁石（Ｃ）であるかを判定する障害物検知制御部（１０）と、障害物検知制御部（１０）による判定結果に応じて、自動駐車用の誘導経路を設定する自動駐車制御部（２０）とを備え、自動駐車制御部（２０）は、縦列駐車による自動駐車において、判定結果が縁石（Ｃ）を示している場合、判定結果が壁（Ｗ）を示している場合に対して車両（１）の切り込みが大きい誘導経路を設定する。

摘要翻译：

停车辅助装置（100）包括一车辆（1）对设置声纳的（4L，4R），声纳（4L，4R）由接收到的反射波的电平值的 通过用每个不同的第一阈值（Lth1）和第二阈值（Lth2），在车辆（1）（O）壁存在的横向（Δd）判断目标距离范围内的障碍物的相比较（ 自动至W）和是或路边（C）判断障碍物检测控制单元是否（10）根据由障碍物检测控制单元（10）的判定结果，设定了自动停车的引导路线 包括停车控制单元（20），自动停车控制器（20），在由平行停车时，自动停车时所述判断结果指示路边（C）中，确定结果指示所述壁（W） 车辆（1）中具有大切口的引导路线被设置为该情况。

公开(公告)号：WO2018009552A1

公开(公告)日：2018-01-11

申请号：PCT/US2017/040734

申请日：2017-07-05

申请人： NAUTO GLOBAL LIMITED

发明人： SATHYANARAYANA, Suchitra ,  SATAZODA, Ravi, Kumar

IPC分类号： G06F9/00 ,  G06F9/32 ,  G06F9/46 ,  G06T7/20

CPC分类号： G06K9/00791 ,  G06K9/00805 ,  G06K9/6223 ,  G06K9/6267

摘要： A method for image analysis, including recording an image sequence at a vehicle system mounted to a vehicle; automatically detecting an object within the image sequence with a detection module; automatically defining a bounding box about the detected object within each image of the image sequence; modifying the image sequence with the bounding boxes for the detected object to generate a modified image sequence; at a verification module associated with the detection module, labeling the modified image sequence as comprising one of a false positive, a false negative, a true positive, and a true negative detected object based on the bounding box within at least one image of the modified image sequence; training the detection module with the label for the modified image sequence; and automatically detecting objects within a second image sequence recorded with the vehicle system with the trained detection module.

摘要翻译： 一种用于图像分析的方法，包括在安装到车辆的车辆系统处记录图像序列; 利用检测模块自动检测图像序列内的物体; 在图像序列的每个图像内自动定义关于检测到的对象的边界框; 用检测对象的边界框修改图像序列以生成修改的图像序列; 在与检测模块相关联的验证模块处，基于修改的图像序列的至少一个图像内的边界框将修改的图像序列标记为包括假阳性，假阴性， 图像序列; 用修改的图像序列的标签对检测模块进行训练; 并用训练的检测模块自动检测车辆系统记录的第二图像序列内的物体。

公开(公告)号：WO2017164017A1

公开(公告)日：2017-09-28

申请号：PCT/JP2017/010285

申请日：2017-03-14

申请人： 株式会社デンソー

发明人： 高木　亮

IPC分类号： G08G1/16 ,  B60R21/00 ,  G06T1/00

CPC分类号： B60Q9/008 ,  B60R21/00 ,  G01S13/867 ,  G01S13/89 ,  G01S13/931 ,  G06K9/00805 ,  G06T1/00 ,  G08G1/16

摘要： ＥＣＵ１０は、撮像装置２１により撮像された画像に基づいて、自車の進行方向に存在する前方車を検知する。また、ＥＣＵ１０は、前記前方車が存在している場合に、前記画像と車両前部又は後部の辞書情報とに基づいて、前記前方車の前側又は後側の車両端部における前記自車の進行方向に対する横方向の大きさを端部横幅として算出する端部幅算出部と、前記自車の進行方向において前記前方車の側部が認識される状況か否かを判定する判定部と、前記前方車の側部が認識される状況であると判定された場合に、前記端部横幅を拡大側に補正して補正横幅を算出する横幅補正部と、を備える。

摘要翻译： ECU 10基于由图像捕获装置21捕获的图像来检测存在于车辆的行驶方向上的前方车辆。 此外，ECU 10，当在前车辆存在，所述图像和所述车辆前部或后部的字典信息，车辆的在前面或后面的前轮的车辆端部的进展的基础上 和端部宽度计算部，其计算的横向相对于所述方向的端部的宽度的范围内，并且其中所述用于确定是否其中在车辆的行进方向前方车辆侧被识别的情况的判定部，所述 以及横向宽度校正单元，其在确定前方车辆的侧部被识别时通过将端部横向宽度校正到放大侧来计算校正横向宽度。