Abstract: A navigation system for one or more vehicles located within a control volume includes: a first camera configured to observe the one or more vehicles and first, second and third reference points disposed within the control volume and having first, second and third known spatial positions, respectively, the first camera being further configured to produce a first output signal from observation of the observed vehicles and reference points; a second camera configured to observe the one or more vehicles and the first, second and third reference points, the second camera being further configured to produce a second output signal from observation of the observed vehicles and reference points; and a processor operatively connected with the first and second cameras and configured to determine a respective spatial position for each of the one or more vehicles from the first and second output signals.

Abstract: A map scanning method including the following step: scanning, by a scanning device, a location area of a vehicle to obtain location data, wherein the scanning device includes a lidar sensor, and the lidar sensor is configured to scan the location area to obtain a scan point cloud data of the location data; comparing, by a processor, the location data with a basic map to obtain a first point cloud area corresponding to the location data in the basic map; Comparing the scan point cloud data with the first point cloud area to calculate a matching degree value; and when the matching degree value is less than an update threshold, updating the first point cloud area according to the scan point cloud data.

Abstract: Systems and methods for updating an electronic map of a facility are disclosed. The electronic map includes a set of map nodes. Each map node has a stored image data associated with a position within the facility. The method includes collecting image data at a current position of a self-driving material-transport vehicle; searching the electronic map for at least one of a map node associated with the current position and one or more neighboring map nodes within a neighbor threshold to the current position; comparing the collected image data with the stored image data of the at least one of the map node and the one or more neighboring map nodes to determine a dissimilarity level. The electronic map may be updated based at least on the collected image data and the dissimilarity level. The image data represents one or more features observable from the current position.

Abstract: A calibration system for a camera and an inertial sensor, respectively rigidly affixed to a vehicle, includes a controller having a processor and tangible, non-transitory memory on which instructions are recorded. The camera is adapted to capture one or more images and the inertial sensor is adapted to obtain motion data, including an angular velocity vector of the vehicle. The controller is configured to receive respective data from the camera and the inertial sensor. The controller is configured to perform adaptive feature selection on the one or more images, including extracting one or more regions of interest respectively from the images and calculating a respective optimal quality level for each of the regions of interest. The camera is aligned to the inertial sensor based in part on the adaptive feature selection and the respective data from the inertial sensor.

Abstract: A controller controls a spacecraft to rendezvous a non-center-of-mass point of the controlled spacecraft with a non-center-of-mass point of an uncontrolled celestial body. The controlled spacecraft and the uncontrolled celestial body form a multi-object celestial system, and the controller produces control commands to thrusters of the controlled spacecraft using a non-linear model predictive control (NMPC) optimizing a cost function over a receding horizon that minimizes an error between coordinates of the non-center-of-mass point of the spacecraft and the non-center-of-mass point of the celestial body subject to joint dynamics of the multi-object celestial system coupled with joint kinematics of the multi-object celestial system.

Abstract: An image processing apparatus includes a representative-distance calculator and a joining processor. The representative-distance calculator generates representative distance values of the basis of a distance image generated from a stereo image including an image of at least one object including a vehicle. The distance image includes distance values of pixels. The joining processor performs a joining process of joining a first image region and a second image region that are defined based on the image of the at least one object in the stereo image, performs a determining process of determining whether a temporarily joined region in which the first image region and the second image region are temporarily joined includes an image of a corner portion of the vehicle, and refrains from performing the joining process when the temporarily joined region includes the image of the corner portion.

Abstract: Method, system for presenting an object on a computing device. A metaphor application on a computing device organizes a user interface based upon a metaphor. The metaphor organizes a document, file, application, or combination thereof based on geospheric direction, geolocation, or both. The metaphor may also organize a document, file, application, data, or a combination thereof based on a solid geometrical figure in three-dimensional Euclidean space. A document, file, application, or any combination thereof may be associated with geophysical direction, a geolocation, or both. The document, file, application, data, or any combination thereof may further be associated with a solid geometrical figure. A presentation object containing data on the document, file, application, data, or combination thereof, and the geospheric direction, geolocation, or both is formatted into data blocks for rendering on a display. The display may be the display screen of the computing device.

Abstract: Disclosed herein are various methods of calibrating relative orientation between spatial orientation sensors, relative orientation between imaging sensor and spatial orientation sensor, and relative orientation between imaging sensors. The relative orientation of the spatial orientation sensors, imaging sensors is calculated with respect to a reference frame.

Abstract: A method of calibrating an alignment model between a first sensor that uses a radar and a second sensor that determines a position of a sensor, and an electronic device performing the method. A radar map for surroundings of the first sensor and the second sensor is generated based on the first sensor and the second sensor, and at least one object located around the first sensor and the second sensor is detected based on pre-generated data. Also, target data corresponding to the object is determined based on the radar map, an error of the target data is calculated based on a shape of the object and the target data, and the alignment model between the first sensor and the second sensor is calibrated based on the error.

Abstract: A Wireless Avionics Intra-Communication (WAIC) system includes a communication network and a WAIC adapter location system. The communication network is configured to exchange data with a portable wireless device via WAIC adapter. The WAIC adapter location system is configured to determine a location of the WAIC adapter, which selectively connects and disconnects the portable device from the communication network based on the location.

Abstract: A communicative lighting system may comprise a plurality of network connected light sources. Each light source may possess a unique identifier permitting light sources to be associated with one another. An input received by one light source may alter the light output of that light source as well as other light sources associated with the light source. The light output of a light source may be altered by changing color, brightening or dimming, blinking, adjusting which of a plurality of light emitting diodes are activated, or otherwise modifying the type or amount of light output.

Abstract: A method for determining a protection radius of a vision-based navigation system includes at least one image sensor able to produce at least one first image at a first instant and one second image at a second instant; and a processing unit coupled to each image sensor and able to implement a processing method able to determine at least one value of motion between the first instant and the second instant on the basis of the first image, of the second image and of a plurality of image data determined on the at least first and second images; the method comprising the following steps: a first step of determining at least one nominal value of uncertainty of the at least one motion value obtained by the processing method for a set comprising the plurality of image data determined; a second step of determining i-th sub-values of uncertainty of the at least one motion value determined by the processing method for an i-th subset defined by removing an i-th part of the image data determined; a third step of determining

Abstract: A method for lane marker recognition includes: providing a filter bank with a plurality of different pairs of filters adapted to detect the left edge and the right edge of a specific type of a lane marker, respectively; receiving an image of a road; dividing the image of a road into a plurality of image segments, wherein each image segment includes at least one row of pixels of the image of a road; and for each of the image segments: applying a plurality of the pairs of filters of the filter bank to the image segment to generate a plurality of filter outputs; and determining which of the filter outputs correspond to a lane marker by using geo-metric information and appearance based information, wherein the geometric in-formation describes allowable dimensions of a determined lane marker, and the appearance based information describes allowable pixel values of a determined lane marker.

Abstract: Disclosed are systems, methods, and non-transitory computer-readable media for adjusting depth of AR content on HUD. A viewing device identifies, based on sensor data, a physical object visible through a transparent display of the vehicle. The sensor data indicates an initial distance of the physical object from the vehicle. The viewing device gathers virtual content corresponding to the physical object and generates an initial presentation of the virtual content based on the initial distance. The viewing device presents the initial presentation of the virtual content on the transparent display at a position on the transparent display corresponding to the physical object. The viewing device determines, based on updated sensor data, an updated distance of the physical object and generates an updated presentation of the virtual content based on the updated distance. The viewing device presents the updated presentation of the virtual content on the transparent display of the vehicle.

Abstract: Systems and methods are provided for transmitting functional safety statistics within a system. A video source produces a video data stream. A functional safety system driver accumulates functional safety statistics from at least one system and writes the functional safety statistics onto an associated system memory. A display sub-system driver writes a frame of the video data stream to the system memory. The display sub-system driver formats the functional safety statistics as video data and appends the functional safety statistics to a portion of the frame of video that is reserved for the functional safety statistics. A display sub-system transmits the frame of the video data stream to a host processor, which extracts the functional safety statistics from the video frame.

Abstract: Method, system for presenting an object on a computing device. A metaphor application on a computing device organizes a user interface based upon a metaphor. The metaphor organizes a document, file, application, or combination thereof based on geospheric direction, geolocation, or both. The metaphor may also organize a document, file, application, data, or a combination thereof based on a solid geometrical figure in three-dimensional Euclidean space. A document, file, application, or any combination thereof may be associated with geophysical direction, a geolocation, or both. The document, file, application, data, or any combination thereof may further be associated with a solid geometrical figure. A presentation object containing data on the document, file, application, data, or combination thereof, and the geospheric direction, geolocation, or both is formatted into data blocks for rendering on a display. The display may be the display screen of the computing device.

Abstract: A managing apparatus provides self-driving of a vehicle, which allows the vehicle to enter a parking space out of a plurality of parking spaces from an outside of a parking lot including the plurality of parking spaces, and which allows the vehicle to leave the parking lot for a predetermined place in the outside. The managing apparatus is provided with: a wait time setting device configured to set a maximum wait time of the vehicle in the predetermined place; and a controller configured to set a driving route for returning the vehicle to the parking lot from the predetermined place on condition that the maximum wait time elapses from an arrival of the vehicle in the predetermined place, and configured to provide the self-driving of the vehicle along the driving route.

Abstract: A method is provided for determining angular relationships from a point of interest to a plurality of peripheral points of interest on a map. One or more cost functions from the point of interest to the plurality of the peripheral points of interest on the map are analyzed. A plurality of vectors emanating from the point of interest to the plurality of the peripheral points of interest on a different representation of the map are displayed. Another method is provided for identifying points of interest on a map. Regions of high interest are identified on the map. Regions of low interest are identified on the map. The regions of high interest are expanded on a different representation of the map. The regions of low interest are contracted by an amount proportional to an amount the regions of high interest are expanded on the different representation of the map.

Abstract: Frame sequences from multiple image sensors may be combined in order to form, for example, an interleaved frame sequence. Individual frames of the combined sequence may be configured a by combination (e.g., concatenation) of frames from one or more source sequences. The interleaved/concatenated frame sequence may be encoded using a motion estimation encoder. Output of the video encoder may be processed (e.g., parsed) in order to extract motion information present in the encoded video. The motion information may be utilized in order to determine a depth of visual scene, such as by using binocular disparity between two or more images by an adaptive controller in order to detect one or more objects salient to a given task. In one variant, depth information is utilized during control and operation of mobile robotic devices.

Abstract: In a system, first and second detection devices are provided in mutually different places in a real space and each configured to image the real space and obtain three-dimensional information including data of a distance to a predetermined point on an object provided in the real space and direction information as to a direction of a predetermined part of the object. The system generates information related to relative positions and installation directions of the first and second detection devices on the basis of: the data of the distance to the predetermined point on the object and the direction information regarding the object, the data of the distance and the direction information being obtained by the first detection device; and the data of the distance to the predetermined point on the object and the direction information regarding the object, the data of the distance and the direction information being obtained by the second detection device.

Abstract: One embodiment is directed to a distributed antenna system (DAS) configured to cause a remote antenna unit deployed near a railroad track to operate in a low-power and/or muted operational state. While the remote antenna unit is operating in the low-power and/or muted operational state, the DAS determines if a train is sufficiently close to a coverage area of the remote antenna unit to trigger a change in an operational state of the remote antenna unit. In response to such a determination, the DAS causes the remote antenna unit to operate in a normal operational state. While the remote antenna unit is operating in the normal operational state, the DAS determines if the train has exited the coverage area of the remote antenna unit. In response to such a determination, the DAS causes the remote antenna unit to operate in the low-power and/or muted operational state.

Abstract: Provided is an embodiment to automatically detect signals in track-bound traffic when track-bound vehicles are traveling on track sections in a track network.

Abstract: The method of the invention comprises: obtaining a sequence of at least two images, with different levels of illumination; extracting the region containing the sign in the image; calculating the luminance values of the signs; and obtaining the difference in luminance of the sign corresponding to the two levels of illumination. The value obtained is the luminance of the sign (11) corresponding to an illumination equal to the difference between the illuminations, or additional illumination. This result is based on the additive property of luminance, according to which the luminance of a sign is the sum of the luminance produced by each source of illumination. A basic illumination device (5), an additional illumination device (7), at least one camera for taking images, and image recording, positioning and synchronism systems are required to implement the method.

Abstract: An abnormality behind a shielding object is discovered. There is provided a monitoring system including a video acquirer, a detector, and a notifier. The video acquirer of the monitoring system acquires a video. The detector of the monitoring system detects entering of a target object into a blind spot generated by a shielding object in the video and appearance of the target object from the blind spot. If the target object does not appear from the blind spot even after a first time elapses since entering of the target object into the blind spot, the notifier of the monitoring system makes a notification.

Abstract: To provide navigation to a destination to a user even in a state in which acquisition of a current position via a positioning module is difficult. There is provided an image processing device including: an image acquisition unit configured to acquire an image onto which a real object including a visual recognition target is projected; a recognition unit configured to recognize relative arrangement of a terminal that has captured the image to the real object by executing image recognition of the recognition target; and a control unit configured to display a display object indicating a direction toward a destination of a user on a screen of the terminal on the basis of the relative arrangement of the terminal recognized by the recognition unit.

Abstract: The present invention is directed to providing an around view monitoring system (AVMS) that may enhance accuracy of an estimation of a camera attitude while a road gradient is present, and an operating method thereof. According to an embodiment of the present invention, the AVMS may calculate a rotation matrix for removing a road gradient component using a camera that has not changed position and generate an estimated rotation matrix from which the road gradient is removed by applying the rotation matrix to an estimated rotation matrix for a camera that has changed position.

Abstract: An autonomous land vehicle in accordance with aspects of the present disclosure includes a land vehicle conveyance system, at least two telecommunication devices, an imaging device configured to capture image data of a surrounding environment, a video encoder configured to encode the image data, one or more processors, and at least one memory storing instructions. The telecommunication devices can perform wireless communication independently of each other and can simultaneously perform wireless communication.

Abstract: A method (100) for generating a geographic coordinate and a device (500) for generating a geographic coordinate. The method (100) comprises: enabling a device to point to a position point (102); obtaining data including a geographic coordinate of the device, a relative height between the device and the position point and pointing information of the device (104); and generating a geographic coordinate of the position point based on the data (106). The device (500) for generating a geographic coordinate comprises a pointing module (502), a data obtaining module (504) and a generating module (506). By enabling the device to point to a desired position point, a geographic coordinate of the desired position point can be obtained rapidly.

Abstract: A communicative lighting system may comprise a plurality of network connected light sources. Each light source may possess a unique identifier permitting light sources to be associated with one another. An input received by one light source may alter the light output of that light source as well as other light sources associated with the light source. The light output of a light source may be altered by changing color, brightening or dimming, blinking, adjusting which of a plurality of light emitting diodes are activated, or otherwise modifying the type or amount of light output.

Abstract: A method and system are disclosed for assisting the driver of a motor vehicle while driving through an intersection or road junction where at least three roads converge in a transition area. A future motion of an own or ego vehicle is estimated, and a future motion of a second vehicle is estimated. The probability of a collision between the own vehicle and the second vehicle is estimated based on the respective future motions. A search for a roadway marking arranged in the transition area is executed. If a roadway marking is found, the future motion of at least one of the own vehicle and second vehicle is estimated based on the detected roadway marking. A risk-reducing manuever is executed when the estimated collision probability exceeds a limiting value.

Abstract: A diagnostic method is provided for a vision sensor of a vehicle. The diagnostic method includes the steps of: determining a first position and a direction of movement of the vehicle; selecting a stationary landmark suitable for being detected by the vision sensor from a landmark database, wherein the landmark is located in front of the vehicle in the direction of movement of the vehicle, wherein a distance from the first position of the vehicle to the landmark is greater than the target range of the vision sensor; detecting the landmark via the vision sensor and, in response to a successful detection of the landmark by the vision sensor; determining a second position of the vehicle; determining an actual range of the vision sensor as the distance between the second position of the vehicle and the landmark.

Abstract: A system and methods for muddy water detection using normalized semantic layers, wherein a spectrum analyzer isolates spectrum bands within an image to produce a set of three normalized differential index images from which a composite color image is created, from which a power band is computed, from which a two-color image is produced, and then filters image components within the two-color representation based on defined criteria.

Abstract: The present invention provides a connected device system, comprising a master network device (2) and a device (5) to be integrated into the connected device system (1). The connected device system is adapted to connect the master network device and the device via a wireless network (4). When being added to the wireless network, the device announces itself in the wireless network and provides an upgrade server in the wireless net-work. The upgrade server makes available for retrieval a digital asset indicative of a characteristic of the device. The master network device detects the device in the wireless network based on the announcement, searches for the upgrade server of the device in the wireless network in response to the detection of the device, and retrieves the digital asset from the upgrade server of the device. This leads to an improved way of integrating the device into the connected device system.

Abstract: An obstacle detection device to be fitted to a mobile vehicle able to move parallel to a reference plane comprises: at least two emitters of electromagnetic beams which are able to form two virtual planes in two different directions that are able to intersect one another and intersect a potential obstacle, at least one image sensor able to produce an image of the intersection of the virtual planes and of the potential obstacle, an image analysis means able to determine the presence of an obstacle, configured to compare the image with a reference image. A detection method employing such a device is also provided.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a storage device, for parking a vehicle in a camera's field of view. The method may include actions of determining that a vehicle is parked within a predetermined distance of a property, obtaining data from a monitoring system component, determining, based on the obtained data, that the vehicle is not parked within a field of view of a camera, determining that (i) the vehicle should be parked within a field of view of the camera and (ii) that an image captured by the camera should include a representation of the vehicle, and performing an operation to instigate movement of the vehicle to a different location that is (i) within a field of view of a camera and (ii) where an image captured by the camera includes a representation of the vehicle.

Abstract: In one embodiment, a client computing device stores a plurality of digital media files. For each digital media files, the client computing device stores data identifying a geographic location associated with the digital media file. When the client computing device receives a request to view digital media files through a graphical user interface, the client computing device uses a GPS receiver to determine a geographic location of the client computing device and an electronic compass circuit to determine an orientation of the client computing device. Based, at least in part, on the location and orientation of the client computing device, the client computing device identifies a plurality of particular geographic locations that are associated with one or more digital media files. The client computing device displays icons corresponding to the plurality of particular geographic locations, ordered by a distance from the client computing device to each of the location.

Abstract: An image processing method includes, when an edit triggering event for a target image is detected, acquiring description information associated with the target image. The description information includes interference information that affects image quality occurred in a shooting process of the target image. The method further includes editing image clips in the target image which are associated with respective interference information of the description information to obtain a processed target image.

Abstract: A system includes: (1) a sensor module configured to detect an object at a predetermined distance and obtain position information of the object relative to the sensor module; (2) a zooming module configured to move at an angle and capture an image of the object; and (3) a controller connected to the sensor module and the zooming module. The controller is configured to derive the angle in accordance with the predetermined distance and the position information of the object, and the controller is configured to control the zooming module to move in accordance with the angle.

Abstract: A positional recording synchronization system can include: creating a time stamped telemetry point for an unmanned aerial vehicle; creating a time stamped recording; creating transformed data from the time stamped recording, the transformed data being tiles for zooming or thumbnails; creating a flightpath array, an image metadata array, and a video metadata array; determining whether entries of the video metadata array match with the flightpath array; determining whether entries of the image metadata array match with the flightpath array; synchronizing the time stamped telemetry point with the time stamped recording based on either the entries of the image metadata array matching the flightpath array, the entries of the visualizer module matching the flightpath array, or a combination thereof; and displaying the time stamped telemetry point as a selection tool for calling, viewing, or manipulating the time stamped recording on a display.

Abstract: A method of calibrating a document scanner contact image sensor scan bar having an LED array to increase LED sensor array output and/or reduce total color scan line cycle time.

Abstract: Methods and systems for navigating a mobile robot through a crowded pedestrian environment based on a trained navigation model are described herein. The trained navigation model receives measurement data identifying nearby pedestrians and the velocity of each nearby pedestrian relative to the mobile robot, and the current position and desired endpoint position of the mobile robot in the crowded pedestrian environment. Based on this information, the trained navigation model generates command signals that cause the mobile robot to adjust its velocity. By repeatedly sampling the velocities of surrounding pedestrians and current location, the navigation model directs the mobile robot toward the endpoint location with a minimum of disruption to the pedestrian traffic flows. The navigation model is trained to emulate desirable pedestrian walking behaviors in a crowded pedestrian environment by tracking the movement of a behavioral trainer through a crowded pedestrian environment.

Abstract: The invention relates to a method for using a driver assistance device of a motor vehicle (1) to detect a mark (7) made on a ground (6) by providing a two-dimensional image by means of an imaging sensor (3) of the driver assistance device (2); by identifying a potential mark (7?) on the basis of the two-dimensional image (11); and by performing a distance measurement by means of a distance sensor (4) of the driver assistance device and also taking a result (17) of the distance measurement as a basis for determining whether the potential mark (7?) is the mark (7) made on the ground (6) or a mark (14) on an object (13) that is on the ground (6).

Abstract: A first augmented-reality (AR) device comprises an optical sensor, a geographic location sensor, an orientation sensor, and a display. The first AR device accesses a first geographic location of the first AR device and an orientation of the first AR device and generates a picture taken at the first geographic location of the first AR device and associated with the orientation of the first AR device. The first AR device retrieves, from a server, transportation information from a second AR device in a vehicle. The server assigns the second AR device to the first AR device. The first AR device forms transportation AR content based on the transportation information and displays the transportation AR content in the display based on the first geographic location and orientation of the first AR device, the transportation information, and the picture generated at the first geographic location of the first AR device.

Abstract: A system and method for using triplet loss for proposal free instance-wise semantic segmentation for lane detection are disclosed. A particular embodiment includes: receiving image data from an image generating device mounted on an autonomous vehicle; performing a semantic segmentation operation or other object detection on the received image data to identify and label objects in the image data with object category labels on a per-pixel basis and producing corresponding semantic segmentation prediction data; performing a triplet loss calculation operation using the semantic segmentation prediction data to identify different instances of objects with similar object category labels found in the image data; and determining an appropriate vehicle control action for the autonomous vehicle based on the different instances of objects identified in the image data.

Abstract: A mobile device is configured as a point and shoot type weapon for gaming, training, or the like. Utilizing the mobile device based weapon, such as a smart phone for example, a user thereof can simply aim the mobile device at a target and shoot. Shooting may be accomplished by tapping the mobile device, activating a button or soft key on the mobile device, making a gesture (e.g., moving a finger as if a trigger is being pulled), providing an audible command (e.g., “shoot”), or the like.

Abstract: A method for detecting a curbstone in an environment of a vehicle. The method includes recognizing at least one line segment that belongs to the curbstone with the aid of image data that are read in by an interface to a camera device of the vehicle. The line segment is projected onto a ground plane of the environment in order to generate a projected line segment. A subset of a plurality of three-dimensionally triangulated points in the environment of the vehicle is assigned to the line segment as a function of a position of the points relative to a position of the camera device, relative to a starting point of the projected line segment, and relative to an end point of the projected line segment. A flank plane of the curbstone is ascertained with the aid of the assigned points.

Abstract: Data streams from multiple image sensors may be combined in order to form, for example, an interleaved video stream, which can be used to determine distance to an object. The video stream may be encoded using a motion estimation encoder. Output of the video encoder may be processed (e.g., parsed) in order to extract motion information present in the encoded video. The motion information may be utilized in order to determine a depth of visual scene, such as by using binocular disparity between two or more images by an adaptive controller in order to detect one or more objects salient to a given task. In one variant, depth information is utilized during control and operation of mobile robotic devices.

Abstract: An image processing apparatus includes an image data acquisition unit configured to acquire image data, a distance information acquisition unit configured to acquire distance information corresponding to the image data, a detection unit configured to detect one or more object regions, each representing an object, from the image data based on the distance information, an extraction unit configured to extract an object region representing an object which does not overlap a specific three-dimensional region, from among detected object regions, based on the distance information, and an image processing unit configured to perform image processing for the image data based on the extracted object region.

Abstract: A method for performing a file transfer comprises receiving a request for data where the data is to be compressed. A checkpoint is established, where the checkpoint occurs before the end of the transmission of the requested data. Also, a compression requirement is established for the requested data. The operations of extracting a segment of the data to be communicated, processing the extracted segment according to the compression requirement, and communicating the processed segment across the network are performed until the checkpoint is reached. Once the established checkpoint is reached, a checkpoint routine is performed by determining whether a desired level of performance in compression is achieved in communicating the processed segments. If the desired level of performance is achieved, each additional segment is processed based upon the compression requirement.

Abstract: A driver assistance systems mountable in a host vehicle while the host vehicle is moving forward with headlights on for detection of obstacles based on shadows. The driver assistance system includes a camera operatively connectible to a processor. A first image frame and a second image frame are captured of a road. A first dark image patch and a second dark image patch include intensity values less than a threshold value. The first and the second dark image patches are tracked from the first image frame to the second image frame as corresponding images of the same portion of the road. Respective thicknesses in vertical image coordinates are measured of the first and second dark image patches responsive to the tracking.