A method for detecting a raised object located within a parking area, using at least two video cameras, which are spatially distributed inside of the parking area, and whose respective visual ranges overlap in an overlapping region; the method including the following steps: recording respective video images of the overlapping region, using the video cameras; analyzing the recorded images, in order to detect a raised object in the recorded video images; the analyzing being carried out exclusively by at least one of the video cameras, inside of the video camera(s). A corresponding system, a parking area and a computer program are also described.

An image monitoring system for vehicle for detecting a left lane area and a right lane area behind a vehicle using a region of interest (ROI) is provided. The system includes a rear camera, a memory configured to store an image obtained from the rear camera, and a controller configured to be set to one of a driving mode and a parking mode based on whether the vehicle is traveling and to detect an event from the image obtained from the rear camera. The controller set to the driving mode sets the ROI corresponding to the left lane area and the right lane area behind the vehicle on the image obtained from the rear camera, while the vehicle is traveling, detects whether an object is located on or enters the ROI, and notifies a driver of the vehicle of the detected result using an alarm.

A charging support device is used in charging a battery of a vehicle having a power reception unit on a bottom surface of the vehicle with power supplied from a power transmission device located outside the vehicle. The charging support device includes a controller that generates an overhead view image that displays a vehicle image of the vehicle and an image of a surrounding area of the vehicle captured by a camera. The controller displays, in the overhead view image, (i) a first mark indicating a position of the power reception unit and (ii) a second mark indicating a position of the power transmission device. The controller varies a color or a size of the second mark in a case where a positional relationship between the position of the power reception unit and the position of the power transmission device achieves a predetermined charging efficiency.

A vision system for a vehicle includes a plurality of cameras and a processor operable to process image data captured by the cameras to generate images derived from image data captured by at least some of the cameras. The processor generates a vehicle representation. A display screen, viewable by a driver of the vehicle, displays the generated images and the vehicle representation as would be viewed from a virtual camera viewpoint exterior to and higher than the vehicle itself. A portion of the vehicle representation is rendered as displayed to be at least partially transparent to enable viewing at the display screen of an object present exterior of the vehicle that would otherwise be partially hidden by non-transparent display of that portion of the vehicle representation.

An apparatus configured to park a vehicle includes a plurality of cameras configured to acquire images of surroundings of the vehicle; a display configured to display an around-view image showing a view around the vehicle that is generated by combining the images acquired by the plurality of cameras; and a processor. The processor is configured to determine a first input that is applied on the around-view image displayed on the display; display, according to the first input that has been applied on the around-view image, a parking guide image representing a user-specified parking region for the vehicle; determine whether there is an actual parking region that matches the displayed parking guide image representing the user-specified parking region; and based on a determination that there is an actual parking region that matches the displayed parking guide image, provide a signal that controls a parking operation of the vehicle.

Tracking an object using an unmanned aerial vehicle is disclosed. A plurality of images showing the object is received from a camera of the unmanned aerial vehicle. A first static characteristic, a second static characteristic, a first dynamic characteristic, and a second dynamic characteristic of the object are determined. The second static characteristic is compared to the first static characteristic, and the second dynamic characteristic is compared to the first dynamic characteristic. It is determined that the second static characteristic is approximately equal to the first static characteristic, and that the second dynamic characteristic is approximately equal to the first dynamic characteristic. Finally, it is determined that the object is moving.

A method and apparatus that detect a leak of vehicle fluid are provided. The method includes determining whether leak detection is active, capturing a first image of an area while parking a vehicle into the area if the leak detection is active, capturing a second image of the area while the vehicle is exiting from the area, comparing the first image and the second image to detect whether the leak is present, and outputting a notification indicating a leak is present if the comparing detects the leak.

A method computes a final image of an area surrounding a vehicle, from at least one image captured in a first angular portion of the area surrounding the vehicle by a camera and a distance between the vehicle and a point in the surrounding area, which distance is determined, in a second angular portion located in one of the first angular portions, by at least one measurement sensor. The method includes capturing an image and, for each captured image, correcting a distortion in the captured image to generate a corrected image; for each corrected image, transforming the perspective of the corrected image using a matrix storing a pre-calculated distance between the camera and a point in the corresponding surrounding area; and adding the transformed images to obtain the final image. The perspective transforming includes adjusting the pre-calculated distance to the distance determined by the measurement sensor.

The present disclosure provides various approaches for smart area monitoring suitable for parking garages or other areas. These approaches may include ROI-based occupancy detection to determine whether particular parking spots are occupied by leveraging image data from image sensors, such as cameras. These approaches may also include multi-sensor object tracking using multiple sensors that are distributed across an area that leverage both image data and spatial information regarding the area, to provide precise object tracking across the sensors. Further approaches relate to various architectures and configurations for smart area monitoring systems, as well as visualization and processing techniques. For example, as opposed to presenting video of an area captured by cameras, 3D renderings may be generated and played from metadata extracted from sensors around the area.

A neural network may be used to determine corner points of a skewed polygon (e.g., as displacement values to anchor box corner points) that accurately delineate a region in an image that defines a parking space. Further, the neural network may output confidence values predicting likelihoods that corner points of an anchor box correspond to an entrance to the parking spot. The confidence values may be used to select a subset of the corner points of the anchor box and/or skewed polygon in order to define the entrance to the parking spot. A minimum aggregate distance between corner points of a skewed polygon predicted using the CNN(s) and ground truth corner points of a parking spot may be used simplify a determination as to whether an anchor box should be used as a positive sample for training.