Connaught Electronics Ltd.

Systems and methods for altering colors of pixels of images produced by cameras using a lens. A plurality of generated images are received, and color channel values and luminance values of image pixels are determined. Benchmark U-color thresholds and V-color thresholds are established based on the color channels and luminance associated with the pixels. A first image generated using a first lens of a first camera is received. U-color values and V-color values associated with first pixels of the first image are determined. A uniformity score associated with the first lens is received. The benchmark U-color and V-color thresholds are altered based on the uniformity score. The U-color values and V-color values of the first pixels are corrected in response to the U-color values being within the altered U-color thresholds and the V-color values being within the altered V-color thresholds.

Methods and systems for displaying image data in a vehicle are disclosed herein. A method for displaying image data include providing initial and final projection surfaces; receiving a video stream and is buffered in a buffer storage, which is updated at a frame rate. The method includes reading out image data at a readout rate from the buffer at an initial readout instance, intermediate readout instances and a final readout instance. The method includes determining respective sets of initial and final sampling points on the projection surfaces. For each of the intermediate readout instances, the method includes determining intermediate sampling points, which interpolate between the final sampling points and the respective initial sampling points, and an intermediate projection surface containing the intermediate sampling points. The method includes projecting the read out image data to the respective intermediate projection surface and a corresponding image is displayed on a display device.

A camera for a motor vehicle is disclosed. The camera includes a housing, wherein a receiving space, in which a lens module of the camera and a circuit board of the camera are arranged, is bounded by the housing. The camera has a longitudinal axis, wherein the circuit board and the lens module are arranged axially spaced from each other in the direction of this longitudinal axis such that a clearance is formed between the circuit board and the lens module. An elastic seal is arranged between the circuit board and the lens module such that the clearance is sealed from the remaining volume space in the housing by the seal.

For computing height information of a curb, line segments approximating the curb in a ground plane are determined, a perpendicular bisector of a first line segment is computed, and a second segment are determined, which intersects the perpendicular bisector. A first plane including the first segment and a second plane including the second segment are determined. A midpoint of the first segment is projected to the second plane and an intersection point of the second segment is projected to the first plane. Depending on respective resulting heights, the first and the second segment are classified as upper and lower, respectively. An upper point of the upper segment is projected to a further plane, which is perpendicular to the ground plane and includes the lower segment, and a height of the projected upper point above the ground plane is determined.

A method for automatic visual perception includes generating first feature maps from a camera image by a first encoder module of a neural network and the first feature maps are transformed into a top view perspective, emitting an ultrasonic pulse into the environment and generating an ultrasonic sensor signal depending on reflected portions of the emitted ultrasonic pulse. The method includes generating a spatial ultrasonic map depending on the ultrasonic sensor signal and generating second feature maps from the ultrasonic map by a second encoder module of the neural network. The method includes fusing the transformed first feature maps and the second feature maps and carrying out a visual perception task by a decoder module of the neural network depending on the fused feature maps.

A MEMS device includes an electrical distribution substrate and a spacer ring extending upward therefrom. An actuator stator is positioned above the electrical distribution substrate and within the spacer ring. An outer frame extends from a floor of the actuator stator. An actuator rotor is suspended above the floor. A sensor is supported by the actuator rotor. A conductive stack is positioned above the spacer ring. A wire electrically connects the sensor to the conductive stack. A plurality of vias extending through the spacer ring and to the electrical distribution substrate, thereby allowing electrical communication between the sensor and the electrical distribution substrate while enabling vertical displacement of the sensor and actuator rotor.

A method includes obtaining image frames from each camera disposed along a vehicle, where each image frame corresponds to a same timestamp. The method further includes constructing a first birds-eye view (BEV) image from each image frame with a first BEV module and constructing a second BEV image from each image frame by Inverse Perspective Mapping (IPM) with a second BEV module. The first BEV module extracts features of an external environment of the vehicle from each image frame, transforms the features to a three-dimensional space, and projects the three-dimensional space onto an overhead two-dimensional plane. Subsequently, a merging module merges the first and second BEV images to produce a hybrid BEV image. Features of an external environment of the vehicle within the hybrid BEV image are detected by a deep learning neural network and the hybrid BEV image is displayed to a user in the vehicle.

In some implementations, the device may include receiving, via an image sensor associated with a vehicle, an image of a parking area. In addition, the device may include determining that the image of the parking area includes an available parking spot; analyzing the image of the parking area to locate a number associated with the available parking spot; determining that the vehicle is entering the available parking spot; creating, in response to the vehicle entering the available parking spot, a generated image that includes the number associated with the available parking spot based on the image of the parking area; and displaying the generated image on a vehicle display in response to the vehicle occupying the available parking spot.

A composite top view of an area surrounding a vehicle is generated utilizing image harmonization and proximity sensor data. Images from various vehicle cameras are received, wherein at least two of the cameras can see the same portion of the environment in an overlapping region. The images are segmented into segments. The overlapping regions include some of the segments of one image and some of the segments of another region to define overlapping segments. Sensor data is received from a plurality of proximity sensors, wherein the sensor data indicates a location of an object outside of the vehicle. The color and/or brightness of the images are harmonized based at least in part on the sensor data. In particular embodiments, weights are placed on the color and/or brightness of the overlapping segments associated with the location of the object. Or these overlapping segments can be removed from the harmonization.

Methods and systems for generating a top view of an area surrounding the vehicle. Images are received from a plurality of vehicle cameras configured to view respective regions of ground outside of the vehicle. Portions of at least some of the regions overlap with portions of another of the regions, defining overlapping portions. Each image is segmented into a respective plurality of segments, wherein the overlapping portions include some of the segments of one image and some of the segments of another image to define overlapping segments. A first harmonization model harmonizes the images based on brightness or color values within the overlapping segments. When an angular orientation of at least one of the plurality of vehicle cameras relative to the ground exceeds a threshold, a second harmonization model is executed in order to account for occlusions or lack of image data when the camera is angled.