Automotive camera market is expected to reach USD 15.1 billion by 2025 (marketsandmarkets), registering CAGR of 10.8% between 2019 and 2025. With increasing awareness around multi camera technologies, automotive companies have started introducing surround view systems into their mid-priced and economic vehicle segments.
A properly implemented surround view system improves safety and driver experience of a vehicle by providing visibility in the immediate vicinity of the vehicle. Automotive surround view system provides better visibility and orientation 360-degrees around the vehicle and helps in mitigating collisions, slow speed incidents, and vehicle damages during lane change, parking, and maneuvering.
Understanding Automotive Surround View Technology
Surround-view camera systems consist of four fish eye cameras with each having >180-degrees horizontal field of view. Automotive cameras in the system are installed in two wing mirrors or side doors, in front and at the rear side of the vehicle and provides 2D and 3D view of the vehicle’s surroundings.
The surround-view camera system performs two primary functions, first is camera calibration and second is merging of multiple video streams into one.
The fish eye cameras produce images from all dimensions of the vehicle and these images are simultaneously sent to ECU where they are processed for correction, combined and stitched together to produce real-time view of the surroundings of the vehicle.
There can be two types of surround view systems:
- 2D Surround View Camera Systems: Two dimensional surround view is a traditional view, which projects a bird’s eye view of the vehicle’s surroundings on the display screen. Original images or video feeds from the cameras are stitched to form the bird’s eye view on a flat surface.
- 3D Surround View Camera Systems: In 3D surround view systems, vehicle and its surroundings are shown in 3D representation, which is in spherical form. Since it is a 3D representation of the surroundings in 360-degrees, the view can be fetched from any angle around the vehicle.
There are two primary functions of Surround view systems:
Surround View System Camera Calibration
Camera calibration can be considered as one of the most challenging tasks while deploying a surround view system in a vehicle. It is the process of calibrating the surround view cameras with the outside environment. Camera calibration ensures proper functioning of the surround view system as improper camera calibration may cause non-aligned image stitching, ghosting and faulty color correction. Majorly OEMs support the installation of cameras (from automotive camera suppliers) into the vehicle and ensures proper camera calibration into the system. Majorly, there are two types of camera calibration parameters:
- Intrinsic Parameters: Intrinsic parameters in camera calibration emerges from the side of automotive cameras suppliers. Since surround view cameras have a larger field of view, it is important to measure parameters as optical centers of the camera, focal length, camera projections, sensor resolution, and lens distortion, as avoiding these parameters may result in distorted image formation.
- Extrinsic Parameters: Extrinsic parameters in camera calibration depends on the relative positions and orientation of the cameras with respect to a common point in the vehicle’s body. Extrinsic parameters in camera calibration is important to define the image overlap regions. Not defining, it may lead to improper overlapping of the surround view image. Extrinsic parameters of the camera calibration can also be affected by the suspension of the vehicle as faulty suspension may lead to in displacement to the camera position.
Image Recognition in Surround View Systems
Image recognition is the key function of surround view system as it is capable of processing the images from multiple surround view cameras in real-time. Automotive image recognition processors play a crucial role in ADAS applications like recognizing vehicles, pedestrians, traffic signs, road line recognition (Lane keeping system, navigation accuracy interpolation), etc. and ultimately leading to assist in automotive surround view systems. New age image recognition processors supported by intelligent vision and image processing algorithms are capable of supporting multiple camera sensors and high-resolution displays in surround view systems. There are many platforms available for image recognition functionalities. Qualcomm’s Eragon™ 820: EIC-Q820-200, which is a cost effective platform and designed to deliver up to 40 percent improvement computing capabilities compared to its predecessors.
Surround view system in automotive applications generates valuable visual information supported by computer vision algorithms to assist in increasing driver’s visibility to the external environment. It also helps in improving ADAS functionalities for autonomous vehicles.
eInfochips as an Automotive IoT Solutions provider assist automotive companies to design and develop vision based ADAS systems through camera-based algorithm porting, sensor fusion for vision, and feature enhancement through code testing.