Vehicle Surround-View Image Stitching for Rear Blind Spot Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing systems for vehicles struggle to combine images from multiple cameras to provide a wide field of view without creating blind spots, especially when the cameras have different angles of view and are positioned to capture rear views on both sides of the vehicle.
Innovation Solution
An image processing apparatus that includes multiple cameras positioned to capture images from different angles, with a controller that combines these images to generate a composite view of the surroundings, reducing blind spots by aligning the angles of view and processing the images to create a seamless rear view display.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If multiple cameras are positioned at different locations to capture rear views, then the field of view is widened, but blind spots are created due to different angles of view
Solution Approach 1:
The patent uses multiple cameras positioned at different spatial locations (center rear, left rear, right rear) to capture images from different dimensions and angles. By combining these multi-dimensional views, the system achieves a comprehensive rear view that eliminates blind spots while maintaining a wide field of view.
Solution Approach 2:
The control unit merges images from multiple cameras positioned at different locations into a single composite rear view image. This combining process integrates the fields of view from all cameras to create a seamless panoramic view that eliminates blind spots while providing a wide overall field of view.
2Loss of information
If images from multiple cameras are combined to generate a wide field of view, then visibility is enhanced, but processing complexity increases
Solution Approach 1:
The patent divides the rear view into multiple image segments captured by different cameras (center, left, right). Each camera captures a specific portion of the rear view, and the control unit processes these segmented images separately before combining them, which simplifies the overall processing complexity compared to using a single complex wide-angle system.
Data Source
AI summary
An image processing apparatus, a mobile object, and a program are provided that are able to generate an image of surroundings of a mobile object without increasing a processing load. An image processing apparatus (10) includes an input unit (101) and a controller (102). The input unit (101) is configured to acquire a first image and a second image. The first image is captured by a first image capture device (11) configured to shoot in a first direction toward rear on a right-hand side relative to a direction of travel of the mobile object. The second image is captured by a second image capture device (12). The second image capture device (12) is disposed away from the first image capture device (11) in the direction of travel and in a left or right direction and is configured to shoot in a second direction toward the rear on the right-hand side. The controller (102) is configured to perform a first control to generate the image of the surroundings of the mobile object based on the first image and the second image.