Vehicle Camera Image Compression Using Depth Vanishing Point
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional vehicle display systems using aspherical mirrors and camera monitoring systems often create a sense of incongruity for drivers due to distorted images and inadequate depth perception, making it difficult for them to accurately grasp the external environment.
Innovation Solution
A visual field support image generation device that uses a geometrically-shaped lens model, such as a vertically long elliptical lens, to compress images in the horizontal direction more than the vertical direction, centered on the depth vanishing point, to create a more natural and less distorted view, while also employing multiple cameras for improved depth perception and gaze-point focused image processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Shape
If horizontal scale correction processing is performed to prevent traveling region dividing lines from being displayed in a curved shape, then the straightness of lane markings is improved, but the naturalness of the displayed image deteriorates due to artificial distortion
Solution Approach 1:
The patent changes the fundamental parameter of image conversion from artificial horizontal/vertical scaling to geometric transformation based on lens model and depth vanishing point. This allows the image to naturally converge at distant points while maintaining straight lane markings, resolving the contradiction between geometric accuracy and natural appearance.
Solution Approach 2:
The patent creates a virtual camera image that copies the optical characteristics of a geometric lens rather than directly processing the flat camera sensor image. This virtual copying approach preserves natural depth perception while achieving straight lane markings through proper perspective transformation.
2Area of stationary object
If the field angle is extended in the horizontal direction using aspherical mirror principles, then the coverage area is improved, but the image distortion increases creating a sense of incongruity
Solution Approach 1:
The patent transforms the image using geometric lens model parameters including focal length, principal point, and depth vanishing point. This mathematical transformation extends the effective field angle while correcting distortion by simulating how light actually travels through a geometric lens, thereby maintaining natural object shapes across the expanded coverage area.
Solution Approach 2:
The patent introduces depth information as an additional dimension in the image transformation process. By using depth vanishing point and perspective transformation, the system handles the extended field angle in three-dimensional space rather than simple two-dimensional scaling, preserving shape accuracy while expanding coverage.
3Measurement precision
If multiple cameras are used to improve depth perception, then the accuracy of spatial understanding is improved, but the device complexity increases
Solution Approach 1:
The patent introduces a virtual camera as an intermediary that synthesizes depth information mathematically from a single physical camera's image. This virtual camera model, based on geometric lens parameters and depth vanishing point, provides accurate depth perception without requiring multiple physical cameras, thereby reducing device complexity while maintaining measurement precision.
Data Source
AI summary
A visual field support image generation device for generating a visual field support image of a vehicle has a camera that captures an image from the vehicle and a processing section. The processing section covers the image captured by the camera to generate the visual filed support image. The image is covered by compressing the captured image so that a compression ratio of the captured image in a horizontal direction becomes higher than a compression ratio of the captured image in a vertical direction by using a depth vanishing point included in the captured image as the center.