Vehicle Image Capture Using Burst Super-Resolution
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face challenges in creating high-resolution images due to demosaicing techniques that introduce artifacts like chromatic aliasing, false gradients, and Moiré patterns, which degrade image quality and hinder effective object detection.
Innovation Solution
The system captures multiple frames of an image with sub-pixel offsets during movement and performs super-resolution computations to create enhanced images with reduced artifacts, using Gaussian radial basis function computations and robustness models to align and accumulate color planes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If demosaicing techniques are used to convey the scene as a color image, then color information is obtained, but chromatic aliasing, false gradients, and Moiré patterns occur leading to poor resolution and artifacts
Solution Approach 1:
The image capture device divides the sensor array into multiple independent color sensor arrays (e.g., red, green, blue channels) rather than using a single sensor with a color filter array. Each color channel is captured separately without requiring demosaicing, thereby eliminating chromatic aliasing and Moiré patterns while maintaining high image resolution.
Solution Approach 2:
The patent introduces multiple independent color sensor arrays as intermediaries to capture different color channels simultaneously. This intermediary approach replaces the traditional single sensor with color filters, allowing direct capture of full-color information at each pixel location without the need for demosaicing operations that cause artifacts.
2Measurement precision
If multiple frames are captured in a burst sequence with sub-pixel offsets, then super-resolution images can be created, but processing time and computational complexity increase
Solution Approach 1:
The system captures multiple frames with sub-pixel offsets in a burst sequence as a preliminary action before final image processing. By pre-capturing the necessary multiple frames with intentional sub-pixel movements, the system prepares all required data for super-resolution computation in advance, enabling efficient processing and reducing overall latency.
Solution Approach 2:
The patent employs periodic burst capture sequences where multiple frames are captured in rapid succession with controlled sub-pixel offsets. This periodic action pattern allows the system to gather sufficient data for super-resolution while maintaining a predictable and optimized processing rhythm, balancing image quality with processing time.
Data Source
AI summary
Described examples relate to an apparatus comprising one or more image sensors coupled to a vehicle and at least one processor. The at least one processor may be configured to capture, in a burst sequence using the one or more image sensors, multiple frames of an image of a scene, the multiple frames having respective, relative offsets of the image across the multiple frames and perform super-resolution computations using the captured, multiple frames of the image of the scene. The at least one processor may also be configured to accumulate, based on the super-resolution computations, color planes and combine, using the one or more processors, the accumulated color planes to create a super-resolution image of the scene.