Autonomous Driving HDR Imaging for Tunnel Contrast Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous driving vehicles face challenges in accurately recognizing geographic features in environments with high brightness contrast, such as tunnels, which can hinder safe predictive driving due to difficulties in understanding the illumination levels inside and outside these areas.
Innovation Solution
An image processing apparatus that generates a synthesized image by adjusting the brightness of objects with significant contrast, using a high dynamic range (HDR) scheme and exposure adjustment based on the difference between the target object and its surroundings, to enhance recognition and reduce unnecessary resource consumption.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If exposure adjustment is performed on all regions of the input image, then the recognition accuracy of target objects is improved, but the computational complexity and processing time increase
Solution Approach 1:
The patent applies exposure adjustment selectively only to regions containing target objects rather than uniformly to the entire image. The image processing apparatus identifies target objects and performs brightness adjustment localized to those specific regions, thereby maintaining recognition accuracy while significantly reducing the computational burden associated with processing the whole image.
Solution Approach 2:
The patent segments the input image into multiple regions based on target object detection. By dividing the image processing task into region-specific operations, the system applies computational resources only where necessary, reducing overall complexity while preserving the recognition accuracy needed for autonomous driving safety.
2Measurement precision
If exposure adjustment is performed on all frames, then the recognition accuracy of target objects is improved, but the processing time increases
Solution Approach 1:
The patent performs exposure adjustment only on frames and regions containing target objects, rather than processing every frame uniformly. This selective approach maintains accurate recognition of tunnel entrances and exits while reducing the total processing time by eliminating redundant computations in frames without relevant objects.
Solution Approach 2:
The patent applies exposure adjustment partially - only to the extent necessary for recognizing target objects, rather than applying it excessively to all image data. This partial action approach ensures sufficient recognition accuracy for safety-critical objects while minimizing processing time delays.
3Measurement precision
If brightness adjustment is applied to enhance target object recognition, then the recognition accuracy improves, but the energy consumption increases
Solution Approach 1:
The patent applies brightness adjustment locally only to regions containing target objects rather than to the entire image. This localized processing maintains the recognition accuracy needed for safe autonomous driving while significantly reducing the energy consumption associated with processing and adjusting the complete image data.
Data Source
Figure 1
Figure 2~3
Figure 4
AI summary
A processor implemented image processing method includes recognizing a target object in a first frame of an input image; adjusting an exposure of a second frame of the input image based on a brightness of the target object; and generating a synthesized image by synthesizing the first frame and the second frame.