Autonomous Driving Object Recognition Under Extreme Illumination
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous driving vehicles face challenges in recognizing objects in environments with extreme illumination changes, such as dark or bright conditions, due to the limited dynamic range of typical cameras, which affects their ability to capture objects accurately.
Innovation Solution
The method involves using a dynamic vision sensor (DVS) to detect object information and adjust the photographic configuration of the camera to enhance image capture in regions where objects are unrecognizable, thereby improving object recognition rates in varying lighting conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a typical camera is used to capture images in autonomous driving, then the device complexity is low, but the object recognition rate deteriorates in environments with extreme illumination changes
Solution Approach 1:
The patent combines a dynamic vision sensor (DVS) with a typical camera to create a hybrid vision system. The DVS detects extreme illumination changes and triggers the camera to capture images only when necessary, merging the advantages of both sensors to improve object recognition rate while managing device complexity.
Solution Approach 2:
The dynamic vision sensor performs preliminary detection of illumination conditions before the camera captures images. By pre-identifying extreme lighting scenarios, the system prepares for optimal image capture timing, improving recognition rate without continuously operating the camera.
2Reliability
If the camera operates continuously to capture objects in all conditions, then the object recognition rate improves, but the use of energy increases
Solution Approach 1:
Instead of continuous operation, the camera performs periodic capture triggered by illumination change events detected by the DVS. This periodic action ensures objects are captured when lighting conditions are favorable while conserving energy during normal conditions.
Solution Approach 2:
The dynamic vision sensor autonomously monitors illumination conditions and self-determines when camera activation is necessary. This self-service mechanism eliminates the need for continuous camera operation, reducing energy consumption while maintaining recognition rate.
3Area of stationary object
If the camera operates in extreme lighting conditions, then the coverage area increases, but the measurement precision deteriorates
Solution Approach 1:
The dynamic vision sensor acts as an intermediary between the environment and the camera, filtering out extreme lighting conditions. It mediates by detecting when illumination is suitable for accurate capture, allowing the camera to maintain measurement precision across the full coverage area.
Data Source
AI summary
Disclosed is an object recognition method including: obtaining a first RGB image by using a camera; predicting at least one first region, in which an object is unrecognizable, in the first RGB image based on brightness information of the first RGB image; determining at least one second region, in which an object exists, from among the at least one first region, based on object information obtained through a dynamic vision sensor; obtaining an enhanced second RGB image by controlling photographic configuration information of the camera in relation to the at least one second region; and recognizing the object in the second RGB image.