Vehicle Object Recognition Without Strobe Illumination
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing object detection techniques for vehicles require strobe illumination, increasing mechanical burden and face difficulties when vehicle positions differ at image capture times, making image comparison challenging.
Innovation Solution
An object recognition device with an image-capturing unit, image generation unit, detection unit, and recognition unit that synthesizes images from different time points to detect objects on a road surface without strobe illumination, using sensors to calculate shift amounts and recognize objects based on luminance differences.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If strobe illumination is used for object detection, then detection reliability is improved, but device complexity and mechanical burden increase
Solution Approach 1:
The patent extracts and eliminates the strobe illumination component from the detection system. Instead of using active illumination, the system utilizes existing environmental light sources (street lamps, vehicle lights) to capture images at different time points, thereby removing the mechanical burden and complexity of mounting and controlling a strobe on the vehicle.
Solution Approach 2:
The system uses the vehicle's own motion and the existing environmental lighting conditions to enable detection. By capturing images at different time points during vehicle movement, the system leverages the vehicle's self-motion and ambient light to create detectable differences without requiring additional illumination equipment.
2Reliability
If images are captured at different time points with strobe illumination, then object detection is enabled, but image comparison becomes difficult when vehicle positions differ significantly
Solution Approach 1:
The system performs preliminary calculations of the vehicle's travel distance between image captures and uses this information to pre-adjust the image coordinates. By compensating for the vehicle's movement in advance through coordinate transformation, the system prepares the images for accurate comparison despite the time interval between captures.
Solution Approach 2:
The system changes the spatial parameters of the images by calculating and applying coordinate transformations based on the vehicle's travel distance. This parameter adjustment aligns the images spatially before comparison, ensuring that corresponding features are properly positioned for accurate detection even when the vehicle has moved between captures.
3Illumination intensity
If strobe illumination is mounted on the vehicle, then illumination is provided, but mechanical burden on the vehicle increases
Solution Approach 1:
The patent removes the strobe illumination system from the vehicle, eliminating the associated mechanical burden and weight. Instead of mounting illumination equipment on the vehicle, the system utilizes existing environmental light sources that are already present in the scene, thereby reducing the vehicle's mechanical load while maintaining detection capability.
Data Source
AI summary
An object recognition device includes; an image-capturing unit mounted to a mobile body; an image generation unit that converts images captured by the image-capturing unit at different time points to corresponding synthesized images as seen vertically downwards from above; a detection unit that compares together a plurality of the synthesized images and detects corresponding regions; and a recognition unit that recognizes an object present upon the road surface from a difference between the corresponding regions.