Mono Camera Object Recognition Using Optical Flux Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing object recognition systems for vehicles rely on stereo cameras, requiring knowledge of vehicle and object motion, which is challenging for static objects and prone to errors in proper motion estimation, especially in central image areas crucial for collision detection.
Innovation Solution
A method using a mono camera that forms and segments optical flux profiles along a predetermined image line to recognize objects without relying on stereo data or proper motion estimation, employing homography modeling to differentiate between background, object, and roadway planes, enabling robust object detection even with moving objects.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If stereo cameras are used for object recognition, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The patent uses a mono camera to capture images and creates virtual depth information through optical flux analysis, effectively copying the depth perception capability of stereo cameras through computational methods rather than physical stereo hardware
Solution Approach 2:
The patent replaces the mechanical stereo camera system with a mono camera system that uses optical flux profile analysis and homography modeling to achieve object recognition, substituting physical depth measurement with computational image processing
2Measurement precision
If proper motion estimation is used in stereo image processing, then object detection accuracy is improved, but reliability deteriorates due to errors in motion estimation
Solution Approach 1:
The patent extracts and analyzes only the optical flux profile along predetermined image lines without requiring full stereo image processing or motion estimation, isolating the essential depth information needed for object recognition while eliminating sources of error
Solution Approach 2:
The patent introduces homography modeling as an intermediary to relate optical flux profiles to depth information, providing a reliable mathematical framework that avoids direct motion estimation while still enabling accurate object detection
3Measurement precision
If stereo image processing is performed, then object recognition capability is improved, but computing power requirements increase
Solution Approach 1:
The patent extracts only the necessary optical flux profile information along specific image lines rather than performing comprehensive stereo image processing, reducing computational load while maintaining object recognition capability
Solution Approach 2:
The patent applies partial action by processing only selected image lines and using simplified homography modeling instead of full stereo reconstruction, achieving sufficient object recognition with reduced computational effort
Data Source
AI summary
A method for recognizing an object includes reading in a first image signal that represents a first camera image recorded by a mono camera, and a second image signal that represents a second camera image recorded by the mono camera. First pixels situated on an image line of the first camera image are selected from the first image signal. Second pixels are identified from the second image signal, the second pixels corresponding to the first pixels. A flux signal is formed using the first pixels and the second pixels, the flux signal representing an optical flux profile for the first pixels situated along the image line. The flux profile represented by the flux signal is segmented into a plurality of segments, each of which represents a plane in the vehicle surroundings. An object signal that represents a recognized object is determined, using the plurality of segments.