Stereo Camera Moving Object Detection via Global Motion Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing pedestrian recognition technologies struggle to detect moving objects effectively when the camera is in motion, which is crucial for preventing traffic accidents involving pedestrians.
Innovation Solution
A device and method that utilize stereo images from moving cameras to select candidate objects, estimate global motions, and determine moving objects by distinguishing between local and global motions using techniques like RANSAC and Lucas-Kanade optical flow, allowing for accurate detection of moving objects even when the camera is moving.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If pedestrian recognition technology uses standard object detection methods, then detection speed is maintained, but detection accuracy deteriorates when the camera is moving
Solution Approach 1:
The patent segments the detection process into three distinct modules: candidate object selection unit that identifies potential objects in stereo images, global motion estimation unit that calculates camera movement, and object determination unit that distinguishes moving objects from stationary background. This segmentation allows each module to specialize in one aspect of the problem, improving overall detection accuracy while maintaining manageable system complexity through modular design.
Solution Approach 2:
The patent introduces global motion estimation as an intermediary component that mediates between the stereo image input and the final object detection output. By first estimating the camera's global motion and then using this information to correct or compensate for motion effects in the detection process, the system achieves accurate moving object detection despite camera movement, resolving the contradiction between accuracy and complexity.
2Measurement precision
If the system distinguishes between local and global motions to detect moving objects, then detection accuracy improves, but processing time increases
Solution Approach 1:
The patent applies preliminary action by performing global motion estimation before the final object determination step. The candidate object selection unit first identifies potential objects, then the global motion estimation unit pre-calculates camera movement characteristics. This preliminary motion estimation allows the object determination unit to quickly distinguish moving objects from stationary background by comparing local object motion against the pre-computed global motion, reducing processing time while maintaining high detection accuracy.
Solution Approach 2:
The patent implements dynamics by adaptively adjusting the detection process based on estimated global motion. When significant camera motion is detected, the system dynamically switches to a motion-compensated detection mode that uses the global motion estimation to correct image coordinates and improve moving object detection. This dynamic approach ensures high detection accuracy only when necessary, optimizing processing time by avoiding full motion analysis in static scenarios.
3Adaptability or versatility
If the system uses stereo images from moving cameras, then moving object detection capability is improved, but system complexity increases
Solution Approach 1:
The patent achieves universality by designing a detection system that handles both stationary and moving camera scenarios through the same architecture. The global motion estimation unit can estimate motion for any camera movement type, and the object determination unit adapts its detection criteria based on the estimated motion. This multi-functional design enables the system to detect moving objects effectively whether the camera is stationary, moving linearly, or rotating, without requiring separate specialized systems for each scenario.
Solution Approach 2:
The patent applies parameter changes by dynamically adjusting detection parameters based on the estimated global motion. The system modifies coordinate systems, motion thresholds, and detection sensitivity levels according to the camera's motion state. For example, when high camera velocity is detected, the system adjusts its motion compensation parameters and detection thresholds to maintain accuracy. This parameter adaptation allows the system to handle diverse camera motion scenarios while maintaining a unified detection framework, improving versatility without proportionally increasing complexity.
Data Source
AI summary
An object detection device may include: a candidate object selection unit suitable for selecting candidate objects from a stereo image transmitted from moving cameras; a global motion estimation unit suitable for estimating global motions representing movement of the moving cameras from the stereo image; and an object determination unit suitable for detecting a moving object from the stereo image based on the candidate objects and the global motions.