Stereo Depth Detection for Objects Inside the Minimal Distance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing stereoscopic systems struggle to detect objects closer than the minimal detectable distance due to hardware limitations, leading to false depth assignments and resource-intensive software solutions that are unsuitable for real-time applications.
Innovation Solution
A method combining a hardware stereo algorithm that scans a full range of disparities with a software stereo algorithm using random sampling and reduced passes to ensure reliable detection of objects at all distances, including those closer than the minimal detectable distance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If hardware stereo algorithm is used to scan fixed disparity values, then processing speed is improved, but objects closer than minimal distance cannot be detected
Solution Approach 1:
The patent segments the depth detection task into two parts: hardware stereo algorithm handles objects at normal distances by scanning fixed disparity values for speed, while software stereo algorithm handles objects closer than minimal distance by checking unlimited disparity values. This segmentation allows each algorithm to operate in its optimal range, resolving the contradiction between speed and detection reliability.
Solution Approach 2:
The patent introduces an intermediary mechanism that combines results from both hardware and software stereo algorithms. The system uses the hardware algorithm's fast processing for most cases and supplements it with the software algorithm's comprehensive detection capability, creating a hybrid solution that maintains both speed and reliability.
2Reliability
If software stereo algorithm checks unlimited disparity values, then detection reliability is improved, but processing speed deteriorates
Solution Approach 1:
The patent applies local quality by having different algorithms operate in different disparity ranges. The hardware stereo algorithm with fixed disparity scanning is used for the majority of cases (objects at normal distances), while the software stereo algorithm with unlimited disparity checking is used only when needed (objects closer than minimal distance). This localized application optimizes overall system performance.
Solution Approach 2:
The patent uses partial action by having the hardware algorithm perform the basic disparity scanning for all objects, and then applying the software algorithm's unlimited disparity checking only partially - specifically for cases where objects are closer than the minimal detectable distance. This avoids the full computational cost of unlimited checking while maintaining detection reliability where needed.
3Measurement precision
If down-sampling is applied to increase disparity range, then minimal distance is reduced, but image quality and texture clarity deteriorate
Solution Approach 1:
The patent uses copying by creating a down-sampled version of the input image specifically for disparity range expansion, while preserving the original high-resolution image for quality-critical operations. This allows the system to benefit from the expanded disparity range of down-sampled images without permanently losing the quality of the original images.
Data Source
AI summary
A method for use in a stereoscopic image generating system, wherein the image generating system comprises a hardware module associated with at least one pair of image capturing devices, at least one memory means and at least one processor, wherein information retrieved from the capturing devices is processed by the at least one processor, which is configured to implement a hardware module stereo algorithm for identifying objects included in the captured scene at a distance that is equal to or greater than a minimal distance defined by the geometry and disparity range of said hardware module, and wherein the at least one processor is further configured to implement a software stereo algorithm adapted to identify objects included in the captured scene at a distance that is less than the minimal distance required for detecting objects by the hardware module stereo algorithm.