Stereo Object Distance Detection With Adaptive Search Ranges
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing object detection systems face challenges in efficiently reducing calculation load while maintaining accuracy, particularly when objects are present in both the upper and lower areas of the screen.
Innovation Solution
An object distance detection device that utilizes a combination of monocular and stereo processing to detect objects, where monocular processing provides initial object detection and distance estimation, and stereo processing focuses on specific areas with limited search ranges to reduce unnecessary calculations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If the search range is narrowed in the upper area only, then calculation load is reduced in cases without neighboring objects, but object detection accuracy deteriorates when objects exist in both upper and lower areas
Solution Approach 1:
The screen is divided into multiple areas (first area and second area) with different search range narrowing strategies. The first area applies search range narrowing to reduce calculation load, while the second area maintains full search range to ensure detection accuracy when objects are present, thus resolving the contradiction between calculation efficiency and detection accuracy.
Solution Approach 2:
The search range is dynamically adjusted based on detection results from the first area. When an object is detected in the first area, the search range in the second area is accordingly narrowed. This dynamic adjustment allows the system to reduce calculation load when unnecessary while maintaining detection accuracy when objects are present.
2Measurement precision
If full search range is maintained in both areas, then object detection accuracy is preserved, but calculation load increases
Solution Approach 1:
The processing area is segmented into a first area and a second area, where different search range strategies are applied. The first area uses narrowed search range to reduce calculation load, while the second area uses full search range to maintain detection accuracy, thus resolving the contradiction between calculation efficiency and detection accuracy.
Solution Approach 2:
The system performs preliminary detection in the first area before processing the second area. This preliminary action allows the system to identify objects early and accordingly adjust the search range in the second area, reducing unnecessary calculations while maintaining detection accuracy.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The device achieves improved object detection accuracy and reduces calculation load by limiting search ranges in stereo processing, enabling efficient and accurate distance detection.
Implementation Method 1
imaging units 101 and 102 each have an optical lens mounted on an image sensor
Implementation Method 2
detects a distance of an object from a parallax between a plurality of captured images by the plurality of imaging units
Data Source
Figure 1~2
Figure 3~4
Figure 5
AI summary
The purpose of the present invention is to provide an imaging device capable of both improving object detection accuracy and reducing calculation load. The present invention comprises: a plurality of imaging units; an object information acquisition unit that obtains object information including distance information for objects; a search conditions setting unit that sets conditions for searching for image elements corresponding to specific image elements inside a standard image captured by one of the plurality of imaging units, same being searched for inside a reference image captured by another imaging unit; and a stereo distance detection unit that performs said search on the basis of conditions set by the search conditions setting unit and detects the object distance on the basis of parallax obtained by the search. The search conditions setting unit sets the search range for searches, on the basis of the detection information for the object.