Object Detection Device Using Stereo Camera Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing object detecting devices face challenges in accurately detecting objects both near and far, leading to erroneous detection or non-detection, particularly in far regions where distance data is scarce, and require lengthy processing times.
Innovation Solution
The object detecting device employs a stereo camera system with a near-far boundary distance setting unit to differentiate between near and far regions, using disparity images from two cameras to accurately detect objects in both areas, with a near object detecting unit processing near regions and a far object detecting unit using pattern matching in the far region, thereby reducing processing time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a device uses distance image data to detect objects, then near objects can be detected accurately, but far objects with small pixels suffer from erroneous detection or non-detection
Solution Approach 1:
The patent divides the detection space into near region and far region based on distance thresholds. Different detection algorithms are applied to each region: disparity-based detection for near objects and pattern matching for far objects. This segmentation allows each algorithm to operate in its optimal range, improving overall detection accuracy and reliability.
Solution Approach 2:
The patent changes the detection parameter from disparity-based measurement to pattern matching based detection for far objects. By switching detection methods according to distance parameters, the system maintains high detection accuracy for both near and far objects, overcoming the limitation of small pixel sizes in far region detection.
2Reliability
If a device processes all distance data to detect objects, then comprehensive detection is achieved, but processing time increases
Solution Approach 1:
The patent segments the detection process into two parallel paths: one for near objects using disparity data and another for far objects using pattern matching on single images. This segmentation enables simultaneous processing of different object types, reducing overall processing time while maintaining comprehensive detection coverage.
Solution Approach 2:
The patent applies partial processing by selecting only the necessary detection method for each region. For far objects, it uses pattern matching on a single image rather than processing all disparity data, which reduces computational load and processing time while still achieving reliable detection.
3Measurement precision
If a device divides distance data into solid object and road surface information, then solid object detection is improved, but division becomes difficult in far regions where distance data decreases
Solution Approach 1:
The patent inverts the traditional approach by not dividing distance data into solid objects and road surfaces. Instead, it directly applies pattern matching to the entire image for far object detection, bypassing the complex division process entirely. This simplifies the detection process while maintaining accurate object detection in far regions.
Solution Approach 2:
The patent extracts only the essential detection information needed for far objects using pattern matching, rather than attempting to fully separate and classify all distance data. This extraction approach reduces processing complexity while focusing on the critical task of detecting far objects regardless of road surface variations.
Data Source
Figure 1
Figure 2
Figure 3(a)~3(b)
AI summary
The purpose of the present invention is to provide an object detecting device which is capable of accurately detecting an object even far away, and of shortening processing time. Provided is an object detecting device (100), comprising: a disparity acquisition unit (116) which compares each image of two cameras (112, 113) and computes a disparity for each pixel; a near-far boundary setting unit (118) which, in a single image of one of the two cameras, sets a boundary (Rb) between a near region (R1) which is close to a vehicle (110) and a far region (R2) which is distant from the vehicle (110) ; a near object detecting unit (119) which detects objects (102, 104) of the near region (R1) on the basis of the disparity; and a far object detecting unit (120) which detects objects (103, 104) of the far region (R2) on the basis of the single image.