Disparity Deriving Apparatus for Weak Texture Objects
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional range-finding methods using disparity information struggle with precision when measuring distances to objects with weak texture, as they rely on ineffective edge detection and produce unreliable disparity values, leading to inaccurate range-finding and object classification.
Innovation Solution
The implementation of a semi-global matching (SGM) method that calculates synthesis costs by aggregating costs from surrounding pixels, enabling the derivation of disparity values for objects with weak texture and improving range-finding precision by generating high-density disparity images.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If edge detection method is used for disparity derivation, then processing speed is improved, but measurement precision deteriorates for objects with weak texture
Solution Approach 1:
The patent changes the parameter used for matching from edge-based features to intensity-based features. Specifically, it uses the absolute difference of intensity values between corresponding pixels in left and right images, rather than relying on edge detection. This parameter change allows the method to work effectively for objects with weak texture where edge detection fails, thereby improving measurement precision while maintaining processing speed.
2Measurement precision
If cost aggregation method is used for weak texture objects, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The patent segments the disparity derivation process into two distinct stages: (1) initial disparity derivation using simple intensity difference for all pixels, and (2) selective cost aggregation only for pixels identified as having weak texture. This segmentation allows the complex cost aggregation operation to be applied only where necessary, improving measurement precision for weak texture objects while limiting the increase in overall processing complexity.
Solution Approach 2:
Instead of applying cost aggregation to all pixels in the image, the patent applies it partially only to pixels identified as having weak texture characteristics. This partial action approach ensures that the computational complexity increase is minimized while still achieving improved measurement precision for the specific problem case of weak texture objects.
3Device complexity
If conventional stereo matching is used, then processing simplicity is maintained, but reliability deteriorates for objects with weak texture
Solution Approach 1:
The patent introduces an intermediary classification step that identifies pixels with weak texture characteristics. This intermediary process acts as a mediator between the simple conventional stereo matching and the more reliable cost aggregation method. By using this intermediary classification, the system can automatically select the appropriate processing method for each pixel, thereby improving reliability for weak texture objects while maintaining overall processing simplicity through automated decision-making.
Data Source
Figure 1
Figure 2A~2C
Figure 3A~4
AI summary
A disparity deriving apparatus (3) for deriving disparity of an object based on a reference image of the object captured at a first image capturing position and a comparison image of the object captured at a second image capturing position includes a calculator (310) to calculate costs between a first reference area in the reference image and each one of corresponding areas corresponding to the first reference area in a given region in the comparison image, and costs between a second reference area, surrounding the first reference area in the reference image, and each one of corresponding areas corresponding to the second reference area in a given region in the comparison image; a synthesizer (320) to synthesize the costs of the first reference area, and the costs of the second reference area as synthesis costs; a deriving unit (330) to derive a disparity value of the object captured in the first reference area based on the synthesis costs synthesized by the synthesizer; and an output unit (340) to output the disparity value derived by the deriving unit and a specific cost of the first reference area identified by the derived disparity value, the specific cost obtained from the costs calculated for the first reference area by the calculator (310).