Parallax Information Processing Through Dynamic-Static Image Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies face challenges in accurately generating parallax information without reducing processing speed, as they either focus on motion areas, leading to reduced resolution and accuracy in static areas, or reduce the entire area to lower resolution, compromising the depth distance calculation.
Innovation Solution
A parallax information generation device that identifies both dynamic and static areas in an image capturing scene, determining a process target area including parts or all of the dynamic area and static area, and performs image processing to generate parallax information, thereby incorporating static areas into the processing without reducing accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If stereo matching is performed only on motion areas, then processing speed is improved, but distance information accuracy in static areas deteriorates
Solution Approach 1:
The patent segments the image processing into two distinct paths: a first processing path for motion areas that performs full-resolution stereo matching, and a second processing path for static areas that uses down-sampled images. This segmentation allows different processing strategies to be applied to different regions, improving overall processing speed while maintaining accuracy where needed.
Solution Approach 2:
The patent applies local quality by using different processing resolutions for different regions. Motion areas receive high-quality full-resolution processing, while static areas use lower-resolution down-sampled processing. This localized quality adjustment optimizes the balance between processing speed and accuracy based on the specific requirements of each region.
2Measurement precision
If the entire image area is processed at full resolution, then measurement precision is improved, but processing speed deteriorates
Solution Approach 1:
The patent divides the image processing into segmented paths based on motion detection results. By identifying motion areas and separating them from static areas, the system can apply appropriate processing intensity to each segment, avoiding unnecessary full-resolution processing in static regions while maintaining accuracy in motion regions.
Solution Approach 2:
The patent applies partial action by performing full-resolution stereo matching only on motion areas rather than the entire image. For static areas, down-sampled processing is sufficient. This partial application of high-resolution processing maintains necessary accuracy while significantly reducing the overall computational burden.
3Productivity
If down-sampled images are used for stereo matching, then processing speed is improved, but resolving power of parallax and depth distance deteriorates
Solution Approach 1:
The patent applies different quality levels locally: down-sampled images are used for static areas where high resolving power is less critical, while full-resolution images are used for motion areas where accurate parallax measurement is essential. This local quality differentiation maintains necessary measurement precision while improving processing speed.
Solution Approach 2:
The patent uses down-sampling as a partial action strategy applied selectively to static areas. By not applying down-sampling to the entire image but only to static regions, the system achieves processing speed improvement without sacrificing parallax resolving power in motion areas where it is most needed.
Data Source
AI summary
A parallax information generation device includes: an imaging unit; a process target area determination unit configured to determine a process target area to be subjected to a predetermined image processing in a base image and a reference image captured by the imaging unit; and an image processing unit configured to perform the predetermined image processing to the process target area to generate parallax information. The process target area determination unit identifies a dynamic area in an image capturing scene, by comparing the images between frames, and determines, as the process target area, an area including a part or the entirety of the dynamic area and a part of a static area that is an area other than the dynamic area.