Depth Masks for Image Segmentation via Disparity Map Refinement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Depth determinations in composite images from stereoscopic digital camera systems are prone to errors due to incompleteness or noise in disparity maps, leading to artifacts in depth mask generation and image segmentation.
Innovation Solution
The implementation of improved disparity map determination and intelligent depth mask generation methods, including noise reduction techniques and outlier rejection, to enhance depth-based image segmentation and filtering in composite plural images.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If disparity maps are used for depth determination in composite images, then depth-based image segmentation can be achieved, but errors and artifacts occur due to incompleteness or noise in the disparity maps
Solution Approach 1:
The patent applies preliminary noise reduction filtering to disparity maps before depth mask generation. By preprocessing the disparity data to remove noise and fill incomplete regions before depth determination, the system prevents errors from propagating into the final depth mask, thereby improving both measurement precision and reliability
Solution Approach 2:
The patent implements iterative refinement where depth mask quality is evaluated and feedback is used to adjust disparity map processing parameters. This feedback loop allows the system to identify and correct errors in depth determination by comparing expected depth patterns with actual results, improving both accuracy and reliability
2Measurement precision
If noise reduction techniques and outlier rejection are applied to disparity maps, then depth-based image segmentation accuracy improves, but processing complexity increases
Solution Approach 1:
The patent divides the complex noise reduction and outlier rejection process into distinct modular stages: initial noise filtering, disparity validation, outlier detection, and iterative refinement. Each stage handles a specific aspect of data cleaning, making the overall complex process more manageable and computationally efficient while maintaining high segmentation accuracy
Solution Approach 2:
The patent dynamically adjusts processing parameters such as noise threshold values, outlier detection sensitivity, and filtering window sizes based on the specific characteristics of the input disparity map. This adaptive parameter adjustment optimizes the balance between segmentation accuracy and processing complexity for different image scenarios
Data Source
AI summary
A depth-based computational photography method includes recording a first image using a stereoscopic camera system, establishing a disparity map, via a processor, of the first image including a determination of pixel distance between features in the first image and the features in a second image, determining a histogram of pixel disparity values from the disparity map, and removing statistical outlier pixel disparity values. The method further includes identifying holes in disparity map of the first image and computationally shrinking holes in the disparity map via averaging disparity map values based on radially sampled areas proximate to the hole and sharpening histogram pixel disparity value peaks by shifting pixel disparity values in a close neighborhood range of a histogram pixel disparity value peak maximum closer to a pixel disparity value at the histogram pixel disparity value peak maximum.