Depth Map Post-Processing for Irregular Edge Repair
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for obtaining depth information, such as active and passive scanning, often result in depth maps with quality issues like black holes and irregular edges, which affect subsequent applications and have not been adequately addressed by current post-processing techniques.
Innovation Solution
A method and device for post-processing depth/disparity maps that involves edge extraction, super pixel-based segmentation, and weighted median filtering to identify and repair irregular edge regions, improving the accuracy and quality of depth maps.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If active scanning or passive scanning methods are used to obtain depth information, then depth maps can be generated, but the depth maps contain quality issues such as black holes and irregular edges
Solution Approach 1:
The patent extracts edge information from the depth map using edge detection algorithms, separating the problematic irregular edge regions from the rest of the depth map for targeted processing. This extraction allows the algorithm to focus specifically on repairing edge artifacts while preserving the overall depth information.
Solution Approach 2:
The patent applies different processing strategies to different regions of the depth map. Regular regions undergo standard processing, while irregular edge regions identified through edge information and superpixel segmentation receive specialized repair treatment using weighted median filtering and superpixel-based region growing, thereby improving local quality without unnecessarily processing the entire map.
2Reliability
If conventional filtering methods are applied to remove noise from depth maps, then some noise can be reduced, but black holes and irregular edges still remain after processing
Solution Approach 1:
The patent segments the depth map into superpixels based on color image information and depth map characteristics. This segmentation divides the image into meaningful regions that respect object boundaries, allowing the algorithm to identify and repair artifacts within coherent regions while maintaining object integrity. The superpixel segmentation is particularly effective for handling irregular edges by grouping pixels into homogeneous regions.
Solution Approach 2:
The patent uses color image information as an intermediary to guide the repair process. The color image provides additional structural information that helps identify irregular edge regions and guides the weighted median filtering process, serving as a mediator between the noisy depth map and the final repaired output.
3Loss of information
If stereo matching is used in passive scanning to obtain depth maps, then more detailed depth information can be obtained, but irregular edges and black holes still exist
Solution Approach 1:
The patent merges multiple information sources including depth map data, color image information, and edge detection results to create a more robust repair process. By combining these different types of information, the algorithm can identify irregular edge regions more accurately and apply appropriate repairs that preserve the detailed depth information while correcting edge artifacts.
Data Source
AI summary
A method and a device for post-processing depth/disparity maps adopt a strategy of combining edge information and segmentation information when detecting irregular edge regions. The method includes dividing a color image into super pixels when performing image segmentation on a color image; partitioning a grayscale range into a preset number of intervals, and for each super pixel, statistically obtaining a histogram of all the pixel points that fall within the intervals; determining, in a current super pixel, whether a ratio of the number of pixels contained in the interval having a maximum interval distribution value, to the total number of pixels in the current super pixel is less than the first threshold; and if so, further dividing the current super pixel using a color-based segmentation method. The disclosed method and device improve accuracy of color image division while ensuring image processing speed, thus improving detection accuracy of the irregular edge regions.