Depth Image Denoising via Layer Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional depth image denoising methods fail to effectively remove noise connected to larger effective connectivity regions, leading to reduced denoising effectiveness and lower image quality.
Innovation Solution
A depth image denoising method that decomposes the original image into multiple layers, denoises each layer to eliminate isolated noise, and merges the denoised layers to remove noise connected to larger effective regions, thereby enhancing the denoising effect.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If conventional denoising method removes only isolated noises with smaller area, then isolated noises are eliminated, but noises connected to effective connectivity region of greater area remain, reducing denoising effect
Solution Approach 1:
The patent segments the depth image into multiple layers based on depth information, allowing different denoising strategies to be applied to different depth ranges. This segmentation enables the system to distinguish between isolated noises and noises connected to effective regions more effectively, resolving the contradiction by treating different noise types differently across depth layers.
Solution Approach 2:
The patent applies different denoising operations to different regions of the depth image based on local characteristics. By analyzing connectivity regions and their relationship with effective regions locally, the system can selectively remove noises while preserving connected effective structures, thereby improving both denoising precision and completeness.
2Productivity
If conventional denoising method directly removes isolated noises, then processing is simple and fast, but connected noises are not eliminated, lowering image quality
Solution Approach 1:
The patent performs preliminary actions by segmenting the depth image into layers and identifying connectivity regions before applying denoising operations. This preliminary analysis enables the system to efficiently distinguish between removable noises and preserved structures, maintaining processing speed while improving image quality through targeted denoising.
Solution Approach 2:
The patent introduces a new dimension by utilizing depth information to create layered representations of the depth image. This dimensional transformation allows the system to analyze noise connectivity in three-dimensional space, enabling more accurate distinction between isolated noises and connected effective regions without significantly increasing processing complexity.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The present disclosure discloses a depth image denoising method. In one embodiment, the depth image denoising method includes the following steps: decomposing an original depth image of a shot object into n layers of depth image, where n is an integer that is greater than or equal to two; denoising on each of the n layers of depth image, to eliminate isolated noise(s) in each of the n layers of depth image; and, merging the denoised n layers of depth image, to obtain a final denoised depth image. Correspondingly, the present disclosure also discloses a depth image denoising apparatus. With the depth image denoising method and/or denoising apparatus according to the present disclosure, not only isolated noise(s) in the original depth image can be eliminated, but also noise(s) which is/are connected to effective connectivity region of greater area can be eliminated effectively, thereby improving denoising effect.