Super-Resolution Reconstruction of Heterogeneous Stereoscopic Images
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for super-resolution reconstruction of stereo images fail to effectively handle heterogeneous stereoscopic images compressed using asymmetric methods, leading to difficulties in 3D reconstruction and information fusion due to the disconnect between left and right views.
Innovation Solution
A super-resolution reconstruction method that utilizes spatial and viewpoint correlations between high-resolution and low-resolution images by narrowing search areas through edge detection and selective patch processing, followed by parametric transformation to paste high-frequency information from high-resolution to low-resolution images, ensuring robustness and universality without extensive neural network training.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If asymmetric compression method is used to compress stereoscopic images, then transmission and storage costs are reduced, but heterogeneity between left and right images is introduced
Solution Approach 1:
The patent transforms the low-resolution image to match the resolution parameters of the high-resolution image through super-resolution reconstruction. This involves changing the resolution parameter of one image to achieve parameter consistency between the pair, thereby resolving the heterogeneity introduced by asymmetric compression while maintaining the compression benefits.
Solution Approach 2:
The patent copies high-frequency information from the high-resolution image to the low-resolution image through patch-based reconstruction. By copying detailed information from one view to reconstruct the other, the method restores homogeneity without requiring both original high-resolution images, thus maintaining compression efficiency.
2Manufacturing precision
If single-image super-resolution reconstruction algorithm is used to process stereo images, then image quality is improved, but the connection between left and right views is destroyed
Solution Approach 1:
The patent merges the advantages of single-image super-resolution with stereo image relationships by combining the high-resolution image and low-resolution image into a unified reconstruction framework. The method uses patches from both images and their corresponding relationships to reconstruct the final high-resolution stereo pair, preserving view connections while improving quality.
Solution Approach 2:
The patent introduces corresponding patch relationships as an intermediary between the high-resolution and low-resolution images. This intermediary mechanism uses the spatial and semantic correspondence between patches in different views to guide the reconstruction process, ensuring that the connection between views is maintained while achieving super-resolution.
3Device complexity
If conventional super-resolution methods are used for heterogeneous stereoscopic images, then reconstruction is simplified, but recovery of high-frequency information is insufficient
Solution Approach 1:
The patent segments the image into multiple patches and processes each patch independently through corresponding relationships. By dividing the reconstruction task into smaller patch-level operations, the method efficiently recovers high-frequency information in each local region while maintaining overall simplicity through the modular patch-based approach.
Solution Approach 2:
The patent applies different reconstruction strategies to different local regions (patches) based on their corresponding relationships. High-frequency information is selectively recovered in regions where correspondence is strong, while maintaining simplicity in regions where correspondence is weak. This local quality approach optimizes information recovery without uniformly increasing complexity.
Data Source
Figure 1~2
AI summary
The invention discloses a method and system for super-resolution reconstruction of heterogeneous stereoscopic images, which comprises :(1) acquires a stereoscopic image pair at a certain moment compressed by an asymmetrical compression method in a binocular vision system, and the stereoscopic image pair includes two views, one of which is a high-resolution image and the other is a low-resolution image; (2) narrows the search area of the high-resolution image and the low-resolution image; (3) determines the high-frequency information of the high-resolution image and the low-resolution image from the search regions of the high-resolution image and the low-resolution image, respectively;(4) pastes the high-frequency information of the high-resolution image to the high-frequency information of the low-resolution image through parametric transformation. The advantages are :(1) the prior knowledge of high-resolution images in heterogeneous stereo images can be directly utilized to recover the high-frequency information of the low-resolution images; (2) the spatial relativity and viewpoint relativity of heterogeneous stereoscopic images can be used for finding the corresponding sub-blocks for the low-resolution images from the high-resolution images to efficiently compensate the heterogeneity between stereoscopic images.