Saliency Map Generation via Dual Similarity Integration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing salient region detection algorithms require previous knowledge or learning data, limiting their ability to evaluate saliency measures in images without pre-existing information, and are not effective for still images.
Innovation Solution
An image processing device and method that generates a saliency map by calculating two similarity maps based on different algorithms, one using dense recovery and the other sparse recovery, integrating these maps to accurately determine the saliency of image regions without prior knowledge, utilizing superpixel images and Bayesian estimation for improved accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a learning-based algorithm is used to detect salient regions, then the saliency measure can be closer to human sense, but previous knowledge or learning data must be prepared in advance
Solution Approach 1:
The algorithm uses the image data itself to generate the saliency map without requiring external learning data. By treating the input image as both the target and the reference (using marginal regions as templates), the system performs self-service detection that eliminates the need for separate learning phases while maintaining accuracy comparable to learning-based methods.
2Measurement precision
If a single algorithm is used to generate the similarity map, then the processing is simpler, but the detection precision is reduced
Solution Approach 1:
The patent combines multiple similarity map generation algorithms (dense recovery and sparse recovery) into a unified framework. By integrating the results from different algorithms through a combination formula, the system achieves higher detection precision while managing complexity through a structured merging approach that leverages the strengths of each individual algorithm.
Data Source
AI summary
An image processing device and a method performed by a computer, the image processing device comprising: a processing unit configured to operate as an image acquisition unit configured to acquire an image; a similarity map generator configured to calculate, with a region constructed with one or a plurality of pixels in the image as a constituent unit, a first similarity map based on a first algorithm, the first similarity map representing a degree of similarity between the region and a marginal region of the image, and calculate a second similarity map based on a second algorithm, the second similarity map representing a degree of similarity between the region and the marginal region of the image; and a saliency map generator configured to integrate the first similarity map and the second similarity map to generate a saliency map.