A Fusion Method of Multi-label Image Annotation Results Based on Rank Minimization
A technology of image annotation and fusion method, applied in the field of fusion of image annotation results, which can solve the problems of different fusion effects and increase the complexity of decision-making fusion
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment Construction
[0048] Refer to attached figure 1 And attached figure 2 , taking the color histogram, Gabor texture and local binary pattern features as examples, the multi-label result fusion method of image annotation based on rank minimization includes the following steps:
[0049] 1) Extract various feature representations of images in the training set, and each image in the training set has semantic tags given in advance;
[0050] 2) Under different feature representations, train their respective supervised learning image annotation models;
[0051] 3) For a new image without semantic tagging words, use the same method to extract multiple feature representations of the image, and use these feature representations to input to the corresponding supervised learning image tagging model to predict its multi-label tagging results;
[0052] 4) Use the rank minimization optimization algorithm to fuse the multi-label results output by multiple models: for the result vectors predicted by the mo...
PUM
Abstract
Description
Claims
Application Information
- R&D Engineer
- R&D Manager
- IP Professional
- Industry Leading Data Capabilities
- Powerful AI technology
- Patent DNA Extraction
Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.
© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com