.png){kind=icon}
.png){kind=icon}
Return
Image Classification via Hierarchical Dictionary Learning
.png){kind=icon}
PRE
AI
DOI：10.1109/ccdc.2019.8832839.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Recently lots of dictionary learning methods have been presented to deal with the issue of image classification, there is still substantial room for improvement. To further improve the performance, a novel Fisher discriminative criterion based structured dictionary learning method is here introduced which contains the main two components. Firstly, category-unique dictionary for all categories in different levels arc learned to obtain the specific information of each category, and category-shared dictionary between all categories are learned in different levels to obtain the common information across different categories; secondly, a discriminative term in terms of Fisher discrimination criterion is here jointly utilized to enhance the discrimination of category-unique dictionary and category-shared dictionary. Experiment results on public available databases are conducted to demonstrate the superiority of the proposed method compared with the state-of-the-art dictionary learning methods.
Keywords:
Structured Dictionary Learning
Fisher Discriminative Criterion
Image Classification
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
P
IF:
0
Papers:
1.8K
Citations:
0
Organization
No organization information available