Structural Texture Similarity for Material Recognition
| Autor: | Jue Lin, Thrasyvoulos N. Pappas |
|---|---|
| Rok vydání: | 2019 |
| Předmět: |
Computer science
business.industry Feature vector ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION Pattern recognition 02 engineering and technology 01 natural sciences Grayscale Texture (geology) Image (mathematics) 010309 optics Similarity (network science) Histogram Steerable filter 0103 physical sciences 0202 electrical engineering electronic engineering information engineering 020201 artificial intelligence & image processing Artificial intelligence Cluster analysis business ComputingMethodologies_COMPUTERGRAPHICS |
| Zdroj: | ICIP |
| DOI: | 10.1109/icip.2019.8803648 |
| Popis: | We propose a new direct approach for material recognition under diverse illumination and viewing conditions based on visual texture. We apply K-means clustering to feature vectors that consist of steerable filter subband statistics and dominant colors of each texture image in order to obtain a small number of exemplars characterizing each material. We then use structural texture similarity metrics and color composition metrics to compare a query texture to the exemplars for material classification. Experimental results using the CUReT database establish the importance of color and demonstrate that five exemplars per texture provide performance comparable to the state of the art. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|