Method for Constructing Texture Features based on an Image Weight Model
| Autor: | Mikhail P. Shleymovich, S. A. Lyasheva, Maya M. Lyasheva, Zinnur M. Gizatullin |
|---|---|
| Rok vydání: | 2021 |
| Předmět: |
Discrete wavelet transform
Brightness Pixel business.industry ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION Pattern recognition Texture (geology) Image (mathematics) Set (abstract data type) Wavelet Computer Science::Computer Vision and Pattern Recognition Artificial intelligence business Energy (signal processing) ComputingMethodologies_COMPUTERGRAPHICS Mathematics |
| Zdroj: | 2021 IEEE Conference of Russian Young Researchers in Electrical and Electronic Engineering (ElConRus). |
| DOI: | 10.1109/elconrus51938.2021.9396549 |
| Popis: | The article describes a statistical method for constructing texture features based on an image weight model. The weight model characterizes the pixels of an image in terms of describing its texture. It is a set of weight values that reflect the significance of the change in brightness in the pixels of the image. A relatively large amount of brightness difference in a pixel characterizes it as a pixel belonging to an edge. The density of edges in different areas of the image, in turn, characterizes the texture of the image as a whole. Thus, weights can also be used to characterize the texture of the image. To construct a weighting model, an algorithm based on the analysis of the energy of the coefficients of the discrete wavelet transform is proposed. The results of comparison of test images using the obtained texture features are presented. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|