Two Dimensional Time Series for Textures
| Autor: | J. P. Gambotto |
|---|---|
| Rok vydání: | 1981 |
| Předmět: |
Series (mathematics)
Computer science business.industry ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION Inverse Pattern recognition Image (mathematics) Autoregressive model Computer Science::Computer Vision and Pattern Recognition Segmentation Artificial intelligence Reference Region business Texture synthesis |
| Zdroj: | Digital Image Processing ISBN: 9789400985452 |
| DOI: | 10.1007/978-94-009-8543-8_13 |
| Popis: | This paper presents an approach to synthesis, analysis and recognition of textures based on two dimensional time series. In this approach, two dimensional autoregressive models are used to generate textures having given statistical properties. Some new estimation techniques are proposed, and used to fit an autoregressive model to a reference region in an image. The interest of a recognition procedure based on two dimensional inverse filtering method is then emphasized. This procedure is applied to natural pictures for classification and segmentation using the second order statistical properties of the textures only. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|