Technical NotePublished
Tópicos sobre Protocolos de Comunicação
Anderson Alves de Albuquerque; Bruno Marques Amaral; Fadul da Costa Bento; Fernanda Santoro Jannuzzi; Fernando Spencer; Marcelo Eller de Amorim; Marcelo Ruela Felippe; M. Maestrelli
2003
Publication details
- Authors
- Anderson Alves de Albuquerque; Bruno Marques Amaral; Fadul da Costa Bento; Fernanda Santoro Jannuzzi; Fernando Spencer; Marcelo Eller de Amorim; Marcelo Ruela Felippe; M. Maestrelli
- Type
- Technical Note
- Status
- Published
- Publication date
- 2003
- Description
- Image analysis usually refers to processing of images with the goal of finding objects in the image. Image segmentation is one of the most critical tasks in automatic image analysis. The non-extensive entropy is a recent development in statistical mechanics and has been considered as a useful measure in describing termostatistical properties of physical systems. In this formalism a real quantity q was introduced as parameter for physical systems that presents long range interactions, long time memories and fractal-type structures. In image processing, one of the most efficient techniques for image segmentation is entropy-based thresholding. This approach uses relative entropy from the information theory considering the gray level image histogram and a theoretical one as probabilities distributions. In this work, we applied non-extensive entropy as formalism for. It was proposed an image thresholding method using relative entropy basis. Some typical information theory results using this method are presented and illustrate the influence of the q in the thresholding process.
- Year
- 2003
- Number
- CBPF-NT-006/03
- Linked to a thesis
- No
- Linked to a grant
- No