Nombre: LUCAS DE ASSIS SOARES
Tipo: PhD thesis
Fecha de publicación: 23/10/2020
Supervisor:
Nombre |
Papel |
|---|---|
| EVANDRO OTTONI TEATINI SALLES | Advisor * |
| PATRICK MARQUES CIARELLI | Co-advisor * |
Junta de examinadores:
| Nombre |
Papel |
|---|---|
| CARLOS ALEXANDRE BARROS DE MELLO | External Examiner * |
| EVANDRO OTTONI TEATINI SALLES | Advisor * |
| KARIN SATIE KOMATI | External Examiner * |
| MARIO SARCINELLI FILHO | Internal Examiner * |
| PATRICK MARQUES CIARELLI | Co advisor * |
Páginas
Sumario: Texture segmentation is a challenging problem in computer vision due to the subjective nature of textures, the variability in which they occur in digital images, their dependence on scale and illumination variation, and the lack of a precise definition of textures in the scientific literature. This thesis proposes a method to segment textures through a binary pixel-wise classification, thereby without the need for a predefined number of textures classes. Using a convolutional neural network, with an encoderdecoder architecture, each pixel is classified as being inside an internal texture region or in a border between
two different textures. The network is trained using the Prague Texture Segmentation Datagenerator and Benchmark and tested using the same dataset, besides the Brodatz textures dataset, and the Describable Textures Dataset. The method is also evaluated on the separation of regions in images from different applications, namely remote sensing images and H&E-stained tissue images. It is shown that the presented method has a good performance on different test sets, can precisely identify borders between texture regions
and does not suffer from over-segmentation.
