Publicação
Automation of road feature extraction from high resolution images
| Resumo: | The detection of road features from remotely sensed images has become a critical factor in maintaining a reliable and updated road network in a country to provide a base reference for transportation, emergency planning, and navigation. With the recent advances of convolutional neural networks in image processing, several publications are devoted to the development of a method for automatically extract roads from satellite images. However, a reliable feature extraction method has not yet been developed with the desired accuracy and precision, and always seems to be a proportionality between the accuracy and the complexity of these developed methods. The aim of this study was therefore to develop an accurate road extraction method without compromising computational efficiency. In this paper, a semantic segmentation neural network that combines the strengths of transfer learning and U-net architecture is proposed with a minimal network complexity. Further, post-processing based on morphological operations and regional properties of the extracted segments were used to remove the noises from the final output. The results have been compared with different automatic classification and segmentation methods and the results of the proposed method produced an F1 score of 0.83 and high accuracy of 95.57%, more accurate and precise than all the other models for the freely available Massachusetts dataset. Finally, the developed method stood superior to the preexisting methods in terms of performance measure and network complexity. |
|---|---|
| Autores principais: | Heva, Prasadi Thilanka Senadeera Kanda Uda |
| Assunto: | Convolutional neural networks U-Net Image Segmentation architecture Road extraction Transfer learning Morphological operations |
| Ano: | 2021 |
| País: | Portugal |
| Tipo de documento: | dissertação de mestrado |
| Tipo de acesso: | acesso aberto |
| Instituição associada: | Universidade Nova de Lisboa |
| Idioma: | inglês |
| Origem: | Repositório Institucional da UNL |
| Resumo: | The detection of road features from remotely sensed images has become a critical factor in maintaining a reliable and updated road network in a country to provide a base reference for transportation, emergency planning, and navigation. With the recent advances of convolutional neural networks in image processing, several publications are devoted to the development of a method for automatically extract roads from satellite images. However, a reliable feature extraction method has not yet been developed with the desired accuracy and precision, and always seems to be a proportionality between the accuracy and the complexity of these developed methods. The aim of this study was therefore to develop an accurate road extraction method without compromising computational efficiency. In this paper, a semantic segmentation neural network that combines the strengths of transfer learning and U-net architecture is proposed with a minimal network complexity. Further, post-processing based on morphological operations and regional properties of the extracted segments were used to remove the noises from the final output. The results have been compared with different automatic classification and segmentation methods and the results of the proposed method produced an F1 score of 0.83 and high accuracy of 95.57%, more accurate and precise than all the other models for the freely available Massachusetts dataset. Finally, the developed method stood superior to the preexisting methods in terms of performance measure and network complexity. |
|---|