Details
| Original language | English |
|---|---|
| Pages (from-to) | 87-94 |
| Number of pages | 8 |
| Journal | ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences |
| Volume | 4 |
| Issue number | 2/W7 |
| Early online date | 16 Sept 2019 |
| Publication status | Published - 2019 |
| Event | 1st Photogrammetric Image Analysis and Munich Remote Sensing Symposium, PIA 2019+MRSS 2019 - Munich, Germany Duration: 18 Sept 2019 → 20 Sept 2019 |
Abstract
We explore the use of semantic segmentation in Digital Terrain Models (DTMS) for detecting manmade landscape structures in archaeological sites. DTM data are stored and processed as large matrices of depth 1 as opposed to depth 3 in RGB images. The matrices usually contain continuous real-valued information upper bound of which is not fixed, such as distance or height from a reference surface. This is different from RGB images that contain integer values in a fixed range of 0 to 255. Additionally, RGB images are usually stored in smaller multidimensional matrices, and are more suitable as inputs for a neural network while the large DTMs are necessary to be split into smaller sub-matrices to be used by neural networks. Thus, while the spatial information of pixels in RGB images are important only locally within a single image, for DTM data, they are important locally, within a single sub-matrix processed for neural network, and also globally, in relation to the neighboring sub-matrices. To cope with the two differences, we apply min-max normalization to each input matrix fed to the neural network, and use a slightly modified version of DeepLabv3+ model for semantic segmentation. We show that with the architecture change, and the preprocessing, better results are achieved.
Keywords
- Deep Learning, Digital Terrain Models, Laser Scanning, Object Detection, Semantic Segmentation
ASJC Scopus subject areas
- Earth and Planetary Sciences(all)
- Earth and Planetary Sciences (miscellaneous)
- Environmental Science(all)
- Environmental Science (miscellaneous)
- Physics and Astronomy(all)
- Instrumentation
Cite this
- Standard
- Harvard
- Apa
- Vancouver
- BibTeX
- RIS
In: ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol. 4, No. 2/W7, 2019, p. 87-94.
Research output: Contribution to journal › Conference article › Research › peer review
}
TY - JOUR
T1 - Semantic segmentation of manmade landscape structures in digital terrain models
AU - Kazimi, B.
AU - Thiemann, F.
AU - Sester, M.
PY - 2019
Y1 - 2019
N2 - We explore the use of semantic segmentation in Digital Terrain Models (DTMS) for detecting manmade landscape structures in archaeological sites. DTM data are stored and processed as large matrices of depth 1 as opposed to depth 3 in RGB images. The matrices usually contain continuous real-valued information upper bound of which is not fixed, such as distance or height from a reference surface. This is different from RGB images that contain integer values in a fixed range of 0 to 255. Additionally, RGB images are usually stored in smaller multidimensional matrices, and are more suitable as inputs for a neural network while the large DTMs are necessary to be split into smaller sub-matrices to be used by neural networks. Thus, while the spatial information of pixels in RGB images are important only locally within a single image, for DTM data, they are important locally, within a single sub-matrix processed for neural network, and also globally, in relation to the neighboring sub-matrices. To cope with the two differences, we apply min-max normalization to each input matrix fed to the neural network, and use a slightly modified version of DeepLabv3+ model for semantic segmentation. We show that with the architecture change, and the preprocessing, better results are achieved.
AB - We explore the use of semantic segmentation in Digital Terrain Models (DTMS) for detecting manmade landscape structures in archaeological sites. DTM data are stored and processed as large matrices of depth 1 as opposed to depth 3 in RGB images. The matrices usually contain continuous real-valued information upper bound of which is not fixed, such as distance or height from a reference surface. This is different from RGB images that contain integer values in a fixed range of 0 to 255. Additionally, RGB images are usually stored in smaller multidimensional matrices, and are more suitable as inputs for a neural network while the large DTMs are necessary to be split into smaller sub-matrices to be used by neural networks. Thus, while the spatial information of pixels in RGB images are important only locally within a single image, for DTM data, they are important locally, within a single sub-matrix processed for neural network, and also globally, in relation to the neighboring sub-matrices. To cope with the two differences, we apply min-max normalization to each input matrix fed to the neural network, and use a slightly modified version of DeepLabv3+ model for semantic segmentation. We show that with the architecture change, and the preprocessing, better results are achieved.
KW - Deep Learning
KW - Digital Terrain Models
KW - Laser Scanning
KW - Object Detection
KW - Semantic Segmentation
UR - http://www.scopus.com/inward/record.url?scp=85084680018&partnerID=8YFLogxK
U2 - 10.5194/isprs-annals-IV-2-W7-87-2019
DO - 10.5194/isprs-annals-IV-2-W7-87-2019
M3 - Conference article
AN - SCOPUS:85084680018
VL - 4
SP - 87
EP - 94
JO - ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
JF - ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
SN - 2194-9042
IS - 2/W7
T2 - 1st Photogrammetric Image Analysis and Munich Remote Sensing Symposium, PIA 2019+MRSS 2019
Y2 - 18 September 2019 through 20 September 2019
ER -