Deep automatic soil roughness estimation from digital images
Soil roughness, defined as the irregularities of the soil surface, yields significant information about soil water storage, infiltration and overland flow and, thus, is a key factor in characterizing the quality of the terrain; it is often used as input in many synthetic general agricultural models...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Taylor & Francis Group
2024-12-01
|
| Series: | European Journal of Remote Sensing |
| Subjects: | |
| Online Access: | https://www.tandfonline.com/doi/10.1080/22797254.2024.2342955 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850127141980602368 |
|---|---|
| author | M. Ivanovici S. Popa K. Marandskiy C. Florea |
| author_facet | M. Ivanovici S. Popa K. Marandskiy C. Florea |
| author_sort | M. Ivanovici |
| collection | DOAJ |
| description | Soil roughness, defined as the irregularities of the soil surface, yields significant information about soil water storage, infiltration and overland flow and, thus, is a key factor in characterizing the quality of the terrain; it is often used as input in many synthetic general agricultural models and in particular in soil moisture estimation models. In this paper, we propose a framework that combines a specific setup for data acquisition with deep convolutional networks for actual estimation. The former relies on projecting a line red laser beam on the analysed soil surface followed by digital color image acquisition. The later, involves two convolutional models that are trained in a supervised manner to predict the soil roughness. The data set was produced in the laboratory both on synthetic and real soil samples. The labels used in the training process are the soil roughness values measured by using a pinboard. The detailed evaluation showed that the error of the automatic precision lies in the range of ground truth deviation, thus validating the proposed procedure. |
| format | Article |
| id | doaj-art-b02f439002f54f79bbc3ff00b919b152 |
| institution | OA Journals |
| issn | 2279-7254 |
| language | English |
| publishDate | 2024-12-01 |
| publisher | Taylor & Francis Group |
| record_format | Article |
| series | European Journal of Remote Sensing |
| spelling | doaj-art-b02f439002f54f79bbc3ff00b919b1522025-08-20T02:33:44ZengTaylor & Francis GroupEuropean Journal of Remote Sensing2279-72542024-12-0157110.1080/22797254.2024.2342955Deep automatic soil roughness estimation from digital imagesM. Ivanovici0S. Popa1K. Marandskiy2C. Florea3MIV Laboratory, Transilvania University of Brasov, Brasov, RomaniaMIV Laboratory, Transilvania University of Brasov, Brasov, RomaniaMIV Laboratory, Transilvania University of Brasov, Brasov, RomaniaMIV Laboratory, Transilvania University of Brasov, Brasov, RomaniaSoil roughness, defined as the irregularities of the soil surface, yields significant information about soil water storage, infiltration and overland flow and, thus, is a key factor in characterizing the quality of the terrain; it is often used as input in many synthetic general agricultural models and in particular in soil moisture estimation models. In this paper, we propose a framework that combines a specific setup for data acquisition with deep convolutional networks for actual estimation. The former relies on projecting a line red laser beam on the analysed soil surface followed by digital color image acquisition. The later, involves two convolutional models that are trained in a supervised manner to predict the soil roughness. The data set was produced in the laboratory both on synthetic and real soil samples. The labels used in the training process are the soil roughness values measured by using a pinboard. The detailed evaluation showed that the error of the automatic precision lies in the range of ground truth deviation, thus validating the proposed procedure.https://www.tandfonline.com/doi/10.1080/22797254.2024.2342955Soil roughnessconvolutional neural networksVGG-11ResNet-18 |
| spellingShingle | M. Ivanovici S. Popa K. Marandskiy C. Florea Deep automatic soil roughness estimation from digital images European Journal of Remote Sensing Soil roughness convolutional neural networks VGG-11 ResNet-18 |
| title | Deep automatic soil roughness estimation from digital images |
| title_full | Deep automatic soil roughness estimation from digital images |
| title_fullStr | Deep automatic soil roughness estimation from digital images |
| title_full_unstemmed | Deep automatic soil roughness estimation from digital images |
| title_short | Deep automatic soil roughness estimation from digital images |
| title_sort | deep automatic soil roughness estimation from digital images |
| topic | Soil roughness convolutional neural networks VGG-11 ResNet-18 |
| url | https://www.tandfonline.com/doi/10.1080/22797254.2024.2342955 |
| work_keys_str_mv | AT mivanovici deepautomaticsoilroughnessestimationfromdigitalimages AT spopa deepautomaticsoilroughnessestimationfromdigitalimages AT kmarandskiy deepautomaticsoilroughnessestimationfromdigitalimages AT cflorea deepautomaticsoilroughnessestimationfromdigitalimages |