Automated Mapping of Cropland Boundaries Using Deep Neural Networks
Accurately identifying the boundaries of agricultural land is critical to the effective management of its resources. This includes the determination of property and land rights, the prevention of non-agricultural activities on agricultural land, and the effective management of natural resources. The...
| Main Author: | |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-09-01
|
| Series: | AgriEngineering |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2624-7402/5/3/97 |
| _version_ | 1797581718590324736 |
|---|---|
| author | Artur Gafurov |
| author_facet | Artur Gafurov |
| author_sort | Artur Gafurov |
| collection | DOAJ |
| description | Accurately identifying the boundaries of agricultural land is critical to the effective management of its resources. This includes the determination of property and land rights, the prevention of non-agricultural activities on agricultural land, and the effective management of natural resources. There are various methods for accurate boundary detection, including traditional measurement methods and remote sensing, and the choice of the best method depends on specific objectives and conditions. This paper proposes the use of convolutional neural networks (CNNs) as an efficient and effective tool for the automatic recognition of agricultural land boundaries. The objective of this research paper is to develop an automated method for the recognition of agricultural land boundaries using deep neural networks and Sentinel 2 multispectral imagery. The Buinsky district of the Republic of Tatarstan, Russia, which is known to be an agricultural region, was chosen for this study because of the importance of the accurate detection of its agricultural land boundaries. Linknet, a deep neural network architecture with skip connections between encoder and decoder, was used for semantic segmentation to extract arable land boundaries, and transfer learning using a pre-trained EfficientNetB3 model was used to improve performance. The Linknet + EfficientNetB3 combination for semantic segmentation achieved an accuracy of 86.3% and an f1 measure of 0.924 on the validation sample. The results showed a high degree of agreement between the predicted field boundaries and the expert-validated boundaries. According to the results, the advantages of the method include its speed, scalability, and ability to detect patterns outside the study area. It is planned to improve the method by using different neural network architectures and prior recognized land use classes. |
| first_indexed | 2024-03-10T23:08:36Z |
| format | Article |
| id | doaj.art-7f2d472f96f64141bc014b322eda9dca |
| institution | Directory Open Access Journal |
| issn | 2624-7402 |
| language | English |
| last_indexed | 2024-03-10T23:08:36Z |
| publishDate | 2023-09-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | AgriEngineering |
| spelling | doaj.art-7f2d472f96f64141bc014b322eda9dca2023-11-19T09:08:38ZengMDPI AGAgriEngineering2624-74022023-09-01531568158010.3390/agriengineering5030097Automated Mapping of Cropland Boundaries Using Deep Neural NetworksArtur Gafurov0Institute of Environmental Sciences, Kazan Federal University, 420097 Kazan, RussiaAccurately identifying the boundaries of agricultural land is critical to the effective management of its resources. This includes the determination of property and land rights, the prevention of non-agricultural activities on agricultural land, and the effective management of natural resources. There are various methods for accurate boundary detection, including traditional measurement methods and remote sensing, and the choice of the best method depends on specific objectives and conditions. This paper proposes the use of convolutional neural networks (CNNs) as an efficient and effective tool for the automatic recognition of agricultural land boundaries. The objective of this research paper is to develop an automated method for the recognition of agricultural land boundaries using deep neural networks and Sentinel 2 multispectral imagery. The Buinsky district of the Republic of Tatarstan, Russia, which is known to be an agricultural region, was chosen for this study because of the importance of the accurate detection of its agricultural land boundaries. Linknet, a deep neural network architecture with skip connections between encoder and decoder, was used for semantic segmentation to extract arable land boundaries, and transfer learning using a pre-trained EfficientNetB3 model was used to improve performance. The Linknet + EfficientNetB3 combination for semantic segmentation achieved an accuracy of 86.3% and an f1 measure of 0.924 on the validation sample. The results showed a high degree of agreement between the predicted field boundaries and the expert-validated boundaries. According to the results, the advantages of the method include its speed, scalability, and ability to detect patterns outside the study area. It is planned to improve the method by using different neural network architectures and prior recognized land use classes.https://www.mdpi.com/2624-7402/5/3/97remote sensing imageryneural networkarable land boundariesSentinel 2 imagesgeospatial datamachine learning |
| spellingShingle | Artur Gafurov Automated Mapping of Cropland Boundaries Using Deep Neural Networks AgriEngineering remote sensing imagery neural network arable land boundaries Sentinel 2 images geospatial data machine learning |
| title | Automated Mapping of Cropland Boundaries Using Deep Neural Networks |
| title_full | Automated Mapping of Cropland Boundaries Using Deep Neural Networks |
| title_fullStr | Automated Mapping of Cropland Boundaries Using Deep Neural Networks |
| title_full_unstemmed | Automated Mapping of Cropland Boundaries Using Deep Neural Networks |
| title_short | Automated Mapping of Cropland Boundaries Using Deep Neural Networks |
| title_sort | automated mapping of cropland boundaries using deep neural networks |
| topic | remote sensing imagery neural network arable land boundaries Sentinel 2 images geospatial data machine learning |
| url | https://www.mdpi.com/2624-7402/5/3/97 |
| work_keys_str_mv | AT arturgafurov automatedmappingofcroplandboundariesusingdeepneuralnetworks |