Deep Learning Applied to Vegetation Identification and Removal Using Multidimensional Aerial Data
When performing structural inspection, the generation of three-dimensional (3D) point clouds is a common resource. Those are usually generated from photogrammetry or through laser scan techniques. However, a significant drawback for complete inspection is the presence of covering vegetation, hiding...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-10-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/20/21/6187 |
| _version_ | 1797549287344701440 |
|---|---|
| author | Milena F. Pinto Aurelio G. Melo Leonardo M. Honório André L. M. Marcato André G. S. Conceição Amanda O. Timotheo |
| author_facet | Milena F. Pinto Aurelio G. Melo Leonardo M. Honório André L. M. Marcato André G. S. Conceição Amanda O. Timotheo |
| author_sort | Milena F. Pinto |
| collection | DOAJ |
| description | When performing structural inspection, the generation of three-dimensional (3D) point clouds is a common resource. Those are usually generated from photogrammetry or through laser scan techniques. However, a significant drawback for complete inspection is the presence of covering vegetation, hiding possible structural problems, and making difficult the acquisition of proper object surfaces in order to provide a reliable diagnostic. Therefore, this research’s main contribution is developing an effective vegetation removal methodology through the use of a deep learning structure that is capable of identifying and extracting covering vegetation in 3D point clouds. The proposed approach uses pre and post-processing filtering stages that take advantage of colored point clouds, if they are available, or operate independently. The results showed high classification accuracy and good effectiveness when compared with similar methods in the literature. After this step, if color is available, then a color filter is applied, enhancing the results obtained. Besides, the results are analyzed in light of real Structure From Motion (SFM) reconstruction data, which further validates the proposed method. This research also presented a colored point cloud library of bushes built for the work used by other studies in the field. |
| first_indexed | 2024-03-10T15:12:32Z |
| format | Article |
| id | doaj.art-039c565557aa4283b48c8759f52d9638 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-10T15:12:32Z |
| publishDate | 2020-10-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-039c565557aa4283b48c8759f52d96382023-11-20T19:10:01ZengMDPI AGSensors1424-82202020-10-012021618710.3390/s20216187Deep Learning Applied to Vegetation Identification and Removal Using Multidimensional Aerial DataMilena F. Pinto0Aurelio G. Melo1Leonardo M. Honório2André L. M. Marcato3André G. S. Conceição4Amanda O. Timotheo5Department of Electronics Engineering, Federal Center for Technological Education of Rio de Janeiro, Rio de Janeiro 20260-100, BrazilDepartment of Electrical Engineering, Federal University of Juiz de Fora, Juiz de Fora 36073-120, BrazilDepartment of Electrical Engineering, Federal University of Juiz de Fora, Juiz de Fora 36073-120, BrazilDepartment of Electrical Engineering, Federal University of Juiz de Fora, Juiz de Fora 36073-120, BrazilDepartment of Electrical Engineering, Federal University of Bahia, Salvador 40210-630, BrazilDepartment of Electrical Engineering, Federal University of Juiz de Fora, Juiz de Fora 36073-120, BrazilWhen performing structural inspection, the generation of three-dimensional (3D) point clouds is a common resource. Those are usually generated from photogrammetry or through laser scan techniques. However, a significant drawback for complete inspection is the presence of covering vegetation, hiding possible structural problems, and making difficult the acquisition of proper object surfaces in order to provide a reliable diagnostic. Therefore, this research’s main contribution is developing an effective vegetation removal methodology through the use of a deep learning structure that is capable of identifying and extracting covering vegetation in 3D point clouds. The proposed approach uses pre and post-processing filtering stages that take advantage of colored point clouds, if they are available, or operate independently. The results showed high classification accuracy and good effectiveness when compared with similar methods in the literature. After this step, if color is available, then a color filter is applied, enhancing the results obtained. Besides, the results are analyzed in light of real Structure From Motion (SFM) reconstruction data, which further validates the proposed method. This research also presented a colored point cloud library of bushes built for the work used by other studies in the field.https://www.mdpi.com/1424-8220/20/21/6187vegetation identification/recognition3D point clouddeep learningUnmanned Aerial Vehiclesstructural analyzes |
| spellingShingle | Milena F. Pinto Aurelio G. Melo Leonardo M. Honório André L. M. Marcato André G. S. Conceição Amanda O. Timotheo Deep Learning Applied to Vegetation Identification and Removal Using Multidimensional Aerial Data Sensors vegetation identification/recognition 3D point cloud deep learning Unmanned Aerial Vehicles structural analyzes |
| title | Deep Learning Applied to Vegetation Identification and Removal Using Multidimensional Aerial Data |
| title_full | Deep Learning Applied to Vegetation Identification and Removal Using Multidimensional Aerial Data |
| title_fullStr | Deep Learning Applied to Vegetation Identification and Removal Using Multidimensional Aerial Data |
| title_full_unstemmed | Deep Learning Applied to Vegetation Identification and Removal Using Multidimensional Aerial Data |
| title_short | Deep Learning Applied to Vegetation Identification and Removal Using Multidimensional Aerial Data |
| title_sort | deep learning applied to vegetation identification and removal using multidimensional aerial data |
| topic | vegetation identification/recognition 3D point cloud deep learning Unmanned Aerial Vehicles structural analyzes |
| url | https://www.mdpi.com/1424-8220/20/21/6187 |
| work_keys_str_mv | AT milenafpinto deeplearningappliedtovegetationidentificationandremovalusingmultidimensionalaerialdata AT aureliogmelo deeplearningappliedtovegetationidentificationandremovalusingmultidimensionalaerialdata AT leonardomhonorio deeplearningappliedtovegetationidentificationandremovalusingmultidimensionalaerialdata AT andrelmmarcato deeplearningappliedtovegetationidentificationandremovalusingmultidimensionalaerialdata AT andregsconceicao deeplearningappliedtovegetationidentificationandremovalusingmultidimensionalaerialdata AT amandaotimotheo deeplearningappliedtovegetationidentificationandremovalusingmultidimensionalaerialdata |