UAV Landing Using Computer Vision Techniques for Human Detection
The capability of drones to perform autonomous missions has led retail companies to use them for deliveries, saving time and human resources. In these services, the delivery depends on the Global Positioning System (GPS) to define an approximate landing point. However, the landscape can interfere wi...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-01-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/20/3/613 |
| _version_ | 1828120059539619840 |
|---|---|
| author | David Safadinho João Ramos Roberto Ribeiro Vítor Filipe João Barroso António Pereira |
| author_facet | David Safadinho João Ramos Roberto Ribeiro Vítor Filipe João Barroso António Pereira |
| author_sort | David Safadinho |
| collection | DOAJ |
| description | The capability of drones to perform autonomous missions has led retail companies to use them for deliveries, saving time and human resources. In these services, the delivery depends on the Global Positioning System (GPS) to define an approximate landing point. However, the landscape can interfere with the satellite signal (e.g., tall buildings), reducing the accuracy of this approach. Changes in the environment can also invalidate the security of a previously defined landing site (e.g., irregular terrain, swimming pool). Therefore, the main goal of this work is to improve the process of goods delivery using drones, focusing on the detection of the potential receiver. We developed a solution that has been improved along its iterative assessment composed of five test scenarios. The built prototype complements the GPS through Computer Vision (CV) algorithms, based on Convolutional Neural Networks (CNN), running in a Raspberry Pi 3 with a Pi NoIR Camera (i.e., No InfraRed—without infrared filter). The experiments were performed with the models Single Shot Detector (SSD) MobileNet-V2, and SSDLite-MobileNet-V2. The best results were obtained in the afternoon, with the SSDLite architecture, for distances and heights between 2.5–10 m, with recalls from 59%–76%. The results confirm that a low computing power and cost-effective system can perform aerial human detection, estimating the landing position without an additional visual marker. |
| first_indexed | 2024-04-11T13:59:11Z |
| format | Article |
| id | doaj.art-43c11984af4541bca97a51952ef22d4b |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-04-11T13:59:11Z |
| publishDate | 2020-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-43c11984af4541bca97a51952ef22d4b2022-12-22T04:20:11ZengMDPI AGSensors1424-82202020-01-0120361310.3390/s20030613s20030613UAV Landing Using Computer Vision Techniques for Human DetectionDavid Safadinho0João Ramos1Roberto Ribeiro2Vítor Filipe3João Barroso4António Pereira5School of Technology and Management, Computer Science and Communication Research Centre, Polytechnic Institute of Leiria, Campus 2, Morro do Lena – Alto do Vieiro, Apartado 4163, 2411-901 Leiria, PortugalSchool of Technology and Management, Computer Science and Communication Research Centre, Polytechnic Institute of Leiria, Campus 2, Morro do Lena – Alto do Vieiro, Apartado 4163, 2411-901 Leiria, PortugalSchool of Technology and Management, Computer Science and Communication Research Centre, Polytechnic Institute of Leiria, Campus 2, Morro do Lena – Alto do Vieiro, Apartado 4163, 2411-901 Leiria, PortugalINESC TEC and University of Trás-os-Montes e Alto Douro, Quinta de Prados, 5001-801 Vila Real, PortugalINESC TEC and University of Trás-os-Montes e Alto Douro, Quinta de Prados, 5001-801 Vila Real, PortugalSchool of Technology and Management, Computer Science and Communication Research Centre, Polytechnic Institute of Leiria, Campus 2, Morro do Lena – Alto do Vieiro, Apartado 4163, 2411-901 Leiria, PortugalThe capability of drones to perform autonomous missions has led retail companies to use them for deliveries, saving time and human resources. In these services, the delivery depends on the Global Positioning System (GPS) to define an approximate landing point. However, the landscape can interfere with the satellite signal (e.g., tall buildings), reducing the accuracy of this approach. Changes in the environment can also invalidate the security of a previously defined landing site (e.g., irregular terrain, swimming pool). Therefore, the main goal of this work is to improve the process of goods delivery using drones, focusing on the detection of the potential receiver. We developed a solution that has been improved along its iterative assessment composed of five test scenarios. The built prototype complements the GPS through Computer Vision (CV) algorithms, based on Convolutional Neural Networks (CNN), running in a Raspberry Pi 3 with a Pi NoIR Camera (i.e., No InfraRed—without infrared filter). The experiments were performed with the models Single Shot Detector (SSD) MobileNet-V2, and SSDLite-MobileNet-V2. The best results were obtained in the afternoon, with the SSDLite architecture, for distances and heights between 2.5–10 m, with recalls from 59%–76%. The results confirm that a low computing power and cost-effective system can perform aerial human detection, estimating the landing position without an additional visual marker.https://www.mdpi.com/1424-8220/20/3/613autonomous deliverycomputer visiondeep neural networksintelligent vehiclesinternet of thingsnext generation servicesreal-time systemsremote sensingunmanned aerial vehiclesunmanned aircraft systems |
| spellingShingle | David Safadinho João Ramos Roberto Ribeiro Vítor Filipe João Barroso António Pereira UAV Landing Using Computer Vision Techniques for Human Detection Sensors autonomous delivery computer vision deep neural networks intelligent vehicles internet of things next generation services real-time systems remote sensing unmanned aerial vehicles unmanned aircraft systems |
| title | UAV Landing Using Computer Vision Techniques for Human Detection |
| title_full | UAV Landing Using Computer Vision Techniques for Human Detection |
| title_fullStr | UAV Landing Using Computer Vision Techniques for Human Detection |
| title_full_unstemmed | UAV Landing Using Computer Vision Techniques for Human Detection |
| title_short | UAV Landing Using Computer Vision Techniques for Human Detection |
| title_sort | uav landing using computer vision techniques for human detection |
| topic | autonomous delivery computer vision deep neural networks intelligent vehicles internet of things next generation services real-time systems remote sensing unmanned aerial vehicles unmanned aircraft systems |
| url | https://www.mdpi.com/1424-8220/20/3/613 |
| work_keys_str_mv | AT davidsafadinho uavlandingusingcomputervisiontechniquesforhumandetection AT joaoramos uavlandingusingcomputervisiontechniquesforhumandetection AT robertoribeiro uavlandingusingcomputervisiontechniquesforhumandetection AT vitorfilipe uavlandingusingcomputervisiontechniquesforhumandetection AT joaobarroso uavlandingusingcomputervisiontechniquesforhumandetection AT antoniopereira uavlandingusingcomputervisiontechniquesforhumandetection |