Litter Detection with Deep Learning: A Comparative Study
Pollution in the form of litter in the natural environment is one of the great challenges of our times. Automated litter detection can help assess waste occurrences in the environment. Different machine learning solutions have been explored to develop litter detection tools, thereby supporting resea...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-01-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/22/2/548 |
| _version_ | 1797490571948851200 |
|---|---|
| author | Manuel Córdova Allan Pinto Christina Carrozzo Hellevik Saleh Abdel-Afou Alaliyat Ibrahim A. Hameed Helio Pedrini Ricardo da S. Torres |
| author_facet | Manuel Córdova Allan Pinto Christina Carrozzo Hellevik Saleh Abdel-Afou Alaliyat Ibrahim A. Hameed Helio Pedrini Ricardo da S. Torres |
| author_sort | Manuel Córdova |
| collection | DOAJ |
| description | Pollution in the form of litter in the natural environment is one of the great challenges of our times. Automated litter detection can help assess waste occurrences in the environment. Different machine learning solutions have been explored to develop litter detection tools, thereby supporting research, citizen science, and volunteer clean-up initiatives. However, to the best of our knowledge, no work has investigated the performance of state-of-the-art deep learning object detection approaches in the context of litter detection. In particular, no studies have focused on the assessment of those methods aiming their use in devices with low processing capabilities, e.g., mobile phones, typically employed in citizen science activities. In this paper, we fill this literature gap. We performed a comparative study involving state-of-the-art CNN architectures (e.g., Faster RCNN, Mask-RCNN, EfficientDet, RetinaNet and YOLO-v5), two litter image datasets and a smartphone. We also introduce a new dataset for litter detection, named PlastOPol, composed of 2418 images and 5300 annotations. The experimental results demonstrate that object detectors based on the YOLO family are promising for the construction of litter detection solutions, with superior performance in terms of detection accuracy, processing time, and memory footprint. |
| first_indexed | 2024-03-10T00:33:55Z |
| format | Article |
| id | doaj.art-e9cdd1d694184cce95fda270c6f69569 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-10T00:33:55Z |
| publishDate | 2022-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-e9cdd1d694184cce95fda270c6f695692023-11-23T15:20:29ZengMDPI AGSensors1424-82202022-01-0122254810.3390/s22020548Litter Detection with Deep Learning: A Comparative StudyManuel Córdova0Allan Pinto1Christina Carrozzo Hellevik2Saleh Abdel-Afou Alaliyat3Ibrahim A. Hameed4Helio Pedrini5Ricardo da S. Torres6Institute of Computing, University of Campinas, Avenue Albert Einstein, Campinas 13083-852, BrazilBrazilian Center for Research in Energy and Materials (CNPEM), Brazilian Synchrotron Light Laboratory (LNLS), Campinas 13083-100, BrazilDepartment of International Business, NTNU—Norwegian University of Science and Technology, Larsgårdsvegen 2, 6009 Alesund, NorwayDepartment of ICT and Natural Sciences, NTNU—Norwegian University of Science and Technology, Larsgårdsvegen 2, 6009 Alesund, NorwayDepartment of ICT and Natural Sciences, NTNU—Norwegian University of Science and Technology, Larsgårdsvegen 2, 6009 Alesund, NorwayInstitute of Computing, University of Campinas, Avenue Albert Einstein, Campinas 13083-852, BrazilDepartment of ICT and Natural Sciences, NTNU—Norwegian University of Science and Technology, Larsgårdsvegen 2, 6009 Alesund, NorwayPollution in the form of litter in the natural environment is one of the great challenges of our times. Automated litter detection can help assess waste occurrences in the environment. Different machine learning solutions have been explored to develop litter detection tools, thereby supporting research, citizen science, and volunteer clean-up initiatives. However, to the best of our knowledge, no work has investigated the performance of state-of-the-art deep learning object detection approaches in the context of litter detection. In particular, no studies have focused on the assessment of those methods aiming their use in devices with low processing capabilities, e.g., mobile phones, typically employed in citizen science activities. In this paper, we fill this literature gap. We performed a comparative study involving state-of-the-art CNN architectures (e.g., Faster RCNN, Mask-RCNN, EfficientDet, RetinaNet and YOLO-v5), two litter image datasets and a smartphone. We also introduce a new dataset for litter detection, named PlastOPol, composed of 2418 images and 5300 annotations. The experimental results demonstrate that object detectors based on the YOLO family are promising for the construction of litter detection solutions, with superior performance in terms of detection accuracy, processing time, and memory footprint.https://www.mdpi.com/1424-8220/22/2/548littermarine littercitizen sciencelitter detectionobject detectionneural networks |
| spellingShingle | Manuel Córdova Allan Pinto Christina Carrozzo Hellevik Saleh Abdel-Afou Alaliyat Ibrahim A. Hameed Helio Pedrini Ricardo da S. Torres Litter Detection with Deep Learning: A Comparative Study Sensors litter marine litter citizen science litter detection object detection neural networks |
| title | Litter Detection with Deep Learning: A Comparative Study |
| title_full | Litter Detection with Deep Learning: A Comparative Study |
| title_fullStr | Litter Detection with Deep Learning: A Comparative Study |
| title_full_unstemmed | Litter Detection with Deep Learning: A Comparative Study |
| title_short | Litter Detection with Deep Learning: A Comparative Study |
| title_sort | litter detection with deep learning a comparative study |
| topic | litter marine litter citizen science litter detection object detection neural networks |
| url | https://www.mdpi.com/1424-8220/22/2/548 |
| work_keys_str_mv | AT manuelcordova litterdetectionwithdeeplearningacomparativestudy AT allanpinto litterdetectionwithdeeplearningacomparativestudy AT christinacarrozzohellevik litterdetectionwithdeeplearningacomparativestudy AT salehabdelafoualaliyat litterdetectionwithdeeplearningacomparativestudy AT ibrahimahameed litterdetectionwithdeeplearningacomparativestudy AT heliopedrini litterdetectionwithdeeplearningacomparativestudy AT ricardodastorres litterdetectionwithdeeplearningacomparativestudy |