Surveyance of the marine environment for beached plastic debris using a novel, hyperspectral shortwave infrared camera on an uncrewed aerial vehicle (UAV)

Plastic waste has been rapidly accumulating in the natural environment for decades since<br/>the commercial introduction of plastic in the 1950s as a disposable material. Millions of<br/>tonnes of durable, persistent plastic waste now contaminate global ecosystems, including<br/>rivers, urban beaches, remote islands, the deep sea, soils, mountains, and very small<br/>pieces (microplastics) have even been detected in the air. Such pervasive plastic debris is<br/>a deadly and common threat to wildlife due to ingestion and entanglement. Large debris<br/>are particularly problematic not only because of these direct consequences, but because<br/>they break down into smaller pieces in the environment. These smaller pieces are more<br/>easily ingested by a greater diversity of species and are more difficult to remediate. It is<br/>therefore sensible to focus prevention, remediation, and monitoring efforts on large, more<br/>easily observable debris. In the past five years, monitoring efforts have increasingly<br/>included the investigation of automated plastic detection via remote sensing technologies<br/>and techniques. This approach has the potential to reduce observer bias (associated with<br/>manual, visual surveys), and realise regular and widespread plastic pollution monitoring,<br/>with the ultimate goal of achieving environmentally relevant reductions in plastic production<br/>and pollution. This thesis contributes to these aims by i) investigating the performance of<br/>novel remote sensing methods for plastic detection, ii) developing and implementing a new<br/>plastic detection algorithm, and iii) evaluating for the first time, the impact of operational and<br/>environmental variables on these specific plastic detection methods.<br/>A large proportion of plastic litter detection investigations have implemented inexpensive<br/>and common digital cameras that operate in the visible light spectrum. This approach is<br/>often limited in capability/reliability by inherent ambiguity, resulting from two-dimensional<br/>imagery of materials of unverifiable chemical composition. This thesis investigates the<br/>detection of shoreline plastic litter with the aerial deployment of a specialised, hyperspectral<br/>camera that is sensitive to shortwave infrared (SWIR) light and that measures signal from<br/>more than ten times as many wavelengths as a conventional digital camera. The rationale<br/>for pursuing this research is further detailed in Chapter 1, as well as thesis structure and<br/>objectives. The chosen SWIR sensing technology was selected from numerous other<br/>sensor technologies introduced in Chapter 2, for its specific relevance to detecting plastic<br/>based on chemical composition. Chapter 3 details the functional principles of this<br/>technology, which sees widespread use in commercial recycling facilities to successfully<br/>sort large volumes of different types of plastic. The criteria for selecting the specific SWIR<br/>camera (“OCI-F SWIR” from BaySpec, San Jose, CA, USA) implemented in this work, and an initial evaluation of its capabilities under a controlled lab environment are also presented<br/>in Chapter 3, prior to assessing field performance in Chapters 4 and 5.<br/>Chapter 4 presents the development of a customised, software-based plastic detection<br/>algorithm, informed by results from Chapter 3 and by methods implemented in the literature.<br/>Spectral Correlation Mapping (SCM) was selected for its computational efficiency and<br/>documented improvement over detection with the commonly applied Spectral Angle<br/>Mapping (upon which SCM is based). This method achieved successful classification of<br/>plastic and non-plastic pixels within imagery of individual items of polyethylene (PE) and<br/>polypropylene (PP) – the most common polymers produced globally – placed on a sandy<br/>shoreline. The images were collected with a distance of approximately one metre between<br/>the camera and the plastic items to measure a strong signal from the plastic and to inform<br/>algorithm parameter selection before increasing image complexity and introducing<br/>environmental variability in Chapter 5.<br/>Imagery analysed in Chapter 5 was collected on different days to evaluate the impacts on<br/>plastic detection performance of: plastic type, cloud cover, beach substrate, plastic’s<br/>exposure to the environment (harvested from domestic and environmental sources), and<br/>the altitude of the camera above the plastic. The camera was first deployed on a “highline”<br/>apparatus and manually conveyed across a fixed transect, at a height of five metres above<br/>a variety of plastic items arranged on the beach below. The purpose of this approach was<br/>to ensure the collection of data under steady, regular movement, which might not be as<br/>easily achieved with an uncrewed aerial vehicle (UAV). SCM was not able to consistently<br/>classify plastic pixels in each transect as PE nor PP when SCM was applied in the same<br/>way it had been in Chapter 4 (to images recorded at a one metre distance, of individual<br/>plastic items). The mathematical factors contributing to this discrepancy were investigated<br/>and modifications were subsequently implemented to the application of SCM, to compare<br/>segmented spectra. An additional, new algorithm called Reflectance Range Analysis (RRA)<br/>was also developed and evaluated. The development of RRA was based on differences<br/>observed in the spectral characteristics of plastic items and substrate measured in the aerial<br/>imagery, and on recommendations in the literature to develop/implement algorithms that<br/>detect the intensity of plastic’s characteristic absorption. Where SCM classifies every image<br/>pixel based on spectral similarity to a selected reference, RRA reveals spectral differences<br/>between image pixels without requiring similarity to a reference.<br/>Chapter 6 concludes this work by summarising the key findings and demonstrating the place<br/>of this research in the evolution of plastic detection with SWIR remote sensing.<br/>Recommendations are made here for augmenting plastic detection capabilities through<br/>further study and improvements to survey and equipment design, as well as leveraging the<br/>strengths of multiple remote sensing technologies. The diversity of current, newly available,<br/>and upcoming sensor technologies offers numerous opportunities to measure, monitor, and<br/>collectively address the problem of human-driven plastic pollution.

Name dc:type.qualificationname

Doctor of Philosophy (awarded by UHI)

Level dc:type.qualificationlevel

Doctoral Thesis

Grantor dc:publisher.institution

University of the Highlands and Islands

Year dc:date.issued

2023

Author dc:creator

• Cocking, Jennifer Laura

Advisors dc:contributor.advisor

• Narayanaswamy, Bhavani
• Williamson, Benjamin

Language dc:language

eng

Identifier

oai:pure.atira.dk:studenttheses/7d159d9c-297b-4817-831e-673ddc4e5e67

OAI identifier oai:identifier

oai:pure.atira.dk:studenttheses/7d159d9c-297b-4817-831e-673ddc4e5e67