UNIC study finds that approximately 66,000 plastic bottles, bags, straws, and food wrappings litter the shorelines of Cyprus.
Endowed with 69 Blue Flag designated beaches, Cyprus needs to invest more in efforts to safeguard its coastlines. Researchers at the University of Nicosia have developed an innovative method to assess the extent of plastic pollution affecting the island’s shorelines, as part of the 4-year PLASTICMED Project. Using artificial intelligence tools, the University’s Marine & Carbon Lab (www.carbonlab.eu) team was able to identify, localise and estimate the amount of plastic detritus dotting the Cypriot coastlines.
Recognising that it is impossible to manually count the thousands of plastics encountered on our seashores, the UNic researchers created an intelligent method based on deep-learning algorithms such as YOLOv5 and YOLACT++. Initially, the tools were trained and tested on the task of identifying the seven (7) most prevalent categories of plastic debris found in the marine environment, which comprise, among others, bottles, bags, buckets, straws, and food wrappings.
Subsequently, the effectiveness of the deep-learning techniques to discern plastics was validated via video footage depicting plastic litter recorded at various beaches in Cyprus. Thanks to its ability to scan large swaths of both coastline and seawater, the intelligent method was able to detect and classify plastics shown in video footage of the marine environment.
Utilising images depicting plastic debris from six beaches in Cyprus, the intelligent technique, based on the YOLACT++ algorithm, traced plastic debris from four categories, namely, bottles, bags, straws, and food wrappings. Considering the total length of the examined beaches, the intelligent method predicted a density of 0.035 plastics/m2. Extrapolating this figure to the full length of the shorelines of Cyprus, which span about 735 km, the proposed method estimated that there are about 66,000 plastic objects weighing a total of about 1,000 kg.
The estimated plastic litter density in Cyprus of 0.035 plastics/m2 is comparable to the number of plastics found in neighbouring countries. For example, the plastic litter density in the Ionian Sea was estimated to be at least 0.03 plastics/m2, while it was reported that there are at least 0.085 plastics/m2 on Turkish shores. Besides estimating the plastic litter density, the dimensions of plastic waste were predicted using image processing. Findings revealed that the predominant size of plastics lying on Cypriot beaches ranged between 10 and 30 cm, which matches the size of common plastics often littering these seashores.
Concluding, one of the project’s goals was to devise a fast and efficient method for recognising and counting plastics across the world’s coasts and oceans. One aspect of the solution to the problem of plastic pollution lies in the need to swiftly detect plastics, which are often dispersed over large stretches on coasts or in the seas. Considering the vast areas that need to be scanned for plastics, it is inconceivable to sample all of the world’s coasts. However, the development of sophisticated technological tools, such as the intelligent method proposed in the present study, may prove instrumental in the fight for addressing plastic pollution both at sea and on land.
The PLASTICMED Project was jointly funded by the Universitas Foundation, the University of Nicosia, the A.G. Leventis Foundation, and the EU’s Horizon 2020 programme. The findings of the study appeared in the prestigious scientific journal of Marine Pollution Bulletin: https://bit.ly/3lbecqz. The research team consisted of Ms. Kyriaki Kylili, Dr. Alessandro Artusi and Dr. Constantinos Hadjistassou.
More info on the Project can be found at: http://www.carbonlab.eu/marine-debris/
Related videos and photos can be viewed at: https://bit.ly/3nIYYuc