OCEAN PLASTIC WASTE DETECTION FROM AERIAL IMAGES USING TRANSFER LEARNING

Abstract

Ocean plastic pollution has become one of the most serious environmental challenges affecting
marine ecosystems, wildlife, and human health. Large amounts of plastic waste enter oceans
every year, making manual monitoring and cleanup difficult and time-consuming. This project
focuses on detecting ocean plastic waste using aerial images captured by drones, satellites, or
aircraft. By applying image processing and machine learning techniques, the system can
automatically identify and classify plastic waste present on the ocean surface.
The proposed approach uses a trained model to analyze aerial images and detect patterns that
distinguish plastic debris from water and other natural objects. Preprocessing techniques such as
image enhancement, noise removal, and segmentation are used to improve detection accuracy. A
machine learning algorithm, such as Convolutional Neural Networks (CNN), is then applied to
classify whether plastic waste is present in the image. The system can highlight the detected
waste areas and provide information that can assist environmental organizations in planning
cleanup operations.
This automated detection method helps in monitoring large ocean areas quickly and efficiently
compared to traditional manual observation methods. The proposed system contributes to
environmental protection by enabling faster detection of plastic pollution and supporting efforts
to reduce marine waste.
The system involves image preprocessing, feature extraction, and classification to distinguish
plastic from non-plastic regions. Transfer learning reduces the need for large datasets and
training time while improving detection accuracy. The model is trained and evaluated using
standard performance metrics such as accuracy, precision, and recall.