We can't find the internet
Attempting to reconnect
Something went wrong!
Hang in there while we get back on track
Conserving Oceans from Plastic Waste using Convolution Neural Network Model
AI summary Read the abstract
Researchers investigated microplastic accumulation in seawater and sediments from a semi-enclosed bay, detecting significant concentrations that varied with season and proximity to urban inputs. The findings highlight how hydrodynamic conditions and human activities jointly shape microplastic distribution in coastal marine environments.
Plastic Pollution is posing a serious threat to the ocean and its diversity to an extreme level. The number of plastic bags, bottle and other plastic materials which are disposed into the ocean not only pollute its water, tend to change its properties but also can kill marine animals and cause the marine diversity to degrade. Internet of Things (IoT) is a new technology which has sensing capabilities and work in a real-time environment. Aquatic IoT sensors can be used in order to detect and tract these plastic objects in the water. By this method, the cleaning of oceans can be done in an effective manner. There are various deep learning models that have the potential to classify the objects in a real-time environment either on land or in water. These models can be used in order to detect plastic waste from marine debris. In this paper a two-step approach has been adopted with the aim to detect and classify plastic objects under ocean water. Firstly, the IoT sensors are deployed deep into the ocean which take pictures of the different regions in the ocean. Next, the VGG16 CNN (Convolutional Neural Network) deep learning model has been adopted for this paper which takes these pictures as input and classify them into plastic, plants, animals and other objects so that amount of plastic waste can be detected. The approach showed 96.5% accuracy as compared to other object detection and classification approaches. More effectiveness in the proposed work can be added by ensembling the random boost classifier and decision tree approach in the future.
More Papers Like This
AnAI-PoweredSystemforReal-TimeUnderwaterPlastic Detection and Classification Using YOLOv7 and VGG16
AI summary Read the abstract
This study proposes a hybrid AI architecture combining YOLOv7 for real-time detection and VGG16 for detailed classification of underwater plastic waste, including bottles, bags, nets, and wrappers, with SMTP notification for automated alerts. The system demonstrates high accuracy and performance under challenging underwater conditions including turbidity and low visibility, offering a practical tool for marine waste monitoring and rapid response.
A novel deep learning model for predicting marine pollution for sustainable ocean management
AI summary Read the abstract
This study developed a deep learning model combining two-step clustering and fully convolutional networks (FCN) to detect and categorize plastic waste in marine and undersea environments, achieving 98.38% accuracy using Kaggle's plastic location dataset. The trained model was applied to an automated floating plastic detection system, demonstrating the potential of AI-driven tools to enhance marine pollution monitoring.
Enhancing marine debris identification with convolutional neural networks
AI summary Read the abstract
A deep learning model was developed to identify and classify marine debris components captured by underwater remotely operated vehicle imagery, addressing the challenge of widely distributed ocean waste including microplastics. The convolutional neural network demonstrated improved accuracy for debris detection and classification compared to conventional image analysis methods.
Plastic Waste on Water Surfaces Detection Using Convolutional Neural Networks
AI summary Read the abstract
Researchers evaluated state-of-the-art convolutional neural network architectures for automatically detecting plastic waste on water surfaces, training models on a dataset representing four categories of plastic litter including plastic bags. The study benchmarked multiple CNN object detection models following extensive dataset preprocessing to determine the most effective approach for automated plastic pollution identification.
Image-Based Detection and Deep Learning Methods for Plastic and Environmental Monitoring
AI summary Read the abstract
Researchers trained computer programs to sort waste photos from the TrashNet image dataset into "harmful" (plastic, mixed trash) and "less harmful" (paper, cardboard, glass, metal) categories. One model, based on VGG16, correctly classified images more than 82% of the time, suggesting such tools could help automate detection of plastic pollution.
Research digests by email
When a large batch of papers lands in the Atlas, we read through it and send a short write-up of what stood out.