International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 11 Issue: 04 | Apr 2024
p-ISSN: 2395-0072
www.irjet.net
HARNESSING DEEP LEARNING FOR UNDERWATER PLASTIC TRASH IDENTIFICATION Javali Chetana1, Jayashree R2, Kamini M3, Siri D H4 1,2,3,4BE, Information Science and Engineering, GSSS Institute of Engineering and Technology for Women-
Mysuru,Karnataka, India
5Assisstant Professor, Dept. of Information Science and Engineering, GSSS Institute of Engineering and Technology
for Women-Mysuru, Karnataka, India ---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - One of the most serious environmental issues is
involved in building an effective detection system. We will discuss the data collection and annotation process, the selection of suitable deep learning architectures, model training and evaluation, deployment in real-world scenarios, and the ongoing need for improvement and public engagement. By harnessing the power of deep learning, we can contribute to the protection and preservation of our precious oceans and the myriad of life they sustain.
ocean pollution; according to a study, most of the plastic trash from land which finds in the ocean is from human waste. These pollutants threaten the animals, the surrounding economy, and the balance of the marine ecosystem. Humans and aquatic life will definitely be affected by this. Though the most common techniques for evaluating and classify plastics seem to work well, they have several limitations. For this reason to simplify removal, it is essential to take advantage of innovative alternatives that are capable of detecting plastics and using the most recent innovations in technology. To locate and identify objects, we investigated the deep learning recognition of objects algorithms YOLO v4 . The epipelagic layers of the water bodies consist of marine plastics. Internetaccessible pictures of marine waste are used to build the databases. Image augmentation made it feasible to add further images to the collection. After the completion of the results, a review is done on the algorithm's performance in addition to the Mean Average Precision of YOLO v4 .
paper look exactly like this document. The easiest way to do this is simply to download the template, and replace(copypaste) the content with your own material. Number the reference items consecutively in square brackets (e.g. [1]). However the authors name can be used along with the reference number in the running text. The order of reference in the running text should match with the list of references at the end of the paper.
2.PROBLEM STATEMENT This project will identify and classify underwater plastic trash from images, aiding in environmental conservation efforts. Researchers can gather valuable data on the types, quantities plastic waste, aiding in the development of effective cleanup and mitigation strategies. This project will create environmental awareness about the conservation efforts and policies to reduce plastic pollution and protect marine life.
Key Words: YOLO v4, plastic trash, ocean pollution, deep learning, training, Image pre-processing.
1.INTRODUCTION The pervasive problem of plastic pollution in our oceans and water bodies poses a significant threat to marine life and ecosystems. To address this issue, innovative solutions are required, and one such solution is leveraging deep learning technologies to detect and monitor plastic trash underwater. Deep learning, a subset of artificial intelligence, has shown remarkable capabilities in image recognition and object detection tasks, making it a promising tool for identifying and tracking plastic debris beneath the water's surface. This application of deep learning involves the development of a computer vision system that can automatically analyze underwater images or video footage, pinpoint plastic waste, and facilitate timely response and conservation efforts. Such a system can assist marine researchers, environmental organizations, and policymakers in comprehending the extent of the problem and taking appropriate actions to mitigate plastic pollution in aquatic environments. In this exploration of detecting plastic trash underwater using deep learning, we will delve into the key components and methodologies
© 2024, IRJET
|
Impact Factor value: 8.226
3.PROPOSED SYSTEM The Proposed System is based on Deep Learning Algorithm using YOLO(You Only look Once). Choose a YOLO-based architecture suitable for object detection tasks. We are opting YOLOv4, version in our project. It is designed and developed in such a way that it provides dynamic response for finding the plastics under the water which helps to the sea divers to easily identify the plastics under the water. The proposed system uses Yolo Model to detect the plastic waste under the water.
|
ISO 9001:2008 Certified Journal
|
Page 2401