A Deep Learning Approach to Plastic Bottle Waste Detection on the Water Surface using YOLOv6 and YOLOv7 DOI Open Access

Naufal Laksana Kirana,

Diva Kurnianingtyas, Indriati Indriati

et al.

Engineering Technology & Applied Science Research, Journal Year: 2024, Volume and Issue: 14(6), P. 18623 - 18630

Published: Dec. 2, 2024

Deep learning is a branch of machine with many layers, such as the You Only Look Once (YOLO) method. From various versions YOLO, YOLOv6 and YOLOv7 are considered more prominent because they achieve high Mean Average Precision (mAP) values. Both YOLO have been implemented into problems, especially in waste detection problem. Plastic bottle one most common types that pollutes Indonesian waters. This study aims to solve this problem by helping sort surface waters applying YOLOv7. FloW-Img was used, obtained on request from Orcaboat website. The dataset consists 500,000 objects 2,000 images. models were evaluated using mAP running time. results show can handle well, values 0.873 0.512, respectively. In addition, (4.21 m/s) has higher speed than (13.7 m/s). However, tests images do not objects, provides better accuracy consistency results, making it suitable for real-world applications demand environments much visual noise.

Language: Английский

Enhanced Real-Time Object Detection using YOLOv7 and MobileNetv3 DOI Open Access
Sara Ennaama, Hassan Silkan, Ahmed Bentajer

et al.

Engineering Technology & Applied Science Research, Journal Year: 2025, Volume and Issue: 15(1), P. 19181 - 19187

Published: Feb. 1, 2025

Object detection serves as a crucial element in computer vision, increasingly relying on deep learning techniques. Among various methods, the YOLO series has gained recognition an effective solution. This research enhances object by merging YOLOv7 with MobileNetv3, known for its efficiency and feature extraction. The integrated model was tested using COCO dataset, which contains over 164,000 images across 80 categories, achieving mAP score of 0.61. Additionally, confusion matrix analysis confirmed accuracy, especially detecting common objects such 'person' 'car' minimal misclassifications. results demonstrate potential proposed to address complexities real-world scenarios, highlighting applicability scientific industrial domains.

Language: Английский

Citations

1
Comparative Analysis of YOLOv8 and YOLOv9 Models for Real-Time Plant Disease Detection in Hydroponics DOI Open Access
Abhishek Tripathi, Vinaya Gohokar, Rupali Kute

et al.

Engineering Technology & Applied Science Research, Journal Year: 2024, Volume and Issue: 14(5), P. 17269 - 17275

Published: Oct. 9, 2024

Plant diseases are a significant threat to modern agricultural productivity. Hydroponic systems also affected for various reasons. Reliable and efficient detection methods essential early intervention management of in hydroponics. This study investigates the use You Only Look Once (YOLO) models, namely YOLOv8 YOLOv9, plant hydroponic environment. A diverse dataset was prepared, comprising images from hydroponics system setup New Disease Image Dataset Kaggle. Custom annotated were used train test models compare their accuracy, processing speed, robustness systems. The results showed that YOLOv9 is slightly better than terms as it achieved 88.38% compared 87.22%, respectively. requires less computational resources takes relatively time real-time disease detection. Therefore, recommended portable devices.

Language: Английский

Citations

2
Maize Leaf Disease Detection using Manta-Ray Foraging Optimization with Deep Learning Model DOI Open Access

S. Vimalkumar,

R.S. Latha

Engineering Technology & Applied Science Research, Journal Year: 2024, Volume and Issue: 14(5), P. 17068 - 17074

Published: Oct. 9, 2024

Maize (corn) is a major and high yield crop, cultivated worldwide although diseases may cause severe reductions. Monitoring identifying maize throughout the growth cycle are crucial tasks. Accurately detecting an issue for farmers who need expertise in plant pathology, while professional diagnosis can be time-consuming expensive. Meanwhile, conventional Deep Learning (DL) image recognition models slowly entering field of disease detection. This paper proposes Intelligent Leaf Disease Detection design using Manta-Ray Foraging Optimization with DL (IMLDD-MRFODL) model. The aim IMLDD-MRFODL method to detect categorize leaf diseases. applies Median Filtering (MF) preprocessing, densely connected network (DenseNet) feature extraction, MRFO technique hyperparameter tuning. exploits Long Short-Term Memory (LSTM) classification. Experimental evaluation was conducted validate approach comparative analysis exhibited superior accuracy proposed method.

Language: Английский

Citations

1
A Recyclable Waste Image Recognition System with YOLOv8 for Children's Environmental Education DOI Open Access

Aiman Fahmi Zambri,

Shuzlina Abdul-Rahman, Norlina Mohd Sabri

et al.

Engineering Technology & Applied Science Research, Journal Year: 2024, Volume and Issue: 14(5), P. 16492 - 16498

Published: Oct. 9, 2024

Rapid economic growth and increasing urban population have led to a significant increase in waste production, raising serious concerns for countries worldwide. As the expands, generation poses numerous environmental public health challenges. This study focuses on educating children about recyclable promote early awareness proper classification habits. Specifically, this investigates performance of YOLOv8 model embed it into recognition system tailored children's management education. Datasets were obtained from Kaggle underwent preprocessing. The findings show that with 100 epochs, an SGD optimizer, batch size 25 achieved best performance, accuracy over 94% low loss 0.367. demonstrated competitive detecting classifying images, highlighting its potential as effective tool educational programs aimed at teaching importance promoting sustainable practices age.

Language: Английский

Citations

0
Design of Automatic Game System Based on Artificial Intelligence for Three Pieces of Chess DOI
Ziwen Li,

Dun Lin,

Kaitao Deng

et al.

Published: Sept. 13, 2024

Language: Английский

Citations

0
A Deep Learning Approach to Plastic Bottle Waste Detection on the Water Surface using YOLOv6 and YOLOv7 DOI Open Access

Naufal Laksana Kirana,

Diva Kurnianingtyas, Indriati Indriati

et al.

Engineering Technology & Applied Science Research, Journal Year: 2024, Volume and Issue: 14(6), P. 18623 - 18630

Published: Dec. 2, 2024

Deep learning is a branch of machine with many layers, such as the You Only Look Once (YOLO) method. From various versions YOLO, YOLOv6 and YOLOv7 are considered more prominent because they achieve high Mean Average Precision (mAP) values. Both YOLO have been implemented into problems, especially in waste detection problem. Plastic bottle one most common types that pollutes Indonesian waters. This study aims to solve this problem by helping sort surface waters applying YOLOv7. FloW-Img was used, obtained on request from Orcaboat website. The dataset consists 500,000 objects 2,000 images. models were evaluated using mAP running time. results show can handle well, values 0.873 0.512, respectively. In addition, (4.21 m/s) has higher speed than (13.7 m/s). However, tests images do not objects, provides better accuracy consistency results, making it suitable for real-world applications demand environments much visual noise.

Language: Английский

Citations

0