International Journal of Research In Science & Engineering最新文献

筛选

英文中文

The Use of Cnn-Based Multitask Learning for Smart Motorcycle Helmet Design 基于 Cnn 的多任务学习在智能摩托车头盔设计中的应用

International Journal of Research In Science & Engineering Pub Date : 2024-08-10 DOI: 10.55529/ijrise.45.1.13

Mahdi Fadil Khaleel, Zainab Mohammed Abdulkareem, Saadaldeen Rashid Ahmed

{"title":"The Use of Cnn-Based Multitask Learning for Smart Motorcycle Helmet Design","authors":"Mahdi Fadil Khaleel, Zainab Mohammed Abdulkareem, Saadaldeen Rashid Ahmed","doi":"10.55529/ijrise.45.1.13","DOIUrl":"https://doi.org/10.55529/ijrise.45.1.13","url":null,"abstract":"By informing and upholding legal guidelines, robotically figuring out motorcycle helmets through video surveillance contributes to improving road protection. It is difficult for current techniques to keep an eye fixed on motorbikes and inform riders from passengers. In order to tackle those problems, we recommend to reveal and recognize motorbikes using a CNN oriented multi- venture gaining knowledge of (MTL) approach, with a focal point on helmet- wearing motorcyclists. We offer the HELMET the dataset, that's made up of ninety one, 000 annotated frames of 10,006 motorbikes at 12 one-of-a-kind Myanmar observation web sites. This dataset may be used as a factor of reference for techniques of detection. Our technique, referred to as MTL, can provide higher accuracy and efficiency by means of combining helmet use categorization with similarity getting to know. Operating at a charge of greater than 8 frames for each 2d (FPS) on hardware, our technique attains a sixty seven.3% F degree in identifying cyclists and their helmet utilization. The effectiveness of learning in obtaining crucial expertise about avenue protection is highlighted by means of this study. In addition, we present a motorbike helmet prepared with earbuds, a charging module, an integrated computer unit, transceivers, and a photograph sensor. This helmet makes use of picture popularity modes in both daylight and midnight occasions to pick out automobiles coming near the wearer. According to experimental consequences, vehicles and buses' registration plates can be recognized with up to175% accuracy in the course of the1day & 70% accuracy at night. The new smart motorbike helmet is supposed to apprehend vehicles in real time inside a five-meter radius, increasing street protection.","PeriodicalId":506436,"journal":{"name":"International Journal of Research In Science & Engineering","volume":"2 5","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-08-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141919879","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

International Journal of Research In Science &amp; Engineering最新文献

International Journal of Research In Science & Engineering最新文献