Real-Time Ground-Level Building Damage Detection Based on Lightweight and Accurate YOLOv5 Using Terrestrial Images

نویسندگان

چکیده

Real-time building damage detection effectively improves the timeliness of post-earthquake assessments. In recent years, terrestrial images from smartphones or cameras have become a rich source disaster information that may be useful in assessing at lower cost. this study, we present an efficient method based on combination with improved YOLOv5. We compiled Ground-level Detection Building Damage Assessment (GDBDA) dataset consisting annotations types, including debris, collapse, spalling, and cracks. A lightweight accurate YOLOv5 (LA-YOLOv5) model was used to optimize efficiency accuracy. particular, Ghost bottleneck added backbone neck modules model, aim reduce size. Convolutional Block Attention Module (CBAM) module enhance recognition effect. addition, regarding scale difference damage, Bi-Directional Feature Pyramid Network (Bi-FPN) for multi-scale feature fusion aggregate features different types. Moreover, depthwise separable convolution (DSCONV) further compress parameters. Based our GDBDA dataset, proposed not only achieved accuracy above 90% targets, but also had smallest weight size fastest speed, which by about 64% 24%, respectively. The performed well datasets regions. overall results indicate realizes rapid detection, meets requirement embedding future.

برای دانلود باید عضویت طلایی داشته باشید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Real-time damage detection of bridges using adaptive time-frequency analysis and ANN

Although traditional signal-based structural health monitoring algorithms have been successfully employed for small structures, their application for large and complex bridges has been challenging due to non-stationary signal characteristics with a high level of noise. In this paper, a promising damage detection algorithm is proposed by incorporation of adaptive signal processing and Artificial...

متن کامل

Real-Time intrusion detection alert correlation and attack scenario extraction based on the prerequisite consequence approach

Alert correlation systems attempt to discover the relations among alerts produced by one or more intrusion detection systems to determine the attack scenarios and their main motivations. In this paper a new IDS alert correlation method is proposed that can be used to detect attack scenarios in real-time. The proposed method is based on a causal approach due to the strength of causal methods in ...

متن کامل

A Real Time Traffic Sign Detection and Recognition Algorithm based on Super Fuzzy Set

Advanced Driver Assistance Systems (ADAS) benefit from current infrastructure to discern environmental information. Traffic signs are global guidelines which inform drivers from near characteristics of paths ahead. Traffic Sign Recognition (TSR) system is an ADAS that recognize traffic signs in images captured from road and show information as an adviser or transmit them to other ADASs. In this...

متن کامل

Building Damage Detection of the 2003 Bam, Iran Earthquake Using Quickbird Images Based on Object-based Classification

In resent years, remote sensing technologies are utilized in disaster management as a means of information gathering in a large scale disaster. QuickBird captured a clear image of Bam City, eight days after the 2003 Bam, Iran, earthquake. The city was also observed by QuickBird about three months before the event. In this paper, we perform building damage detection based on land cover classific...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

ژورنال

عنوان ژورنال: Remote Sensing

سال: 2022

ISSN: ['2315-4632', '2315-4675']

DOI: https://doi.org/10.3390/rs14122763