Wyniki wyszukiwania - BazTech

Ograniczanie wyników

Znaleziono wyników: 2

Liczba wyników na stronie

Wyniki wyszukiwania

Sortuj według:

Ogranicz wyniki do:

udCATS : A Comprehensive Unsupervised Deep Learning Framework for Detecting Collective Anomalies in Time Series

Pham Truong Son, Nguyen Viet Hung, Le Anh Thang, Bui Van Duong

Annals of Computer Science and Information Systems

2022

Vol. 33

201--206

Anomaly detection has recently gained enormous attention from the research community. It is widely applied in many industrial areas, such as information security, financing, banking, and insurance. The data in these fields can mainly be represented as time series data, the corollary being that time series anomaly detection plays an essential role in these applications. Therefore, many authors have tried to solve the problem of collective anomaly detection in time series. They have proposed several approaches, from classical methods such as Isolation Forests to modern deep learning networks such as Autoencoders. However, a comprehensive framework for handling this problem is still lacking. In this work, firstly, we propose using an Attention-based Bidirectional LSTM Autoencoder (Att-BiLSTM-AE) as an anomaly detection model. Furthermore, in the essential part of this paper, we developed a comprehensive unsupervised deep learning framework, udCATS, to solve the problem of detecting collective anomalies in time series. Our experiments show that the Att-BiLSTM-AE outperforms other detection models, and using it within the udCATS framework increases the detection accuracy.

The capability of terrestrial laser scanning for monitoring the displacement of high-rise buildings

Pham Trung Dung, Pham Quoc Khanh, Cao Xuan Cuong, Nguyen Viet Hung, Ngo Sy Cuong

Inżynieria Mineralna

2021

no. 2

495--504

Recently, terrestrial laser scanner (TLS) has been increasingly used to monitor of displacement of high-rise buildings. The main advantages of this technique are time-saving, higher point density, and higher accuracy in comparison with GPS and conventional methods. While TLS is ordinary worldwide, there has been no study of the capability of TLS in monitoring the displacement of high-rise buildings yet in Vietnam. The paper's goal is to build a procedure for displacement monitoring of high-rise buildings and assess the accuracy of TLS in this application. In the experiments, a scanned board with a 60 cm x 60 cm mounted on a moveable monument system is scanned by Faro Focus3D X130. A monitoring procedure using TLS is proposed, including three main stages: site investigation, data acquisition and processing, and displacement determination by the Cloud-to-Cloud method (C2C). As a result, the displacement of the scanned object between epochs is computed. In order to evaluate the accuracy, the estimated displacement using TLS is compared with the real displacement. The accuracy depends on scanning geometry, surface property, and point density conditions. Our results show that the accuracy of the estimated displacement is within ± 2 mm for buildings lower than 50 m of height. Thus, TLS completely meets the accuracy requirements of monitoring displacement in the Vietnam Standards of Engineering Surveying. With such outstanding performance, our workflow of using TLS could be applied to monitor the displacement of high-rise buildings in the reality of geodetic production in Vietnam.