• Korean Journal of Remote Sensing
• /
• v.37 no.4
• /
• pp.719-731
• /
• 2021

This study proposes a two-stage hybrid classification model for crop classification using multi-temporal remote sensing images; the model combines feature embedding by using an autoencoder (AE) with a convolutional neural network (CNN) classifier to fully utilize features including informative temporal and spatial signatures. Long short-term memory (LSTM)-based AE (LAE) is fine-tuned using class label information to extract latent features that contain less noise and useful temporal signatures. The CNN classifier is then applied to effectively account for the spatial characteristics of the extracted latent features. A crop classification experiment with multi-temporal unmanned aerial vehicle images is conducted to illustrate the potential application of the proposed hybrid model. The classification performance of the proposed model is compared with various combinations of conventional deep learning models (CNN, LSTM, and convolutional LSTM) and different inputs (original multi-temporal images and features from stacked AE). From the crop classification experiment, the best classification accuracy was achieved by the proposed model that utilized the latent features by fine-tuned LAE as input for the CNN classifier. The latent features that contain useful temporal signatures and are less noisy could increase the class separability between crops with similar spectral signatures, thereby leading to superior classification accuracy. The experimental results demonstrate the importance of effective feature extraction and the potential of the proposed classification model for crop classification using multi-temporal remote sensing images.

• Journal of the Korean Society for Precision Engineering
• /
• v.13 no.11
• /
• pp.57-63
• /
• 1996

Monitoring of grinding troble occurring during the process is classified into the quantitative data which depends upon a sensor and the qualitative knowledge which relies upon an empirical knowledge. Since grinding operation is highly related with a large amount of functional parameters, it is actually deficulty in copying wiht the grinding troubles through the process. To cope with grinding trouble, it is an effective monitoring systems when occurring the grinding process. The use of neural networks is an effective method of detection and/or monitroing on the grinding trouble. In this paper, four parameters which are derived from the AE(Acoustic Emission) signatures are identified, and grinding monitoring system utilized a back propagation learning algorithm of PDP neural networks is presented.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2011.04a
• /
• pp.582-587
• /
• 2011

Acoustic Emission(AE) is widely used for early detection of faults for rotating machinery in these days because of its high sensitivity. AE signal has to need for transferring to low frequency range for the spectrum analysis included the fault mechanism. In transferring process, we lose a lot of fault information caused by unusable signal processing method. Discrete Wavelet Transform(DWT) is a method of signal processing for AE signatures, but the pattern of its mother function is not optimized with AE signals. So, we can lose the fault information when we want to use the DWT for AE signal. Therefore, in this paper, we will propose a novel pattern for DWT mother function, which is optimized with AE signals. And it will be applied to compare the results of DWT by daubechie and novel pattern.

• Smart Structures and Systems
• /
• v.16 no.1
• /
• pp.213-222
• /
• 2015

Early detection and precise location of leakage is of great importance for life-cycle maintenance and management of municipal pipeline system. In the past few years, acoustic emission (AE) techniques have demonstrated to be an excellent tool for on-line leakage detection. Regarding the multi-mode and frequency dispersion characteristics of AE signals propagating along a pipeline, the direct cross-correlation technique that assumes the constant AE propagation velocity does not perform well in practice for acoustic leak location. This paper presents an improved cross-correlation method based on wavelet transform, with due consideration of the frequency dispersion characteristics of AE wave and the contribution of different mode. Laboratory experiments conducted to simulate pipeline gas leakage and investigate the frequency spectrum signatures of AE leak signals. By comparing with the other methods for leak location identification, the feasibility and superiority of the proposed method are verified.

• Structural Engineering and Mechanics
• /
• v.75 no.6
• /
• pp.723-736
• /
• 2020

In this paper, acoustic emission (AE) and pattern recognition are utilized to identify the AE signal signatures caused by propagation of stress corrosion cracking (SCC) in a 304 stainless steel plate. The surface of the plate is under almost uniform tensile stress at a notch. A corrosive environment is provided by exposing the notch to a solution of 1% Potassium Tetrathionate by weight. The Global b-value indicated an occurrence of the first visible crack and damage stages during the SCC. Furthermore, a method based on linear regression has been developed for damage identification using AE data.

• Journal of KIISE:Computer Systems and Theory
• /
• v.35 no.4
• /
• pp.172-177
• /
• 2008

In this paper, we design an authentication protocol based on Fuzzy Signature Vault Scheme using a light signature feature extraction method for user convenience and efficiency of electronic commerce. The signature is used broadly in electronic commerce because it is one of the simple and low-cost biometric items. However, signature has a problem that there are few low-cost and safe protocols. To solve this problem, we design a feature extraction method which is adequate for characters of signature and Fuzzy Vault Scheme. In addition, we design and analyze an efficient authentication protocol with some parameters used in this procedure. The followings are advantages when this protocol is applied to authentication procedure; 1) using convenient and low-cost signatures, 2) being possible to verify users with spending only about second for signature processing and authentication, 3) one time on transmission for sign-in and verification and 4) getting user authentication with secret value at the same time.

• Journal of Astronomy and Space Sciences
• /
• v.31 no.2
• /
• pp.159-167
• /
• 2014

Solar variability is widely known to affect the interplanetary space and in turn the Earth's electromagnetical environment on the basis of common periodicities in the solar and geomagnetic activity indices. The goal of this study is twofold. Firstly, we attempt to associate modes by comparing a temporal behavior of the power of geomagnetic activity parameters since it is barely sufficient searching for common peaks with a similar periodicity in order to causally correlate geomagnetic activity parameters. As a result of the wavelet transform analysis we are able to obtain information on the temporal behavior of the power in the velocity of the solar wind, the number density of protons in the solar wind, the AE index, the Dst index, the interplanetary magnetic field, B and its three components of the GSM coordinate system, $B_X$, $B_Y$, $B_Z$. Secondly, we also attempt to search for any signatures of influence on the space environment near the Earth by inner planets orbiting around the Sun. Our main findings are as follows: (1) Parameters we have investigated show periodicities of ~ 27 days, ~ 13.5 days, ~ 9 days. (2) The peaks in the power spectrum of $B_Z$ appear to be split due to an unknown agent. (3) For some modes powers are not present all the time and intervals showing high powers do not always coincide. (4) Noticeable peaks do not emerge at those frequencies corresponding to the synodic and/or sidereal periods of Mercury and Venus, which leads us to conclude that the Earth's space environment is not subject to the shadow of the inner planets as suggested earlier.