• International Journal of Aeronautical and Space Sciences
• /
• 제6권1호
• /
• pp.8-16
• /
• 2005

This paper describes target detection performance of millimeter wave radar that exits on non-stationary target detection schemes in the ground clutter conditions. The comparison of various CFAR process schemes such as CA(Cell-Average)-CFAR, GO(Greatest Of)/SO(Smallest Of)-CFAR, and OS(Order Statistics)-CFAR performance are applied. Using matlab software, we show the performance and loss between target detection probability and signal to noise ratio. This paper concludes the OS-CFAR process performance is better than any others and satisfies the optimal detection probability without loss of detection in the homogeneous clutter, When range bins increase.

• Computers and Concrete
• /
• 제32권6호
• /
• pp.615-623
• /
• 2023

Crack detection is an essential method to ensure the safety of dam concrete structures. Low-quality crack images of dam concrete structures limit the application of neural network methods in crack detection. This research proposes a modified attentional mechanism model to reduce the disturbance caused by uneven light, shadow, and water spots in crack images. Also, the focal loss function solves the small ratio of crack information. The dataset collects from the network, laboratory and actual inspection dataset of dam concrete structures. This research proposes a novel method for crack detection of dam concrete structures based on the U-Net neural network, namely AF-UNet. A mutual comparison of OTSU, Canny, region growing, DeepLab V3+, SegFormer, U-Net, and AF-UNet (proposed) verified the detection accuracy. A binocular camera detects cracks in the experimental scene. The smallest measurement width of the system is 0.27 mm. The potential goal is to achieve real-time detection and localization of cracks in dam concrete structures.

• 대한원격탐사학회지
• /
• 제36권5_2호
• /
• pp.929-937
• /
• 2020

본 논문에서는 의미론적 분할을 위한 딥러닝 기술 중의 하나인 UNet++ 모델을 이용하여 다시기 위성영상의 변화지역을 탐지하고자 하였다. 다양한 손실함수에 대한 학습결과를 분석하기 위하여, 이진 교차 엔트로피, 자카드 변수에 의하여 학습된 UNet++ 모델에 의한 변화탐지 결과를 평가하였다. 또한, 딥러닝 모델의 결과는 WorldView-3 위성영상을 활용하여 기존의 화소기반 변화탐지 기법의 결과와 비교하여 평가하였다. 실험결과, 손실함수의 특성에 따라서 딥러닝 모델의 성능이 달라질 수 있음을 확인하였으나, 기존 기법들과 비교하여 우수한 결과를 나타내는 것도 확인하였다.

• Journal of the Korean Wood Science and Technology
• /
• 제36권5호
• /
• pp.24-32
• /
• 2008

Artificial brown-rot decay was induced to two wood species, Pinus densiflora and Pinus radiata. A modified direct inoculation method was used and the decay indicators of mass loss and two compressive mechanical properties, maximum compressive strength (MCS) and compressive stiffness, were estimated over the period of 8 weeks of fungal exposure. Measurable mass loss occurred 2 weeks after the fungal attack, with 15% to 22% of the loss occurring 8 weeks after fungal exposure with Fornitopsis palustris and Gloeophyllurn trabeurn. Mechanical properties proved to be far more sensitive than mass loss detection: approximately five to six times by quantity. Of the two mechanical properties, MCS was more sensitive to and consistent with progressive brown-rot decay. An ultrasonic test was performed to determine the feasibility and accuracy of this method for nondestructive detection of brown-rot decay. The ultrasonic test is highly sensitive at qualitative detection of the early stages of brown-rot decay.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제16권1호
• /
• pp.245-265
• /
• 2022

In response to problems such as insufficient extraction information, low detection accuracy, and frequent misdetection in the field of Thangka image defects, this paper proposes a YOLOv5 prediction algorithm fused with the attention mechanism. Firstly, the Backbone network is used for feature extraction, and the attention mechanism is fused to represent different features, so that the network can fully extract the texture and semantic features of the defect area. The extracted features are then weighted and fused, so as to reduce the loss of information. Next, the weighted fused features are transferred to the Neck network, the semantic features and texture features of different layers are fused by FPN, and the defect target is located more accurately by PAN. In the detection network, the CIOU loss function is used to replace the GIOU loss function to locate the image defect area quickly and accurately, generate the bounding box, and predict the defect category. The results show that compared with the original network, YOLOv5-SE and YOLOv5-CBAM achieve an improvement of 8.95% and 12.87% in detection accuracy respectively. The improved networks can identify the location and category of defects more accurately, and greatly improve the accuracy of defect detection of Thangka images.

• Current Optics and Photonics
• /
• 제8권1호
• /
• pp.105-111
• /
• 2024

We investigate the effect of photon loss on pair production in a hybrid atom-light system. The loss of light field not only affects the generation of photons, but also prevents the generation of atomic collective excitation, although the atoms are not influenced directly. We propose an unbalanced homodyne detection of the number of atomic collective excitation that overcomes the challenge caused by counting uncertainty in practical measurement. In discussion, we show that the intermode correlations and the number correlation is closely related to the initial input state, while the quadrature correlations are independent of the initial state and always exhibit opposite intermode correlations even in the presence of loss.

• 대한기계학회논문집B
• /
• 제30권7호
• /
• pp.638-645
• /
• 2006

Recently, we reported that significant ion loss occurred prior to its detection in the conventional single particles mass spectrometry and more seriously the loss is ion-kinetic-energy-dependent. These lead to significant error in the measured chemical composition of nanoparticles. Here we attempted to design a novel ion optics that is capable of 100% detection of ions generated from single nanoparticle. Using a commercial software SIMION, we simulated the trajectories of ions launched at different speeds inside the previous single particle mass spectrometer We tested how affect changes in shape of repelling plate, adding Einzel lens, substitution of tube electrode between extraction and acceleration grids. As a results, we could find a best design by assembling the trials in the present condition.

• 제어로봇시스템학회논문지
• /
• 제16권9호
• /
• pp.827-832
• /
• 2010

Many researchers have tried to detect the falling and to reduce the injury associated with falling. Normally the method of detection of a loss of balance is more efficient than that of a compensatory motion in order to predict the falling. The detection algorithm of the loss of balance was composed of three main parts: parts of processing of measured data, construction of an internal model and detection of the loss of balance. The internal model represented a simple dynamic motion balancing with two rear legs of a four-legged chair and was a simplified model of a central nervous system of a person. The internal model was defined by the experimental data obtained within a fixed time interval, and was applied to the detecting algorithm to the end of the experiment without being changed. The balancing motion controlled by the human brain was improved in process of time because of the experience accruing to the brain from controlling sensory organs. In this study a reconstruction method of the internal model was used in order to improve the success rate and the detecting time of the algorithm and was changed with time the same as the brain did. When using the reconstruction method, the success rate and the detecting time were 95 % and 0.729 sec, respectively and those results were improved by about 7.6 % and 0.25 sec in comparison to the results of the paper of Ahmed and Ashton-Miller. The results showed that the proposed reconstruction method of the internal model was efficient to improve the detecting performance of the algorithm.

• 한국수자원학회:학술대회논문집
• /
• 한국수자원학회 2019년도 학술발표회
• /
• pp.160-160
• /
• 2019

Water is a limited source that needs to be properly managed and distributed to the ever-growing population of the world. Rapid urbanization and development have increased the overall water demand of the world drastically. However, there is loss of billions of liters of water every year due to leakages in water distribution systems. Such water loss means significant financial loss for the utilities as well. World bank estimates a loss of $14 billion annually from wasted water. To address these issues and for the development of efficient and reliable leakage management techniques, high efforts have been made by the researchers and engineers. Over the past decade, various techniques and technologies have been developed for leakage management and leak detection. These include ideas such as pressure management in water distribution networks, use of Advanced Metering Infrastructure, use of machine learning algorithms, etc. For leakage detection, techniques such as acoustic technique, and in recent yeats transient test-based techniques have become popular. Smart Water Grid uses two-way real time network monitoring by utilizing sensors and devices in the water distribution system. Hence, valuable real time data of the water distribution network can be collected. Best results and outcomes may be produced by proper utilization of the collected data in unison with advanced detection and management techniques. Long term reduction in Non Revenue Water can be achieved by detecting, localizing and repairing leakages as quickly and as efficiently as possible. However, there are still numerous challenges to be met and future research works to be conducted in this field.

• Smart Structures and Systems
• /
• 제30권6호
• /
• pp.673-686
• /
• 2022

Deep learning algorithms for Structural Health Monitoring (SHM) have been extracting the interest of researchers and engineers. These algorithms commonly used loss functions and evaluation indices like the mean square error (MSE) which were not originally designed for SHM problems. An updated loss function which was specifically constructed for deep-learning-based structural damage detection problems has been proposed in this study. By tuning the coefficients of the loss function, the weights for damage localization and quantification can be adapted to the real situation and the deep learning network can avoid unnecessary iterations on damage localization and focus on the damage severity identification. To prove efficiency of the proposed method, structural damage detection using convolutional neural networks (CNNs) was conducted on a truss bridge model. Results showed that the validation curve with the updated loss function converged faster than the traditional MSE. Data augmentation was conducted to improve the anti-noise ability of the proposed method. For reducing the training time, the normalized modal strain energy change (NMSEC) was extracted, and the principal component analysis (PCA) was adopted for dimension reduction. The results showed that the training time was reduced by 90% and the damage identification accuracy could also have a slight increase. Furthermore, the effect of different modes and elements on the training dataset was also analyzed. The proposed method could greatly improve the performance for structural damage detection on both the training time and detection accuracy.