• Journal of Information Processing Systems
• /
• 제19권5호
• /
• pp.614-630
• /
• 2023

Dead trees significantly impact forest production and the ecological environment and pose constraints to the sustainable development of forests. A lightweight YOLOv4 dead tree detection algorithm based on unmanned aerial vehicle images is proposed to address current limitations in dead tree detection that rely mainly on inefficient, unsafe and easy-to-miss manual inspections. An improved logarithmic transformation method was developed in data pre-processing to display tree features in the shadows. For the model structure, the original CSPDarkNet-53 backbone feature extraction network was replaced by MobileNetV3. Some of the standard convolutional blocks in the original extraction network were replaced by depthwise separable convolution blocks. The new ReLU6 activation function replaced the original LeakyReLU activation function to make the network more robust for low-precision computations. The K-means++ clustering method was also integrated to generate anchor boxes that are more suitable for the dataset. The experimental results show that the improved algorithm achieved an accuracy of 97.33%, higher than other methods. The detection speed of the proposed approach is higher than that of YOLOv4, improving the efficiency and accuracy of the detection process.

• 대한전기학회논문지:전력기술부문A
• /
• 제50권6호
• /
• pp.275-278
• /
• 2001

This paper presents an algorithm for detecting current transformer (CT) saturation. At the instants of beginning (or end) of saturation, as a magnetizing inductance of the core is changed significantly, the shapes of the secondary current are also changed significantly though secondary currents are continuous the instants. At the instants, the second-order of third-order difference of the secondary current has big values. Thus, the third difference of the current is used to detect the beginning/end of CT saturation in this paper. If the magnitude of third-order difference of the secondary current is larger than a threshold value, the CT begins of ends saturation at the instants. The proposed detection method is unaffected by the amount of residual flux. The results of various tests with residual flux from -80% to +80% indicate satisfactory performance of the method.

• 대한전기학회논문지:전력기술부문A
• /
• 제53권7호
• /
• pp.382-388
• /
• 2004

The most widely used primary protection for the internal fault detection of power transformers is current percentage differential relaying(PDR). However, the harmonic components could be decreased by magnetizing inrush when there have been changes to the material of iron core or its design methodology. The higher the capacitance of high voltage status and underground distribution, the more differential current includes the second harmonic component during occurrence of an internal fault. Therefore, the conventional harmonic restraint methods need modification. This paper proposes an advanced protective relaying algorithm by fluxt-differential current slope characteristic and trend of voltage and differential current. To evaluate the performance of proposed algorithm, we have made comparative studies of PDR fuzzy relaying, and DWT relaying. The paper is constructed power system model including power transformer, utilizing the WatATP99, and data collection is made through simulation of various internal faults and inrush. As the results of test. the new proposed algorithm was proven to be faster and more reliable.

• 대한전기학회논문지:시스템및제어부문D
• /
• 제55권3호
• /
• pp.115-118
• /
• 2006

In the steel industry, a steel melting electric arc furnace is used to produce both carbon and alloy steels. Steel scrap which is charged into the furnace is heated by means of electric arc between graphite electrodes and the scrap. In this melting process, current is supplied to the furnace through HCC(high current cable) which connect the furnace and transformer. Four HCCs are assigned to each phase in our system to divide the current. Since a sudden cable breaking result in the shutdown of melting process, an aging detection of HCC is very important for both an improvement of productivity and cost reduction. In this paper, the aging of the HCC is estimated by using the current ratio between four HCCs.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제16권7호
• /
• pp.2131-2153
• /
• 2022

To achieve accurate detection of tuberculosis (TB) areas in chest radiographs, we design a chest X-ray TB area detection algorithm. The algorithm consists of two stages: the chest X-ray TB classification network (CXTCNet) and the chest X-ray TB area detection network (CXTDNet). CXTCNet is used to judge the presence or absence of TB areas in chest X-ray images, thereby excluding the influence of other lung diseases on the detection of TB areas. It can reduce false positives in the detection network and improve the accuracy of detection results. In CXTCNet, we propose a channel attention mechanism (CAM) module and combine it with DenseNet. This module enables the network to learn more spatial and channel features information about chest X-ray images, thereby improving network performance. CXTDNet is a design based on a sparse object detection algorithm (Sparse R-CNN). A group of fixed learnable proposal boxes and learnable proposal features are using for classification and location. The predictions of the algorithm are output directly without non-maximal suppression post-processing. Furthermore, we use CLAHE to reduce image noise and improve image quality for data preprocessing. Experiments on dataset TBX11K show that the accuracy of the proposed CXTCNet is up to 99.10%, which is better than most current TB classification algorithms. Finally, our proposed chest X-ray TB detection algorithm could achieve AP of 45.35% and AP50 of 74.20%. We also establish a chest X-ray TB dataset with 304 sheets. And experiments on this dataset showed that the accuracy of the diagnosis was comparable to that of radiologists. We hope that our proposed algorithm and established dataset will advance the field of TB detection.

• 순환기질환의공학회:학술대회논문집
• /
• 순환기질환의공학회 2006년도 제6회 학술대회
• /
• pp.38-39
• /
• 2006

• 전력전자학회:학술대회논문집
• /
• 전력전자학회 2011년도 추계학술대회
• /
• pp.279-280
• /
• 2011

In order to convert the mechanical movement of a linear generator to electrical power, the amateur current of the generator is controlled in accordance to the mover position. A linear encoder, usually used for direct detection of the mover position, not only is vulnerable to mechanical vibration, but also imposes significant constraint on the mechanical design of the generator system. Thus, this study proposes a method for indirect estimation of the mover position with emfs induced in amateur coils. The estimation algorithm is validated with simulation study.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2007년도 추계학술대회 논문집 전력기술부문
• /
• pp.75-77
• /
• 2007

Current percentage differential relaying has been recognized as the principal basis of main protection for stator windings of AC generator. The DWT has merit of obtaining frequency characteristics in time domain. In order to compensate for DFT's defects, we proposed fault detection algorithm using DWT. This paper describes a comparative analysis about conventional DFT-based DFR and advanced DWT-based relaying.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 1998년도 하계종합학술대회논문집
• /
• pp.955-957
• /
• 1998

Switched reluctance motor(SRM) has the advantages of simple structure, low rotor inertia, and high poer rate per unit volume. However, position sensor isessential in SRM in order to synchronize the phase excitation to the rotor position. The position sensors increase the cost of drive system, and tend to reduce system reliability. This paper investigtes the speed control of sensorless SRM. The proposed system consists of position detection circuit, dwell angle controller, digital logic commutator, PI speed controller and 4-phase inverter. The performances in the proposed system are verified through the experiment.

• Journal of the Korean Data and Information Science Society
• /
• 제18권4호
• /
• pp.847-857
• /
• 2007

This paper provides a statistical algorithm which detects COD (chemical oxygen demand) measurement machine error on real-time. For this we propose to use regression model fitting and check its validity against the current observations. The main idea is that the normal regression relation between COD measurement and other parameters inside the machine will be violated when the machine is out of order.