• Journal of the Korea Institute of Military Science and Technology
• /
• v.25 no.4
• /
• pp.381-389
• /
• 2022

Detecting partially occluded objects is difficult due to the appearances and shapes of occluders are highly variable. These variabilities lead to challenges of localizing accurate bounding box or classifying objects with visible object parts. To address these problems, we propose a two-stage part-based attention approach for robust object detection under partial occlusion. First, our part attention network(PAN) captures the important object parts and then it is used to generate weighted object features. Based on the weighted features, the re-weighted object features are produced by our reinforced PAN(RPAN). Experiments are performed on our collected military vehicle dataset and synthetic occlusion dataset. Our method outperforms the baselines and demonstrates the robustness of detecting objects under partial occlusion.

• Journal of Advanced Marine Engineering and Technology
• /
• v.39 no.5
• /
• pp.577-585
• /
• 2015

This paper deals with development of an autopilot system for leisure yacht based on NMEA 2000 network and android platform. The developed system can operate both for manual steering and automatic navigation mode. In automatic steering mode, after manipulation of commands which are NMEA 0183 sentences by android platform, the developed system translates and sends the packets through NMEA 2000 network. Then the controller which is connected to NMEA 2000 network receives the commands and controls the boat's rudder system automatically. The automatic steering mode is achieved by cooperation of two controllers; one for controlling the rudder system, and the other for controlling the vessel's heading. To control the vessel's rudder and heading angle two PID controllers are developed with an adjustable dead-band gain. Also, in order to eliminate the steady-state error occurred by applying dead-band, an integral controller which specifically supervises the system's behavior inside the dead-band area is developed. In this paper, at the first stage, simulations are accomplished using computer in order to examine the feasibility of the proposed based on simulation results. In the next step, the system on a real hydraulic steering model is implemented and at the end the performance examination by implementing it on a real boat and doing test navigation is executed.

• Proceedings of the KIEE Conference
• /
• 1999.07b
• /
• pp.988-990
• /
• 1999

Using neural network approach, the diagnosis of faults in industrial process that requires observing multiple data simultaneously are studied in this paper. Two-stage diagnosis is proposed as the basic structure. The first stage detects the dynamic trend of each measurements and the second stage diagnosis the faults. This paper makes up for the disadvantage of neural about unknown faults. The potential of this approach is demonstrated in simulation using a model of tank reactor.

• The Korean Journal of Applied Statistics
• /
• v.16 no.2
• /
• pp.321-333
• /
• 2003

Self-organizing map (SOM), a unsupervised learning neural network, has been developed by T. Kohonen since 1980's. Main application areas were pattern recognition and text retrieval. Because of that, it has not been spread to statisticians until late. Recently, SOM's are frequently drawn in data mining fields. Kohonen's SOM, however, needs improvements to become a statistician's standard tool. First, there should be a good guideline as for the size of map. Second, an enhanced visualization mode is wanted. In this study, principal components self-organizing map (PC-SOM), a modification of Kohonen's SOM, is proposed to meet such needs. PC-SOM performs one-dimensional SOM during the first stage to decompose input units into node weights and residuals. At the second stage, another one-dimensional SOM is applied to the residuals of the first stage. Finally, by putting together two stages, one obtains two-dimensional SOM. Such procedure can be easily expanded to construct three or more dimensional maps. The number of grid lines along the second axis is determined automatically, once that of the first axis is given by the data analyst. Furthermore, PC-SOM provides easily interpretable map axes. Such merits of PC-SOM are demonstrated with well-known Fisher's iris data and a simulated data set.

• Earthquakes and Structures
• /
• v.4 no.5
• /
• pp.509-525
• /
• 2013

In this paper, learning capabilities of two types of Arterial Neural Networks, namely hierarchical neural networks and Generalized Regression Neural Network were used in a two-stage approach to develop a method for generating spatial varying accelerograms from acceleration response spectra and a distance parameter in which generated accelerogram is desired. Data collected from closely spaced arrays of seismographs in SMART-1 array were used to train neural networks. The generated accelerograms from the proposed method can be used for multiple support excitations analysis of structures that their supports undergo different motions during an earthquake.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.9
• /
• pp.3712-3729
• /
• 2020

Smoke detection is helpful for early fire detection. With its large coverage area and low cost, vision-based smoke detection technology is the main research direction of outdoor smoke detection. We propose a two-stage smoke detection method combined with block Deep Normalization and Convolutional Neural Network (DNCNN) and visual change image. In the first stage, each suspected smoke region is detected from each frame of the images by using block DNCNN. According to the physical characteristics of smoke diffusion, a concept of visual change image is put forward in this paper, which is constructed by the video motion change state of the suspected smoke regions, and can describe the physical diffusion characteristics of smoke in the time and space domains. In the second stage, the Support Vector Machine (SVM) classifier is used to classify the Histogram of Oriented Gradients (HOG) features of visual change images of the suspected smoke regions, in this way to reduce the false alarm caused by the smoke-like objects such as cloud and fog. Simulation experiments are carried out on two public datasets of smoke. Results show that the accuracy and recall rate of smoke detection are high, and the false alarm rate is much lower than that of other comparison methods.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.18 no.3
• /
• pp.551-569
• /
• 2024

With the rapid development of electric vehicles (EVs) industry, EV charging service becomes more and more important. Especially, in the case of suddenly drop of air temperature or open holidays that large-scale EVs seeking for charging devices (CDs) in a short time. In such scenario, inefficient EV charging scheduling algorithm might lead to a bad service quality, for example, long queueing times for EVs and unreasonable idling time for charging devices. To deal with this issue, this paper propose a Deep-Q-Network (DQN) based two-stage scheduling method for the large-scale EVs charging service. Fine-grained states with two delicate neural networks are proposed to optimize the sequencing of EVs and charging station (CS) arrangement. Two efficient algorithms are presented to obtain the optimal EVs charging scheduling scheme for large-scale EVs charging demand. Three case studies show the superiority of our proposal, in terms of a high service quality (minimized average queuing time of EVs and maximized charging performance at both EV and CS sides) and achieve greater scheduling efficiency. The code and data are available at THE CODE AND DATA.

• Korea-Australia Rheology Journal
• /
• v.14 no.4
• /
• pp.189-201
• /
• 2002

The interaction between hydrophobically modified hydroxyethyl cellulose (hmHEC), containing approximately 1 wt% side-alkyl chains of $C_{16}$, and an anionic sodium dodecyl sulphate (SDS) surfactant was investigated. For a semi-dilute solution of 0.5 wt% hmHEC, the previously observed behaviour of a maximum in solution viscosity at intermediate SDS concentrations, followed by a drop at higher SDS concentrations, until above the cmc of surfactant when the solution resembles that of the unsubstituted polymer, was confirmed. Additionally, a two-phase region containing a hydrogel phase and a water-like supernatant was found at low SDS concentrations up to 0.2 wt%, a concentration which is akin to the critical association concentration, cac, of SDS in the presence of hmHEC. Above this concentration, SDS molecules bind strongly to form mixed micellar aggregates with the polymer alkyl side-chains, thus strengthening the network junctions, resulting in the observed increase in viscosity and elastic modulus of the solution. The shear behaviour of this polymer-surfactant complex during steady and step stress experiments was examined In great detail. Between SDS concentrations of 0.2 and 0.25 wt%, the shear viscosity of the hmHEC-polymer complex network undergoes shear-induced thickening, followed by a two-stage shear-induced fracture or break-up of the network. The thickening is thought to be due to structural rearrangement, causing the network of flexible polymers to expand, enabling some polymer hydrophobic groups to be converted from intra- to inter-chain associations. At higher applied stress, a partial local break-up of the network occurs, while at even higher stress, above the critical or network yield stress, a complete fracture of the network into small microgel-like units, Is believed to occur. This second network rupture is progressive with time of shear and no steady state in viscosity was observed even after 300 s. The structure which was reformed after the cessation of shear is found to be significantly different from the original state.

• The Journal of the Convergence on Culture Technology
• /
• v.8 no.2
• /
• pp.385-391
• /
• 2022

Quality evaluation of beef carcasses is an important issue in the livestock industry. Recently, through the AI monitor system based on artificial intelligence, the quality manager can receive help in making accurate decisions based on the analysis of beef carcass images or result information. This artificial intelligence dataset is an important factor in judging performance. Existing datasets may have different surface orientation or resolution. In this paper, we proposed a two-stage classification model that can efficiently manage the grades of beef carcass image using deep learning. And to overcome the problem of the various conditions of the image, a new dataset of 1,300 images was constructed. The recognition rate of deep network for 5-grade classification using the new dataset was 72.5%. Two-stage evaluation is a method to increase reliability by taking advantage of the large difference between grades 1++, 1+, and grades 1 and 2 and 3. With two experiments using the proposed two stage model, the recognition rates of 73.7% and 77.2% were obtained. As this, The proposed method will be an efficient method if we have a dataset with 100% recognition rate in the first stage.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.25 no.4
• /
• pp.27-44
• /
• 2000

In this study we consider a spares provisioning problem for repairable items in which a parts inventory system is incorporated. If a machine fails, a replacement part must be obtained at the parts inventory system before the failed machine enters the repair center. The inventory policy adopted at the parts inventory system is the (S, Q) policy. Operating times of the machine before failure, ordering lead times and repair times are assumed to follow a two-stage Coxian distribution. For this system, we develop an approximation method to obtain the performance measures such as steady state probabilities of the number of machines at each station and the probability that a part will wait at the parts inventory system. For the analysis of the proposed system, we model the system as a closed queueing network and analyze it using a product-form approximation method. A recursive technique as well as an iterative procedure is used to analyze the sub-network. Numerical tests show that the approximation method provides fairly good estimation of the performance measures of interest.