• Journal of Advanced Navigation Technology
• /
• v.26 no.6
• /
• pp.522-527
• /
• 2022

Object tracking is a field of signal processing that sequentially tracks the location of an object based on the previous-time location estimations and the present-time observation data. In this paper, we propose an adaptive scaling neural network that can track and adjust the scale of the input data with three recursive neural network (RNN) submodules. To evaluate object tracking performance, we compare the proposed system with the Kalman filter and the maximum likelihood object tracking scheme under an one-dimensional object movement model in which the object moves with piecewise constant acceleration. We show that the proposed scheme is generally better, in terms of root mean square error (RMSE) performance, than maximum likelihood scheme and Kalman filter and that the performance gaps grow with increased observation noise.

• Earthquakes and Structures
• /
• v.3 no.6
• /
• pp.839-852
• /
• 2012

Earthquake records are often analyzed in various earthquake engineering problems, making time-frequency analysis for such records of primary concern. The best tool for such analysis appears to be based on wavelet functions; selection of which is not an easy task and is commonly carried through trial and error process. Furthermore, often a particular wavelet is adopted for analysis of various earthquakes irrespective of record's prime characteristics, e.g. wave's magnitude. A wavelet constructed based on records' characteristics may yield a more accurate solution and more efficient solution procedure in time-frequency analysis. In this study, a low-pass reconstruction filter is obtained for each earthquake record based on multi-resolution decomposition technique; the filter is then assigned to be the normalized version of the last approximation component with respect to its magnitude. The scaling and wavelet functions are computed using two-scale relations. The calculated wavelets are highly efficient in decomposing the original records as compared to other commonly used wavelets such as Daubechies2 wavelet. The method is further advantageous since it enables one to decompose the original record in such a way that a clear time-frequency resolution is obtained.

• Korean Journal of Ecology and Environment
• /
• v.40 no.2
• /
• pp.264-272
• /
• 2007

Runoff of non-point pollutants has affected bad influence to water quality of river as reaching within short time. For this reason, reducing them prior to reaching aquatic systems or treating them after collection from discharge process of pollutants are desirable for efficient treatment of pollutants. This study was carried out to develop an ecotechnological method to prevent further aggravation of water quality by non-point source through vegetation filter strips. This study has placed a focus on improving infiltration capacity of soil for the optimum condition of vegetation filter strips. Therefore, we used titled soil filter strips instead of vegetation filter strips in this study. The three types of soil tilter strips were used in a bench scale experiment before applying to the field. The reduction efficiency of pollutants in soil filter strips (SS $84.5{\sim}92.5%$, BOD $67.9{\sim}80.6%$, T-N $43.4{\sim}76.6%$, T-P $40.6{\sim}87.4%$, Cu $28.3{\sim}48.1%$ Fe $92.1{\sim}97.7%$, Pb $81.4{\sim}97.3%$) was much higher than that of the controled group. And non-point pollutants reduction efficiency by soil filter strip's forms was estimated to be distinguishing in order of bio material, mixture of sand and gravel and lastly the whole gravel. In the event, the whole reduction efficiency of pollutants on the soil filter strips disclosed good results.

• IEIE Transactions on Smart Processing and Computing
• /
• v.3 no.3
• /
• pp.128-134
• /
• 2014

A real-time, continuous-scale image interpolation method is proposed based on a bilinear interpolation with directionally adaptive low-pass filtering. The proposed algorithm was optimized for hardware implementation. The ordinary bi-linear interpolation method has blocking artifacts. The proposed algorithm solves this problem using directionally adaptive low-pass filtering. The algorithm can also solve the severe blurring problem by selectively choosing low-pass filter coefficients. Therefore, the proposed interpolation algorithm can realize a high-quality image scaler for a range of imaging systems, such as digital cameras, CCTV and digital flat panel displays.

• Journal of Mechanical Science and Technology
• /
• v.20 no.12
• /
• pp.2284-2291
• /
• 2006

The present study investigates gas residence time and mixing characteristics for various swirl numbers generated by injection of secondary air into a lab-scale cylindrical combustor. Fine dust particles and butane gas were injected into the test chamber to study the gas residence time and mixing characteristics, respectively. The mixing characteristics were evaluated by standard deviation value of trace gas concentration at different measurement points. The measurement points were located 25 mm above the secondary air injection position. The trace gas concentration was detected by a gas analyzer. The gas residence time was estimated by measuring the temporal pressure difference across a filter media where the particles were captured. The swirl number of 20 for secondary air injection angle of 5$^{\circ}$ gave the best condition: long gas residence time and good mixing performance. Numerical calculations were also carried out to study the physical meanings of the experimental results, which showed good agreement with numerical results.

• Journal of Korean Society of Water and Wastewater
• /
• v.24 no.4
• /
• pp.377-386
• /
• 2010

To identify operating parameters of the up-flow filtering system, which is available to discharge filtering residue after the rain, developed for treatment of urban storm runoff, lab scale test was carried out. Removal efficiency of SS was 68.7%, 62.2%, and 58.6% at the area roading rate of 2.46m/h, 4.68m/h, and 10m/h, respectively, filtering device is desirable to operate at the lower than 4.68m/h of area roading rate to get higher level of 60% SS removal efficiency. The removal efficiency of SS was 57.1% ~ 68.7% at the raw water SS of 100mg/L ~ 600mg/L, and the SS in treated water was maintained at the constant level through the elapsed time. It is indicate that filtering device can guarantee a certain level of effluent water quality at various raw water quality. The removal efficiency of SS to the depth of filter media was 68.3%, 78.6% at the filter depth of 10 cm, 20cm respectively. The final treated water quality was showed 30.2mg/L of CODMn, 1.60mg/L of TN and 0.25mg/L of TP. The average removal efficiencies by filtering device developed in this research were recorded slightly lower levels than other research. The main reason of these results were the first, the filter depth of the media used in this test was shallow, the second, the kind of filter media in discharge port of residue. More research to kind of filter media, filter packing rate, select of media for residue discharge port should be go on to produce optimum operating condition. The result of this study would be valuable for the application of filtration device to control of urban storm water.

• Proceedings of the KIEE Conference
• /
• 2004.05a
• /
• pp.141-143
• /
• 2004

This paper presents an algorithm in which a feature map is built and localization of a mobile robot is carried out for indoor environments. The algorithm proposes an approach which extracts scale-invariant features of natural landmarks from a pair of stereo images. The feature map is built using these features and updated by merging new landmarks into the map and removing transient landmarks over time. And the position of the robot in the map is estimated by comparing with the map in a database by means of an Extended Kalman filter. This algorithm is implemented and tested using a Pioneer 2-DXE and preliminary results are presented in this paper.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.10 no.11
• /
• pp.5624-5638
• /
• 2016

A novel method based on Surf points is proposed to detect and lock-track single ground target in aerial videos. Videos captured by moving cameras contain complex motions, which bring difficulty in moving object detection. Our approach contains three parts: moving target template detection, search area estimation and target tracking. Global motion estimation and compensation are first made by grids-sampling Surf points selecting and matching. And then, the single ground target is detected by joint spatial-temporal information processing. The temporal process is made by calculating difference between compensated reference and current image and the spatial process is implementing morphological operations and adaptive binarization. The second part improves KALMAN filter with surf points scale information to predict target position and search area adaptively. Lastly, the local Surf points of target template are matched in this search region to realize target tracking. The long-term tracking is updated following target scaling, occlusion and large deformation. Experimental results show that the algorithm can correctly detect small moving target in dynamic scenes with complex motions. It is robust to vehicle dithering and target scale changing, rotation, especially partial occlusion or temporal complete occlusion. Comparing with traditional algorithms, our method enables real time operation, processing $520{\times}390$ frames at around 15fps.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.1
• /
• pp.305-326
• /
• 2019

Discriminative correlation filter (DCF) based tracking algorithms have recently shown impressive performance on benchmark datasets. However, amount of recent researches are vulnerable to heavy occlusions, irregular deformations and so on. In this paper, we intend to solve these problems and handle the contradiction between accuracy and real-time in the framework of tracking-by-detection. Firstly, we propose an innovative strategy to combine the template and color-based models instead of a simple linear superposition and rely on the strengths of both to promote the accuracy. Secondly, to enhance the discriminative power of the learned template model, the spatial regularization is introduced in the learning stage to penalize the objective boundary information corresponding to features in the background. Thirdly, we utilize a discriminative multi-scale estimate method to solve the problem of scale variations. Finally, we research strategies to limit the computational complexity of our tracker. Abundant experiments demonstrate that our tracker performs superiorly against several advanced algorithms on both the OTB2013 and OTB2015 datasets while maintaining the high frame rates.

• The Journal of the Acoustical Society of Korea
• /
• v.14 no.2
• /
• pp.50-61
• /
• 1995

A time scale modification method yielding rate-modified speech while conserving the characteristic of speech was implemented in real-time using a goneral purpose digital signal processor. Time scale modification changed pronunciation speed only, producing a time difference between the input signal and the modified signal, making it impossible to implement it in real-time. In this thesis, a system was implemented to remove the time difference between the input and modified signals. Speech signals slowed down or speeded up by a physical time scale modification method, such as adjusting the motor speed of the cassett tape recorder, was used as the input signal. Physical modification that controled only the inter speed of the cassette tape player distorted the pitch period of the original speech. In this study, a real-time system was implemented so that the pitch-distorted speech was reconstructed back to the original by fractional sampling pitch shifting using an FIR filter, and this signal was time scale modified to match the cassette tape recorder motor speed using SOLA time-scale medification. In experiments using speech signals medifiedby the proposed method, results obtained using a 16-bit resolution ADSP2101 processor and using computer simulations employing floating point operations showed about the same average frame signal-to-noise ratio of about 20 dB.