• The Journal of Korea Robotics Society
• /
• v.18 no.3
• /
• pp.346-351
• /
• 2023

With the release of numerous open driving datasets, the demand for domain adaptation in perception tasks has increased, particularly when transferring knowledge from rich datasets to novel domains. However, it is difficult to solve the change 1) in the sensor domain caused by heterogeneous LiDAR sensors and 2) in the environmental domain caused by different environmental factors. We overcome domain differences in the semi-supervised setting with 3-stage model parameter training. First, we pre-train the model with the source dataset with object scaling based on statistics of the object size. Then we fine-tine the partially frozen model weights with copy-and-paste augmentation. The 3D points in the box labels are copied from one scene and pasted to the other scenes. Finally, we use the knowledge distillation method to update the student network with a moving average from the teacher network along with a self-training method with pseudo labels. Test-Time Augmentation with varying z values is employed to predict the final results. Our method achieved 3rd place in ECCV 2022 workshop on the 3D Perception for Autonomous Driving challenge.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.19 no.1
• /
• pp.327-334
• /
• 2024

In this paper, we developed a system with a near-infrared LED light source with a wavelength of 950nm to acquire palm vein images and a white LED light source to acquire palm creases based on Raspberry Pi. In addition, we implemented a unique pattern-extractable image processing technology that can prevent counterfeiting and enhance security of mixed creases and palmprints through image pre-processing (Gray scaling, Histogram Equalization, Blurring, Thresholding, Thinning) for the acquired vein and palm images, and secured a source technology that can be used in a security-enhanced system.

• Tunnel and Underground Space
• /
• v.30 no.3
• /
• pp.214-225
• /
• 2020

Uniaxial compressive strength (UCS) of rock is one of the important factors to determine the advance speed during shield TBM tunnel excavation. UCS can be obtained through the Geotechnical Data Report (GDR), and it is difficult to measure UCS for all tunneling alignment. Therefore, the purpose of this study is to predict UCS by utilizing TBM machine driving data and machine learning technique. Several machine learning techniques were compared to predict UCS, and it was confirmed the stacking model has the most successful prediction performance. TBM machine data and UCS used in the analysis were obtained from the excavation of rock strata with slurry shield TBMs. The data were divided into 8:2 for training and test and pre-processed including feature selection, scaling, and outlier removal. After completing the hyper-parameter tuning, the stacking model was evaluated with the root-mean-square error (RMSE) and the determination coefficient (R²), and it was found to be 5.556 and 0.943, respectively. Based on the results, the sacking models are considered useful in predicting rock strength with TBM excavation data.

• The Journal of the Acoustical Society of Korea
• /
• v.31 no.7
• /
• pp.427-434
• /
• 2012

The overlap-add technique based on waveform similarity (WSOLA) method is known to be an efficient high-quality algorithm for time scaling of speech signal. The computational load of WSOLA is concentrated on the repeated normalized cross-correlation (NCC) calculation to evaluate the similarity between two signal waveforms. To reduce the computational complexity of WSOLA, this paper proposes a fast NCC computation method, in which NCC is obtained through pre-calculated sum tables to eliminate redundancy of repeated NCC calculations in the adjacent regions. While the denominator part of NCC has much redundancy irrespective of the time-scale factor, the numerator part of NCC has less redundancy and the amount of redundancy is dependent on both the time-scale factor and optimal shift value, thereby requiring more sophisticated algorithm for fast computation. The simulation results show that the proposed method reduces about 40%, 47% and 52% of the WSOLA execution time for the time-scale compression, 2 and 3 times time-scale expansions, respectively, while maintaining exactly the same speech quality of the conventional WSOLA.

• Journal of Ocean Engineering and Technology
• /
• v.28 no.5
• /
• pp.387-395
• /
• 2014

In this paper, a numerical analysis for an internal turret moored vessel located at a 400-m water depth is conducted. The target vessel has an internal turret that is located at the 0.2 Lpp position from the fore-side, with $3{\times}4$ complex mooring lines installed around the turret circumference. To investigate the motion response of the vessel and the structural reliability of the lines, model tests were conducted. The KRISO ocean basin has a water depth of 3.2 m, which represents 192m using a scaling of 1:60. In order to precisely represent the real-scale condition, equivalent mooring lines needed to be designed. Truncated mooring lines were designed to supplement the restriction of the flume's water depth and increase the reliability of the model testing. These truncated mooring lines were composed of two different chains in order to match the pre-tension, simultaneously restoring the curve and variation in the effective line tension. The static similarities were compared using a static pull-out test and free decaying test, and the dynamic similarities were matched via a regular wave test and combined environments test. Consequently, the designed truncated mooring system could represent the prototype mooring system relatively well in the aspects of kinematics and dynamics.

• Magazine of the Korea Concrete Institute
• /
• v.11 no.2
• /
• pp.221-230
• /
• 1999

Corrosion of rebars can occur if there are cracks, moisture and availability of oxygen or carbonation proceeds, chloride penetrates and diffuses in concrete. Once rebars in concrete corrodes, subsequently accompanied with scaling, spalling in concrete cover. As a result of them, the RC structure is seriously deteriorated. In this study, theoretical review and experiments for cathodic protection(CP) have been performed to control corrosion of rebars in concrete contained chlorides and pre-crack. For CP the impressed current system was applied, the protection effect was investigated when rebars was directly contacted with salt water due to crack and open to much chlorides in concrete. In order to investigate the effect of protection, when CP was energized for 1 year, half-cell potential, potential-decay with current density, corrosion ratio, etc. were measured. With the cathodic protection by impressed current system, the depolarized values of all specimen were met NACE Standard, the effect of 34~84% of the ratio of corrosion area and 84~86% of cross-section reduction were calculated.

• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.47 no.5
• /
• pp.1-9
• /
• 2010

In this paper, the novel architecture of Intra-prediction & DCTQ hardware, which can process for the Full HD image($1980{\times}1088$@30fps) in realtime, is proposed. The cycle optimization method for the overall cycle of prediction, transform, scaling, descaling, and reconstruction is proposed. To reduce the cycle in the $4{\times}4$ prediction, the quantization process is performed during the prediction cycle and pre-selection of 2 modes among the 9 modes is performed to reduce the hardware area. To reduce the hardware of $16{\times}16$ and $8{\times}8$ prediction, the sharing logic between 2 prediction is utilized. The proposed architecture can process the 30frame/sec of full HD image in 108 MHz clock and operate 425 cycle for one macroblock.

• Membrane and Water Treatment
• /
• v.11 no.4
• /
• pp.257-274
• /
• 2020

There are many previous review articles are available to summarize either the characterization methods of effluent organic matter (EfOM) or the individual control treatment options. However, there has been no attempt made to compare in parallel the physicochemical treatment options that target the removal of EfOM from biological treatments. This review deals with the recent progress on the characterization of EfOM and the novel technologies developed for EfOM treatment. Based on the publications after 2010, the advantages and the limitations of several popularly used analytical tools are discussed for EfOM characterization, which include UV-visible and fluorescence spectroscopy, Fourier transform infrared spectroscopy (FTIR), size exclusion chromatography (SEC), and Fourier transform-ion cyclotron resonance-mass spectrometry (FT-ICR-MS). It is a recent trend to combine an SEC system with various types of detectors, because it can successfully track the chemical/functional composition of EfOM, which varies across a continuum of different molecular sizes. FT-ICR-MS is the most powerful tool to detect EfOM at molecular levels. However, it is noted that this method has rarely been utilized to understand the changes of EfOM in pre-treatment or post-treatment systems. Although membrane filtration is still the preferred method to treat EfOM before its discharge due to its high separation selectivity, the minimum requirements for additional chemicals, the ease of scaling up, and the continuous operation, recent advances in ion exchange and advanced oxidation processes are greatly noteworthy. Recent progress in the non-membrane technologies, which are based on novel materials, are expected to enhance the removal efficiency of EfOM and even make it feasible to selectively remove undesirable fractions/compounds from bulk EfOM.

• Journal of Periodontal and Implant Science
• /
• v.30 no.1
• /
• pp.203-212
• /
• 2000

Putrefactive activity within the oral cavity is the principal cause of halitosis. The most common intraoral sites of oral malodor production are tongue, interdental and subgingival areas. The other foci may include faulty restorations, sites of food impaction and abscesses. Periodontal disease frequently involves pathological oral malodor, which is caused mainly by volatile sulfur compounds(VSC), such as hydrogen sulfide, methyl mercaptan, and dimethyl sulfide. The purpose of this study is to evaluate the association between oral malodor and periodontal status. Volatile sulfur compounds in mouth air were estimated by portable sulfide monitor($Halimeter^{TM}$). The results were as follows : 1. The levels of volatile sulfur compounds were significantly greater in a periodontitis group than in a control group(P<0.01). The amounts of VSC in mouth air from patients with periodontal involvement were four times greater than those of the control group. 2. The significant positive correlation was found between VSC concentrations and the number of pocket depth above 4mm(P<0.01), but correlation between VSC concentrations and plaque score was not statistically significant(P>0.05). 3. In the periodontitis group, VSC concentrations of pre-treatment significantly decreased after scaling and root planing(P<0.01). 4. No statistically significant correlation was found between VSC concentrations and sex / age in the periodontitis group. The above results indicate that periodontal disease may play a role as an important factor of oral malodor and deep periodontal pockets are a source of volatile sulfur compounds.

• International Journal of Computer Science & Network Security
• /
• v.22 no.5
• /
• pp.175-181
• /
• 2022

COVID-19 is an acute respiratory syndrome that affects the host's breathing and respiratory system. The novel disease's first case was reported in 2019 and has created a state of emergency in the whole world and declared a global pandemic within months after the first case. The disease created elements of socioeconomic crisis globally. The emergency has made it imperative for professionals to take the necessary measures to make early diagnoses of the disease. The conventional diagnosis for COVID-19 is through Polymerase Chain Reaction (PCR) testing. However, in a lot of rural societies, these tests are not available or take a lot of time to provide results. Hence, we propose a COVID-19 classification system by means of machine learning and transfer learning models. The proposed approach identifies individuals with COVID-19 and distinguishes them from those who are healthy with the help of Deep Visual Embeddings (DVE). Five state-of-the-art models: VGG-19, ResNet50, Inceptionv3, MobileNetv3, and EfficientNetB7, were used in this study along with five different pooling schemes to perform deep feature extraction. In addition, the features are normalized using standard scaling, and 4-fold cross-validation is used to validate the performance over multiple versions of the validation data. The best results of 88.86% UAR, 88.27% Specificity, 89.44% Sensitivity, 88.62% Accuracy, 89.06% Precision, and 87.52% F1-score were obtained using ResNet-50 with Average Pooling and Logistic regression with class weight as the classifier.