• Analytical Science and Technology
• /
• v.28 no.6
• /
• pp.453-459
• /
• 2015

Split-flow thin cell fractionation (SPLITT fractionation, SF) is a particle separation technique that allows continuous (and thus a preparative scale) separation into two subpopulations based on the particle size or the density. In SF, there are two basic performance parameters. One is the throughput (TP), which was defined as the amount of sample that can be processed in a unit time period. Another is the fractionation efficiency (FE), which was defined as the number % of particles that have the size predicted by theory. Full-feed depletion mode (FFD-SF) have only one inlet for the sample feed, and the channel is equipped with a flow stream splitter only at the outlet in SF mode. In conventional FFD-mode, it was difficult to extend channel due to splitter in channel. So, we use large scale splitter-less FFD-SF to increase TP from increase channel scale. In this study, a FFD-SF channel was developed for a large-scale fractionation, which has no flow stream splitters (‘splitter less’), and then was tested for optimum TP and FE by varying the sample concentration and the flow rates at the inlet and outlet of the channel. Polyurethane (PU) latex beads having two different size distribution (about 3~7 µm, and about 2~30 µm) were used for the test. The sample concentration was varied from 0.2 to 0.8% (wt/vol). The channel flow rate was varied from 70, 100, 120 and 160 mL/min. The fractionated particles were monitored by optical microscopy (OM). The sample recovery was determined by collecting the particles on a 0.1 µm membrane filter. Accumulation of relatively large micron sized particles in channel could be prevented by feeding carrier liquid. It was found that, in order to achieve effective TP, the concentration of sample should be at higher than 0.4%.

• Journal of Life Science
• /
• v.18 no.8
• /
• pp.1049-1052
• /
• 2008

Microbial production of cellulolytic enzymes by Aspergillus niger in solid state fermentation (SSF) and submerged state fermentation (SF) in laboratory scale was compared. Czapek Dox liquid broth amended with cellulose (0.5%) was used for cultivation in SF, whereas rice bran was used as a solid support, moistened with cellulose, amended Czapek Dox broth for growth in SSF. The production of Carboxymethyl cellulase, Filter paperase and ${\beta}$-Glucosidase was monitored at regular intervals. The peak production of the enzymes occurred within 3 days of incubation in SSF as against $\geq$ 7 days in SF. SSF gave higher yields of enzymes in comparison to SF. Maximum titres of 0.40, 0.62 and 0.013 U/ml in respect of FPase, CMCase and ${\beta}$-glucosidase in SSF were recovered as against 0.13, 0.06 and 0.0013 U/ml in SF respectively, at their respective peak time intervals. Hence, SSF appeared to be a better choice for production of cellulolytic enzymes by Aspergillus niger.

• Journal of Korean Society of Environmental Engineers
• /
• v.30 no.6
• /
• pp.601-608
• /
• 2008

The objective of this study was to evaluate the effect of ozonation as pretreatment on the removal of dissolved or biodegradable organic matter(DOM or BOM), the variance of DOM fractionation, and microbial regrowth by pilot-scale granular activated carbon processes in which adsorption and biodegradability was proceeding due to long time operation. Regardless of point of ozonation applied, GAC processes with ozonation(i.e., Ozonation combined with GAC Filter-adsorber; Pre O$_3$ + F/A, Ozonation combined with GAC adsorber; Post O$_3$ + GAC) compared with GAC processes without ozonation(i.e., GAC Filter-adsorber; F/A, GAC adsorber; GAC) removed approximately 10 to 20% more of DOC, hydrophilic DOM(HPI), BDOC and AOC after long period of operation that biological activity was assumed to happen. Ozonation was not found to have a significant effect on the removal of DOC, but caused the decrease of AOC by approximately 20%. It was found that the fixed bacterial biomass on GAC media did not show a significant difference between the GAC with ozonation and GAC without ozonation as pre-treatment, whereas the HPC of column effluent was more biostable at Post O$_3$ + GAC compared with F/A or GAC.

• Journal of KIISE:Computing Practices and Letters
• /
• v.7 no.3
• /
• pp.260-272
• /
• 2001

In this paper, we have designed a region-based image retrieval system, FRIP(Finding Region In the Pictures). This system includes a robust image segmentation scheme using color and texture direction and retrieval scheme based on features of each region. For image segmentation, by using a circular filter, we can protect the boundary of round object and merge stripes or spots of objects into body region. It also combines scaled and shifted color coordinate and texture direction. After image segmentation, in order to improve the storage management effectively and reduce the computation time, we extract compact features from each region and store as index. For user interface, by the user specified constraints such as color-care / don't care. scale-care / dont care, shape-care / dont care and location-care / dont care, the overal/ matching score is estimated and the top Ie nearest images are reported in the ascending order of the final score.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.51 no.9
• /
• pp.139-147
• /
• 2014

In this paper, we present a novel method for simultaneous detection and tracking of multiple objects using dual-layer particle filtering. The proposed dual-layer particle sampling (DLPS) algorithm consists of parent-particles (PP) in the first layer for detecting multiple objects and child-particles (CP) in the second layer for tracking objects. In the first layer, PPs detect persons using a classifier trained by the intersection kernel support vector machine (IKSVM) at each particle under a randomly selected scale. If a certain PP detects a person, it generates CPs, and makes an object model in the detected object region for tracking the detected object. While PPs that have detected objects generate CPs for tracking, the rest of PPs still move for detecting objects. Experimental results show that the proposed method can automatically detect and track multiple objects, and efficiently reduce the processing time using the sampled particles based on motion distribution in video sequences.

• Journal of Korean Society for Geospatial Information Science
• /
• v.14 no.3 s.37
• /
• pp.47-53
• /
• 2006

In this contribution, an integration of seamless navigation system for the pedestrian is introduced. To overcome the GPS outages in various situations, multi-sensor of GPS, INS, electronic barometer and compass are considered in one Extented Kalman filter. Especially, the integrated system is designed for low-cost for the practical applications. Therefore, a MEMS IMU is considered, and the low quality of the heading is compensated by the electronic compass. In addition, only the pseudoranges from GPS measurements are considered for possible real-time application so that the degraded height is also controlled by a barometer. The mathematical models for each sensor with systematic errors such as biases, scale factors are described in detail and the results are presented in terms of a covariance analysis as well as the position and attitude errors compared to the high-grade GPS/INS combined solutions. The real application scenario of GPS outage is also investigated to assess the feasible accuracy with respect to the outage period. The description on the current status of the development and future research directions are also stated.

• Proceedings of the Membrane Society of Korea Conference
• /
• 1994.06a
• /
• pp.45-58
• /
• 1994

Over the last 20 years, membrane separation processes, such as reverse osmosis, ultrafiltration and microfiltration, have been widely adopted by different industries. Commercial uses of membrane have displaced conventional separation processes, such as distillation, evaporation, precoat filter and so on. Membrane separation processes are often more capital and energy efficient when compared with conventional separation processes. Membrane devices and systems are almost always compact and modular. These are the well-known advantages of membrane separation processes. The disadvantage of the membrane process is that the process does not have scale merit and thus the membrane process is suitable for the small and middle size applications. Energy saving is, of course, the biggest advantage of the membrane process, and in many industries the membrane processes are employed because of this reason. Membrane process has other big advantage. In many applications membrane processes provide much higher quality of product than conventional processes. The example is ultrapure water production by membrane processes in semiconductor industry. Conventional technologies never offer such good quality of pure water. If you can obtain both energy saving and higher quality of product at the same time by membrane processes, this is the best application of membrane processes. One example is the concentration of orange juice by membrane, which has already been commercialized in Japan. Comparing with the conventional vacuum evaporation process, juice concentrated by the membrane process has much better taste and flavor and the energy consumption in the membrane process is much less than the evaporation process. In this paper, first membrane separation technology will be classified and then Japanese membrane manufacturers and new modules and devices under development in Japan will be introduced. Fourth energy saving in membrane process will be discussed and finally practical applications of membrane processes in Japan will be shown.

• KSBB Journal
• /
• v.15 no.5
• /
• pp.469-473
• /
• 2000

We studied the removal of toluene vapors in a lab-scale biofiter. Biofiltration was performed in a column fed in a downflow manner with contaminated air at ambient conditions. The column was packed with a mixture of peat and calstone(5:3 vol. Ratio), which was inoculated with microbes of selected stains(Pseudomonas putida type A). The microorganisms were immobilized on the filter media and biofilms were formed. The fiofilter was operated at various inlet toluene concentrations for days, and treated up to a maximum elimination capacity of $20 g/m^3hr$ at an inlet load of $30 g/m^3hr$, which corresponds to removal efficiencies in the range 20∼90% and a gas retention time of 1 to 2 min. The pressure drop was almost negligible over the biofilter columns, amounting to only $1.062 cmH_2O/m$ and appreciably smaller than other studies. The effects of operating conditions such as flow rate, inlet toluene concentration and moisture content on the performance of the biofilter were sequentially investigated.

• Proceedings of the Korea Committee for Ocean Resources and Engineering Conference
• /
• 2003.05a
• /
• pp.141-148
• /
• 2003

This paper presents a rotating ann test for assessment of an underwater hybrid navigation system for a semi-autonomous underwater vehicle. The navigation system consists of an inertial measurement unit (IMU), an ultra-short baseline (USBL) acoustic navigation sensor and a doppler velocity log (DVL) accompanying a magnetic compass. The errors of inertial measurement units increase with time due to the bias errors of gyros and accelerometers. A navigational system model is derived to include the error model of the USBL acoustic navigation sensor and the scale effect and bias errors of the DVL, of which the state equation composed of the navigation states and sensor parameters is 25 in the order. The conventional extended Kalman filter was used to propagate the error covariance, update the measurement errors and correct the state equation when the measurements are available. The rotating ann tests are conducted in the Ocean Engineering Basin of KRISO, KORDI to generate circular motion in laboratory, where the USBL system was absent in the basin. The hybrid underwater navigation system shows good tracking performance against the circular planar motion. Additionally this paper checked the effects of the sampling ratio of the navigation system and the possibility of the dead reckoning with the DVL and the magnetic compass to estimate the position of the vehicle.

• Journal of Electrical Engineering and Technology
• /
• v.9 no.2
• /
• pp.739-748
• /
• 2014

In this paper, we propose pyramid appearance and global structure action descriptors on both RGB and depth motion history images and a model-free method for human action recognition. In proposed algorithm, we firstly construct motion history image for both RGB and depth channels, at the same time, depth information is employed to filter RGB information, after that, different action descriptors are extracted from depth and RGB MHIs to represent these actions, and then multimodality information collaborative representation and recognition model, in which multi-modality information are put into object function naturally, and information fusion and action recognition also be done together, is proposed to classify human actions. To demonstrate the superiority of the proposed method, we evaluate it on MSR Action3D and DHA datasets, the well-known dataset for human action recognition. Large scale experiment shows our descriptors are robust, stable and efficient, when comparing with the-state-of-the-art algorithms, the performances of our descriptors are better than that of them, further, the performance of combined descriptors is much better than just using sole descriptor. What is more, our proposed model outperforms the state-of-the-art methods on both MSR Action3D and DHA datasets.