• Journal of IKEEE
• /
• v.21 no.4
• /
• pp.388-396
• /
• 2017

Since there are many variables such as various poses, illuminations and occlusions in a face detection problem, a high performance detection system is required. Although CNN is excellent in image classification, CNN operatioin requires high-performance hardware resources. But low cost low power environments are essential for small and mobile systems. So in this paper, the CPU-FPGA integrated system is designed based on 3-stage cascade CNN architecture using small size FPGA. Adaptive Region of Interest (ROI) is applied to reduce the number of CNN operations using face information of the previous frame. We use a Field Programmable Gate Array(FPGA) to accelerate the CNN computations. The accelerator reads multiple featuremap at once on the FPGA and performs a Multiply-Accumulate (MAC) operation in parallel for convolution operation. The system is implemented on Altera Cyclone V FPGA in which ARM Cortex A-9 and on-chip SRAM are embedded. The system runs at 30FPS with HD resolution input images. The CPU-FPGA integrated system showed 8.5 times of the power efficiency compared to systems using CPU only.

• The Korean Journal of Zoology
• /
• v.30 no.1
• /
• pp.53-70
• /
• 1987

The morphological study on the parenchymal cells in the adult planaria performed to observe their cytochemical and ultrastructural characteristics. The results are as follows. Nine types of cells are found in parenchyma. 1. Free parenchymal cell: These cells contain several chromatoid bodies around the nucleus. Heterochromatins are evenly dispersed in large nucleus. These cells are abundant in free ribosomes. 2. Fixed parenchymal cells: These cells have well-developed granular endoplasmic reticulum, mitochondria and Golgi complex but they contain the cytosols exhibiting electron-lucencies. 3. Rhabdite-forming cells: These cells contain the electron-dense rhabdite granules of up to about 0.3 x 0.9 $\mu$m in size. Rhabdite-forming cells have well-developed cell organelles, granular endolplasmic reticulum, mitochondria and Golgy complex. 4. A-type of basophilic granule cells: These cells contain irregularly-shaped granules exhibiting alcianophilia. These granules surrounded by a limited membrane, approximately 1.4 x 0.7 $\mu$m in size, are accumulated in the cytoplasm. 5. C-type of basophilic granule cells: These cells contain electron-dense granules of less than 0.2 $\mu$m in size, which exhibit PAS- positive reaction. This type of granule is also found in the muscle layer of parenchyma. 6. D-type of basophilic granule cells: This type of granule cell occurs only in the parenchyma around reproductive organ. The granules have cytochemical characteristics that they exhibit strongly positive reaction with PAS and weakly eosinophilic property. These electron-dense granules, which are 0.2 to 0.6 $\mu$m in length, have oval shapes. 7. E-type of basophilic granule cells: These cells are found only in the parenchyma around re productive organ. The granules contained in a small number in the cell, exhibit PAS-positive reaction and have an average size of 0. 2pm. 8. Eosinophilic granule cells: These cells contain a large number of eosinophilic granules which have relatively diverse sizes from 0.3 x 0.2 to 0.8 x 0.4 $\mu$m. Most of granules are round or irregularly-shaped and highly electrondense. These cells have an array of well-developed granular endoplasmic reticulum of which cisternae are distened. 9. Transparent granule cells contain electron-lucent granules which exhibit negative reactions with three kinds of cytochemical methods used in this experiment.

• Radiation Oncology Journal
• /
• v.27 no.2
• /
• pp.91-102
• /
• 2009

Purpose: To compare the accuracy and efficacy of EDR2 film, a 2D ionization chamber array (MatriXX) and an amorphous silicon electronic portal imaging device (EPID) in the pre-treatment QA of IMRT. Materials and Methods: Fluence patterns, shaped as a wedge with 10 steps (segments) by a multi-leaf collimator (MLC), of reference and test IMRT fields were measured using EDR2 film, the MatriXX, and EPID. Test fields were designed to simulate leaf positioning errors. The absolute dose at a point in each step of the reference fields was measured in a water phantom with an ionization chamber and was compared to the dose obtained with the use of EDR2 film, the MatriXX and EPID. For qualitative analysis, all measured fluence patterns of both reference and test fields were compared with calculated dose maps from a radiation treatment planning system (Pinnacle, Philips, USA) using profiles and $\gamma$ evaluation with 3%/3 mm and 2%/2 mm criteria. By measurement of the time to perform QA, we compared the workload of EDR2 film, the MatriXX and EPID. Results: The percent absolute dose difference between the measured and ionization chamber dose was within 1% for the EPID, 2% for the MatriXX and 3% for EDR2 film. The percentage of pixels with $\gamma$%>1 for the 3%/3 mm and 2%/2 mm criteria was within 2% for use of both EDR2 film and the EPID. However, differences for the use of the MatriXX were seen with a maximum difference as great as 5.94% with the 2%/2 mm criteria. For the test fields, EDR2 film and EPID could detect leaf-positioning errors on the order of -3 mm and -2 mm, respectively. However it was difficult to differentiate leaf-positioning errors with the MatriXX due to its poor resolution. The approximate time to perform QA was 110 minutes for the use of EDR2 film, 80 minutes for the use of the MatriXX and approximately 55 minutes for the use of the EPID. Conclusion: This study has evaluated the accuracy and efficacy of EDR2 film, the MatriXX and EPID in the pre-treatment verification of IMRT. EDR2 film and the EPID showed better performance for accuracy, while the use of the MatriXX significantly reduced measurement and analysis times. We propose practical and useful methods to establish an effective QA system in a clinical environment.

• Journal of Bio-Environment Control
• /
• v.18 no.3
• /
• pp.177-184
• /
• 2009

For the purpose of reducing the cost for greenhouse operation by acquiring the electric power necessary for it, this study installed a solar photovoltaic system on the roof of the building adjacent to green-houses and experimentally examined the quantity of power generation based on weather conditions. The results of the study are as per the below: The maximum, average and minimum temperature while the experiment was conducted was $0.4{\sim}34.1,\;-6.1{\sim}22.2$, and $-14.1{\sim}16.7^{\circ}C$ respectively, and the solar radiation was $28.8MJ{\cdot}m^{-2}$ (maximum), $14.9MJ{\cdot}m^{-2}$ (average), and $0.6MJ{\cdot}m^{-2}$ (minimum). The quantity of electric power didn't increase in proportion to the quantity of solar radiation and instead, it was almost consistent around 750W. Daily maximum, average and minimum consumption of electric power was 5.2kWh, 2.5kWh and 0kWh respectively. Based on the average electric power consumption of the system used for this experiment, it was sufficient in case the capacity and the working time of a hot blast heater are small, but it was short in case they are big. In case the capacity of the hot blast heater is big, the average electric power quantity will be sufficient for array area $21m^2$, about three times of the present area. In summer when the temperature of the array becomes high, the generation of electric power didn't increase in proportion to the quantity of solar radiation, but this experiment result shows a high correlation between two factors (coefficient of correlation 0.84).

• 한국신재생에너지학회:학술대회논문집
• /
• 2010.06a
• /
• pp.59.1-59.1
• /
• 2010

Aluminum-induced crystallization (AIC) of amorphous silicon (a-Si) is studied with the structure of a glass/Al/$SiO_2$/a-Si, in which the $SiO_2$ layer has micron-sized laser holes in the stack. An oxide layer between aluminum and a-Si thin films plays a significant role in the metal-induced crystallization (MIC) process determining the properties such as grain size and preferential orientation. In our case, the crystallization of a-Si is carried out only through the key hole because the $SiO_2$ layer is substantially thick enough to prevent a-Si from contacting aluminum. The crystal growth is successfully realized toward the only vertical direction, resulting a crystalline silicon grain with a size of $3{\sim}4{\mu}m$ under the hole. Lateral growth seems to be not occurred. For the AIC experiment, the glass/Al/$SiO_2$/a-Si stacks were prepared where an Al layer was deposited on glass substrate by DC sputter, $SiO_2$ and a-Si films by PECVD method, respectively. Prior to the a-Si deposition, a $30{\times}30$ micron-sized hole array with a diameter of $1{\sim}2{\mu}m$ was fabricated utilizing the femtosecond laser pulses to induce the AIC process through the key holes and the prepared workpieces were annealed in a thermal chamber for 2 hours. After heat treatment, the surface morphology, grain size, and crystal orientation of the polycrystalline silicon (pc-Si) film were evaluated by scanning electron microscope, transmission electron microscope, and energy dispersive spectrometer. In conclusion, we observed that the vertical crystal growth was occurred in the case of the crystallization of a-Si with aluminum by the MIC process in a small area. The pc-Si grain grew under the key hole up to a size of $3{\sim}4{\mu}m$ with the workpiece.

• Asian-Australasian Journal of Animal Sciences
• /
• v.19 no.6
• /
• pp.816-824
• /
• 2006

This study was conducted to evaluate the shoot height at which the yield and nutritive value of wormwood (Artemisia montana) is optimized in order to provide information on its potential to support animal production (Experiment 1). A second objective was to determine how the essential oil (EO) concentration in wormwood hay and silage differ (Experiment 2). In Experiment 1, Artemisia montana was harvested at five different shoot heights (20, 40, 60, 80 and 100 cm) from triplicate $1.8{\times}1.8m$ plots. Dry matter (DM) yield was measured at each harvest date and the harvested wormwood was botanically separated into leaf, stalk and whole plant fractions and analyzed for chemical composition and in vitro dry matter digestibility (DMD). Values for total digestible nutrients (TDN), digestible energy (DE) and metabolizable energy (ME) were subsequently calculated using prediction equations. Dry matter yields of stalk and whole plant increased linearly (p<0.001) and leaf yield increased quadratically (p<0.01) with shoot height, whereas the leaf/stalk ratio decreased linearly (p<0.001). As shoot height increased, there was a linear increase (p<0.001) in leaf DM, ether extract (EE) and neutral detergent fiber (NDF) contents and a quadratic increase (p<0.05) in leaf acid detergent fiber (ADF) and nitrogen free extract (NFE) contents, and stalk and whole plant DM (p<0.001), organic matter (OM, p<0.01 and p<0.05), NDF (p<0.001 and 0.05) and NFE (p<0.05) contents. However, there were decreases in leaf crude protein content (CP, quadratic, p<0.001) and stalk and whole plant EE content (linear, p<0.001), CP (quadratic, p<0.05) and ash (quadratic, p<0.05) contents. Digestibility of DM and TDN, and DE and ME value in leaves were not affected by increasing shoot height, but these measures linearly decreased (p<0.001) in stalk and whole plant. In Experiment 2, the hay had higher DM and CP concentrations, but lower EE concentration than the silage. Essential oil (EO) content in wormwood silage (0.49 g/100 g DM) was higher (p<0.05) than that in wormwood hay (0.32 g/100 g DM). Wormwood hay contained 25 essentail oils (EO) including camphor (10.4 g/100 g), 1-borneol (11.6 g/100 g) and caryophyllene oxide (27.7 g/100 g), and wormwood silage had 26 EO constituents including 3-cyclohexen-1-ol (8.1 g/100 g), trans-caryophyllene (8.6 g/100 g) and ${\gamma}$-selinene (16.8 g/100 g). It is concluded that the most ideal shoot height for harvesting wormwood is 60 cm based on the optimization of DM yield and nutritive value. Wormwood silage had a greater quantity and array of EO than wormwood hay.

• KSBB Journal
• /
• v.19 no.6 s.89
• /
• pp.471-477
• /
• 2004

Oldenlandia diffusa is a Chinese medicinal herb with antitumor activity capable of suppressing the growth of some cancer cell lines. Oleanolic acid and ursolic acid are triterpenoid compounds that exist in Oldenlandia diffusa. Recently, these have been noted for anti-inflammatory, anti-cancer, and hepato-protective effects. Application of both plant growth regulators, 2,4-D and kinetin, was found to be essential for the initiation of callus and suspension cells. Leaf blades of Oldenlandia diffusa was transformed into callus on Schenk and Hildebrandt medium supplemented with 0.5 mg/L 2,4-D and 0.1 mg/L kinetin, while optimum initiation condition for suspension cells of Oldenlandia diffusa was determined to be 0.75 mg/L 2,4-D and 0.1 mg/L kinetin. Chromatographic separation of oleanolic acid from its derivatives was achieved using Rexchrom S5-100-ODS column. Analytical conditions for oleanolic acid were determined as follows: flow rate at 1.0 mL/min, UV length at 200 nm and mobile phase of $80\%$ acetonitrile and $20\%$ water. Production of secondary metabolites was found to be increased by the treatment with elicitors or signal transducers. The maximum production of oleanolic acid was 99.6 mg/L in cultures with 0.5 mM salicylic acid. It is 1.74 times higher than that of control.

• Progress in Medical Physics
• /
• v.19 no.4
• /
• pp.247-255
• /
• 2008

A coded aperture camera has been developed to improve the signal-to-noise ratio (SNR) while keeping the spatial resolution of a pinhole gamma camera. The purpose of this study was to optimize a coded aperture camera and to evaluate its possibility for thyroid imaging by Monte Carlo simulation. A clinical gamma camera, a pinhole collimator with 1.0 mm hole diameter, and a $79{\times}79$ modified uniformly redundant array (MURA) mask were designed using GATE (Geant4 Application for Tomographic Emission). The penetration ratio, spatial resolution, integral uniformity and signal-to-noise ratio (SNR) were simulated and evaluated as a function of the mask thickness. The spatial resolution of the coded aperture camera was consistent with the various mask thickness, SNR showed a maximum value at 1.2 mm mask thickness and integral uniformity was improved by increasing mask thickness. Compare to the pinhole gamma camera, the coded aperture camera showed improved SNR by a factor of 30 while keeping almost the same spatial resolution. In this simulation study, the results indicated that high spatial resolution and ultra-high SNR of the thyroid imaging are feasible using a coded aperture camera.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.22 no.11
• /
• pp.2594-2603
• /
• 1997

A novel method of adaptive beam forming is presented in this paper. The proposed technique provides for a suboptimal beam pattern that increases the Signal to Noise/Interference Ratio (SNR/SIR), thus, eventually increases the capacity of the communication channel, under an assumption that the desired signal is dominant compared to each component of interferences at the receiver, which is precoditionally achieved in Code Division Multiple Access (CDMA) mobile communications by the chip correlator. The main advantages of the new technique are:(1)The procedure requires neither reference signals nor training period, (2)The signal interchoerency does not affect the performance or complexity of the entire procedure, (3)The number of antennas does not have to be greater than that of the signals of distinct arrival angles, (4)The entire procedure is iterative such that a new suboptimal beam pattern be generated upon the arrival of each new data of which the arrival angle keeps changing due tot he mobility of the signal source, (5)The total amount of computation is tremendously reduced compared to that of most conventional beam forming techniques such that the suboptimal beam pattern be produced at vevery snapshot on a real-time basis. The total computational load for generating a new set of weitht including the update of an N-by-N(N is the number of antenna elements) autocovariance matrix is $0(3N^2 + 12N)$. It can further be reduced down to O(11N) by approximating the matrix with the instantaneous signal vector.

• Geophysics and Geophysical Exploration
• /
• v.6 no.2
• /
• pp.57-63
• /
• 2003

For the dectection of small cavity in the hard rock, we investigated the feasibility of crosswell travel-time tomography and Kirchhoff migration technique. In travel-time tomography, first arrival anomaly caused by small cavity was investigated by numerical modeling based on the knowledge of actual field information. First arrival delay was very small (<0.125 msec) and detectable receiver offset range was limited to 4m with respect to $1\%$ normalized first arrival anomaly. As a consequence, it was turned out that carefully designed survey array with both sufficient narrow spatial spacing and temporal (<0.03125 msec) sampling were required for small cavity detection. Also, crosswell Kirchhoff migration technique was investigated with both numerical and real data. Stack section obtained by numerical data shows the good cavity image. In crosswell seismic data, various unwanted seismic events such as direct wave and various mode converted waves were alto recorded. To remove these noises und to enhance the diffraction signal, combination of median and bandpass filtering was applied and prestack and stacked migration images were created. From this, we viewed the crosswell migration technique as one of the adoptable method for small cavity detection.