• The KIPS Transactions:PartB
• /
• v.10B no.1
• /
• pp.73-80
• /
• 2003

In this paper we propose an algorithm which matches the corresponding feature points between the reference image and the search image. We use Harris's corner detector to find the feature points in both image. For each feature point in the reference image, we can extract the candidate matching points as feature points in the starch image which the normalized correlation coefficient goes greater than a threshold. Finally we determine a corresponding feature points among candidate points by using dynamic programming. In experiments we show results that match feature points in synthetic image and real image.

• Journal of Microbiology
• /
• v.38 no.4
• /
• pp.250-254
• /
• 2000

This study confirmed that white rot fungus Irpex lacteus was able to metabolize 2,4,6-trinitrotoluene (TNT) with two different initial transformations. In one metabolic pathway of TNT a nitro group was removed from the aromatic ring of TNT. Hydride-Meisenheimer complexes of TNT (H$\^$-/-TNT), colored dark redo were confirmed as the intermediate in this transformation by comparison with the synthetic compounds. 2,4-Dinitrotoluene as a following metabolic product was detected, and nitrite produced by denitration of H$\^$-/-TNT supported this transformation. In the other TNT pathway, nitro groups in TNT were successively reduced to amino groups via hydroxylamines. Hydroxylamino-dinitrotoluenes and amino-dinitrotoluenes were identified as the intermediates. The activity of a membrane-associated aromatic nitroreductase was detected in the cell-free extract of I. lacteus. This enzyme catalyzed the nitro group reduction of TNT with NADPH as a cofactor, Enzyme activity was not observed in the presence of molecular oxygen.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2019.05a
• /
• pp.525-528
• /
• 2019

Atmospheric particle removal is a challenging task and attacks wide interests in computer vision filed. In this paper, we proposed a single image snow removal framework based on deep residual networks. According to the fact that there are various snow sizes in a snow image, the inception module which consists of different filter kernels was adopted to extract multiple resolution features of the input snow image. Except the traditional mean square error loss, the perceptual loss and total variation loss were employed to generate more clean images. Experimental results on synthetic and realistic snow images indicated that the proposed method achieves superior performance in respect of visual perception and objective evaluation.

• Journal of Korea Multimedia Society
• /
• v.25 no.11
• /
• pp.1582-1590
• /
• 2022

Due to the situation of the widespread of the coronavirus, which causes the problem of lack of face image data occluded by masks at recent time, in order to solve the related problems, this paper proposes a method to generate face images with masks using a combination of generative adversarial networks and spatial transformation networks based on CNN model. The system we proposed in this paper is based on the GAN, combined with multi-scale convolution kernels to extract features at different details of the human face images, and used Wasserstein divergence as the measure of the distance between real samples and synthetic samples in order to optimize Generator performance. Experiments show that the proposed method can effectively put masks on face images with high efficiency and fast reaction time and the synthesized human face images are pretty natural and real.

• Food Engineering Progress
• /
• v.13 no.4
• /
• pp.282-288
• /
• 2009

A dark brownish pigment from fruit body of Auricularia auricula was isolated and characterized in this report. The pigment was obtained with a yield of 0.61%(w/w) by alkaline extraction and subsequent purification steps. It showed the positive $FeCl_3$ test which was the indication of phenolic compounds. A synthetic melanin showed a similar spectrometric characteristics to the pigment extract regarding a characteristic UV absorption between 200-250 nm and infrared absorptions profiles in the finger print region including absorption peaks at 1701 and 1624 $cm^{-1}$. Its element analysis indicated that its atomic copmposition is close to that of DOPA melanin (eumelanin). With the result of its antioxidant activity in the TNBT (5-thio-2-nitrobenzoic acid) assay, we concluded that the dark brownish pigment from A. auricula is a melanin-like compound having a powerful antioxidative activity.

• The Korean Journal of Food And Nutrition
• /
• v.32 no.1
• /
• pp.41-49
• /
• 2019

Carotene, xanthophyll, carotenoid anthocyan, phycopyrine, chlorophyll, and monascus pigments are used as natural coloring agents since they are more stable to human body than synthetic coloring agents. Among them, monascus pigments are a natural red pigment produced by the Monascus purpureus. For the development of edible paint using natural pigment, Monascus purpureus strain was cultured at a temperature of $35^{\circ}C$ for 15 days on a PDYA plate and liquid medium to produce a red pigment. In addition, a large amount of the red pigment was extracted from Hongkuk Koji in parallel with water extraction and ultrasonic wave extraction. At this time, the yield of ultrasonic extract was 2~4 times higher. Thus, Monascus purpureus strains, etc. were prepared by freeze-drying powder. In conclusion, natural paints made with red pigments have enabled the development of been edible paints that can be used as eco-friendly materials with good viscosity, enhanced spread ability and coloration.

• The Journal of Korea Robotics Society
• /
• v.19 no.2
• /
• pp.196-202
• /
• 2024

The ability to understand given environments and plan a sequence of actions leading to goal state is crucial for personal service robots. With recent advancements in deep learning, numerous studies have proposed methods for state representation in planning. However, previous works lack explicit information about relationships between objects when the state observation is converted to a single visual embedding containing all state information. In this paper, we introduce graph-based state representation that incorporates both object and relationship features. To leverage these advantages in addressing the task planning problem, we propose a Graph Neural Network (GNN)-based subgoal prediction model. This model can extract rich information about object and their interconnected relationships from given state graph. Moreover, a search-based algorithm is integrated with pre-trained subgoal prediction model and state transition module to explore diverse states and find proper sequence of subgoals. The proposed method is trained with synthetic task dataset collected in simulation environment, demonstrating a higher success rate with fewer additional searches compared to baseline methods.

• Korean Journal of Optics and Photonics
• /
• v.19 no.6
• /
• pp.389-393
• /
• 2008

Enhanced reconstruction of heavy occluded objects was represented using estimation of variance in computational integral imaging. The system is analyzed to extract information of enhanced reconstruction from an elemental images set. To obtain elemental images with enhanced resolution, low focus error, and large depth of focus, synthetic aperture integral imaging (SAII) utilizing a digital camera has been adopted. The focused areas of the reconstructed image are varied with the distance of the reconstruction plane. When an occluded object is occluded heavily, an occluded object can not be reconstructed by removing the occluding object. To obtain reconstruction of the occluded object by remedying the effect of heavy occlusion, the statistical technique has been adopted.

• Korean Chemical Engineering Research
• /
• v.54 no.1
• /
• pp.89-93
• /
• 2016

In this study, we investigated the separation behavior of the anticancer agent paclitaxel and its semi-synthetic precursor 10-deacetylpaclitaxel (10-DAP) from plant cell cultures. As a result of sequential separation/purification performed by biomass extraction with solvent, liquid-liquid extraction, adsorbent treatment, hexane precipitation, and fractional precipitation, the adsorbent treatment was found to be the most effective in separating and recovering 10-DAP from paclitaxel. The optimal adsorbent type, crude extract/adsorbent ratio, and adsorbent treatment temperature were sylopute, 1:1.5 (w/w), and $20^{\circ}C$, respectively. The separation/recovery of 10-DAP from paclitaxel was 74.1% in adsorbent treatment process under optimal conditions.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.24 no.12
• /
• pp.1205-1214
• /
• 2013

Recently spaceborne SAR systems are increasing image resolution and frequency. As a high quality image resolution, the wider bandwidth is required and a wideband signal generator with RF component is very complicated and RF impairments of device is increased. Therefore, it is very important to improve performance by reducing these errors. In this study, the transmission signal of the wideband signal generator is applied to the phase noise, IQ imbalance, ripple gain, nonlinear model of high power amplifier. And we define possible structures of wideband signal generator and measure the PSLR and ISLR for the performance assesment. Also, we extract error of the amplitude and phase from the waveform and use a quadratic polynomial curve fitting and examine the performance change due to nonlinear device. Finally, we apply a high power amplifier predistortion method for non-linear error compensation. And we confirm that distortion in the output of the amplifier by intermodulation component is decreased by 15 dB.