• International Journal of Computer Science & Network Security
• /
• v.24 no.10
• /
• pp.206-212
• /
• 2024

In the last 10 years, artificial intelligence (AI) has shown more predictive accuracy than humans in many fields. Its promising future founded on its great performance increases people's concern about its black-box mechanism. In many fields, such as medicine, mistakes lacking explanations are hardly accepted. As a result, research on interpretable AI is of great significance. Although much work about interpretable AI methods are common in classification tasks, little has focused on segmentation tasks. In this paper, we explored the interpretability on a Deep Retinal Image Understanding (DRIU) network, which is used to segment the vessels from retinal images. We combine the Grad Class Activation Mapping (Grad-CAM), commonly used in image classification, to generate saliency map, with the segmentation task network. Through the saliency map, we got information about the contribution of each layer in the network during predicting the vessels. Therefore, we adjusted the weights of last convolutional layer manually to prove the accuracy of the saliency map generated by Grad-CAM. According to the result, we found the layer 'upsample2' to be the most important during segmentation, and we improved the mIoU score (an evaluation method) to some extent.

• Journal of KIISE:Software and Applications
• /
• v.30 no.10
• /
• pp.925-937
• /
• 2003

The input of digital TV software is the Transport Stream, which utilizes the moving picture compression technique, MPEC-2 (Moving Picture Experts Groups-2). MPEG-2 TS consists of the complicated table hierarchy and internal relationships, as well as the various restrictions with regards to system standard of digital TV software in determining the field values of transport stream. However, the general MPEG-2 TS generation tool produces transport streams solely based on the MPEG-2 TS specification itself, and does not consider the interaction between modulo features or modules themselves, which construct digital TV software. In this paper, we propose a method to systematically generate MPEG-2 TS test data, namely‘Test Stream’, for digital TV software. We present the experiment result where the test stream derived from our developed tool according to the proposed method was applied to the actual digital TV software settop-box, and then analyze the result. Apart from other existing MPEG-2 TS generation tools, the advantage of our method is that not only is it capable of generating various levels of test streams including digital TV software module, integration, and system testing, but also detecting errors and analyzing their causes.

• Aerospace Engineering and Technology
• /
• v.4 no.2
• /
• pp.50-59
• /
• 2005

This paper described performance test of 1st prototype "Tail-Fan" system. KARI(Korea Aerospace Research Institute) developed "Tail-Fan" system as a kind of helicopter anti-torque system(ATS) and designed ATS performance test-rig for Tail-Fan system performance test. For Tail-Fan system performance tests, firstly, test-rig operation tests were carried out for verification of design specifications. And natural frequencies of fan blade and test-rig were measured respectively. To find the operation rotating speed for performance tests, vibration tests using accelerometers on tail gear box(TGB) were carried out. Through the fanplot and vibration test results, rotational speed for Tail-Fan performance test to avoid a resonance were found and performance tests were carried out in flight conditions. We analyzed test data by non-dimensionalization. Through this results, 1st prototype "Tail-Fan" system was satisfied with design requirements.

• Journal of Microbiology and Biotechnology
• /
• v.17 no.12
• /
• pp.1930-1936
• /
• 2007

A sequential optimization strategy based on statistical experimental designs was employed to enhance the production of cyclosporin A (CyA) by Tolypocladium inflatum DSMZ 915 in a submerged culture. A 2-level Plackett-Burman design was used to screen the bioprocess parameters significantly influencing CyA production. Among the 11 variables tested, sucrose, ammonium sulfate, and soluble starch were selected, owing to their significant positive effect on CyA production. A response surface methodology (RSM) involving a 3-level Box-Behnken design was adopted to acquire the best process conditions. Thus, a polynomial model was created to correlate the relationship between the three variables and the CyA yield, and the optimal combination of the major media constituents for cyclosporin A production, evaluated using the nonlinear optimization algorithm of EXCEL-Solver, was as follows (g/l): sucrose, 20; starch, 20; and ammonium sulfate, 10. The predicted optimum CyA yield was 113 mg/l, which was 2-fold the amount obtained with the basal medium. Experimental verification of the predicted model resulted in a CyA yield of 110 mg/l, representing 97% of the theoretically calculated yield.

• International Journal of Industrial Entomology and Biomaterials
• /
• v.12 no.2
• /
• pp.57-61
• /
• 2006

Suppressor of cytokine signaling (SOCS) is known to be as a negative feedback regulator in Janus kinase signal transducer and activator of transcription signaling. Highly conserved SOCS box domain was cloned from a Korean malaria vector, Anopheles sinensis. Sequence analysis indicates that it has identity to Anopheles gambiae (96%), Aedes aegypti (94%), Drosophila melanogaster (78%), Mus musculus (72%) and Homo sapiens (72%), respectively. Tissue specificity RT-PCR demonstrated that the expression level of AsSOCS transcript was high at abdomen, midgut, and ovary, whereas developmental expression patterns showed that the level of AsSOCS was high at egg, early pupae, and adult female. On the other hand, RT-PCR analysis after bacterial challenge showed that SOCS mRNA was strongly induced in larvae. In addition, it was also induced by various immune elicitors such as lipoteicoic acid, CpG-DNA, and laminarin. It seems that AsSOCS, repressor of JAK-STAT pathway, is highly conserved in mosquito, and may play an important role in mosquito innate immune response.

• Journal of Korean Society for Atmospheric Environment
• /
• v.22 no.E2
• /
• pp.49-56
• /
• 2006

This pilot study focuses on emissions characterization of air pollutants produced from incineration of some municipal solid wastes (MSWs). The MSWs incinerated by an electric furnace maintained up to $600^{\circ}C$ included food, paper, and plastic wastes. The pollutants analyzed in this study included concentrations of volatile organic compounds (VOCs), bottom ash contents, and heavy metals extracted from the bottom ash of each waste. The VOCs identified were classified based on their chemical structure. The total emissions of VOCs produced from incineration of the papers were identified as the highest followed by those from the plastics and the food wastes. Aliphatic alkenes were major VOC compounds produced from incineration of plastic or food wastes, while furans were major VOCs produced from incineration of papers. The second major VOCs produced from incineration of food, plastics, and papers were aromatics. In particular, hazardous air pollutants such as benzene were produced with considerable amount of emission concentration. The bottom ash contents of papers were usually much higher than those of food or plastic wastes. The bottom ash contents produced from incineration of food and plastics were much lower than those of other MSWs. In analysis of heavy metals extracted by an ultrasonic method from the bottom ashes of the papers, high concentrations of heavy metals were identified from incineration of newspapers and box (cardboard). In addition, it was identified that the general public might be exposed to considerable amounts of lead concentrations during incineration processes and uses of paper cup and from ashes.

• Journal of Microbiology
• /
• v.44 no.6
• /
• pp.641-648
• /
• 2006

We previously reported that Skp1, a component of the Skp1-Cullin-F-box protein (SCF) complex essential for the timely degradation of cell cycle proteins by ubiquitination, physically interacts with Bfa1, which is a key negative regulator of the mitotic exit network (MEN) in response to diverse checkpoint-activating stresses in budding yeast. In this study, we initially investigated whether the interaction of Skp1 and Bfa1 is involved in the regulation of the Bfa1 protein level during the cell cycle, especially by mediating its degradation. However, the profile of the Bfa1 protein did not change during the cell cycle in skp1-11, which is a SKP1 mutant allele in which the function of Skp1 as a part of SCF is completely impaired, thus indicating that Skp1 does not affect the degradation of Bfa1. On the other hand, we found that the skp1-12 mutant allele, previously reported to block G2-M transition, showed defects in mitotic exit and cytokinesis. The skp1-12 mutant allele also revealed a specific genetic interaction with ${\Delta}bfa1$. Bfa1 interacted with Skp1 via its 184 C-terminal residues (Bfa1-D8) that are responsible for its function in mitotic exit. In addition, the interaction between Bfa1 and the Skp1-12 mutant protein was stronger than that of Bfa1 and the wild type Skp1. We suggest a novel function of Skp1 in mitotic exit and cytokinesis, independent of its function as a part of the SCF complex. The interaction of Skp1 and Bfa1 may contribute to the function of Skp1 in the mitotic exit.

• Journal of the korean society of oceanography
• /
• v.33 no.3
• /
• pp.113-122
• /
• 1998

The requisite numbers of sample replicates for the population study of soft-bottom benthos were estimated from survey data on the Songdo tidal flat and subtidal zone in Youngil Bay, Korea. Large numbers of samples were taken; two-hundred-fifty 0.02 m$^2$ box corers and fifty 0.1m$^2$ van Veen grabs were taken on the Songdo tidal flat and in Youngil Bay, respectively. The effect of sampler size on sampling efforts was investigated by pooling the unit samples in pairs, fours, eights, etc. The requisite number of sample replicates (n$_r$) was determined by sample variance (s$^2$) and mean (m) function (n$_r$:s$^2$/P$^2$m$^2$), at P=0.2 level, in which s$^2$ and m were calculated from the counts of individuals collected. For example, seven samples of 0.02 m$^2$ corer for the intertidal and two samples of 0.1 m$^2$ van Veen grab for subtidal fauna were required to estimate the total density of community. The smaller sampler size was more efficient than larger ones when sampling costs were compared on the basis of the total sampling area. The requisite number of sample replicates was also predicted ($\^{n}$n$_r$) by substituting $\^{s}$$^2$ obtained from the regression of s$^2$ against m using the Taylor's power law ($\^{s}$$^2$:am$^b$). The regression line of survey data on s$^2$ and m plotted on log scale was well fitted to the Taylor's power law (r$^2$${\geq}$0.95, p<;0.001) over the whole range of m. The exponent b was, however, varied when it was estimated from m which was categorized into classes by its scale. The fitted exponent b was large when both density class and the sampler size were large. The number of sample replicates, therefore, could be more significantly estimated, if regression coefficients (a and b) would be calculated from sample variance and mean categorized into density classes.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.22 no.1
• /
• pp.73-79
• /
• 2009

In this paper, the structural analysis and optimization of the door supporting rail structure in the header box located at the top of the aluminum passenger safety door(PSD) at the subway station, which opens and closes regularly, is performed. In case the simple fixed boundary condition is used for the bolt fixture on the supporting rail where the glass door is moving, excessive stresses are obtained. Therefore, more realistic finite element modeling of the bolts is used at the bolt fixture in the whole structure in order to obtain the more physically acceptable FEM results. As a result, fatigue life of twenty years of the structure is obtained to satisfy the design object. Also the optimal design of cross section of the rail part is performed using the response surface method and 15% of weight of the supporting rail part on the door is reduced.

• Journal of Korean Society of Steel Construction
• /
• v.17 no.5 s.78
• /
• pp.511-518
• /
• 2005

The behavior of steel curved bridges is more complicated than straight bridges, thus the analysis, design and construction process of curved bridges require much more attention. In design of curved bridges, the grillage analysis using general structural analysis program or special program is mainly used. Comparative study in coherence between these analytical results and actual behavior of curved bridges has been rarely conducted. To study the behaviour of curved bridges and verify the current design method, field measurements and analyses using general structural analysis program and 3-D refined analysis program were carried out for simple and continuous bridges in this study. The study focused on the behavior of curved steel bridges during construction. Measured and analytical results had quantitative difference mutually, but there were qualitatively similar. Stress variations in transverse direction of flange were observed and grillage analysis models yielded more conservative values than 3-D refined analysis models.