• The Journal of the Korea Contents Association
• /
• v.21 no.3
• /
• pp.23-32
• /
• 2021

Semantic image segmentation technology in the field of computer vision is a technology that classifies an image by dividing it into pixels. This technique is also rapidly improving performance using a machine learning method, and a high possibility of utilizing information in units of pixels is drawing attention. However, this technology has been raised from the early days until recently for 'lack of detailed segmentation' problem. Since this problem was caused by increasing the size of the label map, it was expected that the label map could be improved by using the edge map of the original image with detailed edge information. Therefore, in this paper, we propose a post-processing algorithm that maintains semantic image segmentation based on learning, but modifies the resulting label map based on the edge map of the original image. After applying the algorithm to the existing method, when comparing similar applications before and after, approximately 1.74% pixels and 1.35% IoU (Intersection of Union) were applied, and when analyzing the results, the precise targeting fine segmentation function was improved.

• The Transactions of the Korea Information Processing Society
• /
• v.4 no.9
• /
• pp.2315-2332
• /
• 1997

In this paper, we implement a content based image retrieval system for image searching by visual features from the image databases on WWW (world wide web). The image retrieval system finds the images that contain the most similar color regions after the system automatically extracts color features from the input image. We can select one of two query methods which use a full image of $4{\times}4$ 16 sketched color region. The image similarity is calculated on the histogram intersection distance and the histogram Euclidean distance. As the experimental results show that the two different query types provide the precision/recall 0.84/0.92 and 0.85/0.93 respectively, this retrieval system has been able to obtain high performance and validity.

• Proceedings of the KSRS Conference
• /
• 2002.10a
• /
• pp.610-614
• /
• 2002

Time-mean and absolute geostrophic velocities of the Kuroshio current south of Japan are derived from TOPEX/Poseidon altimeter data using a Gaussian jet model. When compared with simultaneous measurements from a shipboard acoustic Doppler current profiler (ADCP) at two intersection points, the altimetric and ADCP absolute velocities correlate well with the correlation of 0.55 to 0.74. The time-mean velocity is accurate to 1 cm s$^{-1}$ to 5 cm s$^{-1}$. The errors in the absolute and the mean velocities are similar to those reported previously far other currents. The comparable performance suggests the Gaussian jet model is a promising methodology for determining absolute geostrophic velocities, noting that in this region the Kuroshio does not meander sufficiently, which provides unfavorable environment for the performance of the Gaussian jet model.

• Journal of Information Technology Applications and Management
• /
• v.9 no.3
• /
• pp.1-12
• /
• 2002

Due to the fast development in computer and communication technologies, a video is now being more widely used than ever in many areas. The current information analyzing systems are originally built to process text-based data. Thus, it has little bits problems when it needs to correctly represent the ambiguity of a video, when it has to process a large amount of comments, or when it lacks the objectivity that the jobs require. We would like to purpose an algorithm that is capable of analyze a large amount of video efficiently. In a video, divided areas use a region growing and region merging techniques. To sample the color, we translate the color from RGB to HSI and use the information that matches with the representative colors. To sample the shape information, we use improved moment invariants(IMI) so that we can solve many problems of histogram intersection caused by current IMI and Jain. Sampled information on characteristics of the streaming media will be used to find similar frames.

• Journal of the Korean Mathematical Society
• /
• v.48 no.5
• /
• pp.953-967
• /
• 2011

We show that the characteristic function $1S_{\underline{\alpha}}$ of any Harder-Narasimhan strata $S{\underline{\alpha}}\;{\subset}\;Coh_X^{\alpha}$ belongs to the spherical Hall algebra $H_X^{sph}$ of a smooth projective curve X (defined over a finite field $\mathbb{F}_q$). We prove a similar result in the geometric setting: the intersection cohomology complex IC(${\underline{S}_{\underline{\alpha}}$) of any Harder-Narasimhan strata ${\underline{S}}{\underline{\alpha}}\;{\subset}\;{\underline{Coh}}_X^{\underline{\alpha}}$ belongs to the category $Q_X$ of spherical Eisenstein sheaves of X. We show by a simple example how a complete description of all spherical Eisenstein sheaves would necessarily involve the Brill-Noether stratas of ${\underline{Coh}}_X^{\underline{\alpha}}$.

• Journal of Korean Society of Transportation
• /
• v.12 no.2
• /
• pp.47-63
• /
• 1994

The purposes of this study are to identify Traffic Flow characteristics and to develop approach delay model of unsignalized intersection based on the travel speed in the conflicting area. The results of this study are as following ; 1. The cumulative frequency distributions of Left-turning speed show a few differences among approaches and they are distributed to lower range of speeds. On the other hand, those of through speed show obvious differences among bounds. The similar results also show in the analysis of Percentile speed. 2. The effectiveness of conflicting movements to travel speed in the conflicting area are analyzed using regression analysis. Left-turning speed model shows that Left-and Right-Conflicting speed. Through-speed model is also developed, when approaching through volume is less than 420vph. 3. Since the lost time due to the acceleration stop, and decelerlation is occured in the conflicting area, approach delay model is delivered using the travel speed models under the condition of small queuing delay.

• Korean Institute of Interior Design Journal
• /
• v.16 no.6
• /
• pp.65-76
• /
• 2007

This research is progressed on the premise that there is a basic rule to organize a space in form generation process. In this study first, antecedent researches are inquired, and formal vocabularies related with form generation in space design are founded. Generally in the organizing process of spatial form designer could have selectively combined formal vocabularies consciously or unconsciously, and these formal vocabularies related with form generation are classified with 'formal elements' and 'spatial relationships'. First, as formal elements '2-dimensional line elements(rectilinear, oblique-linear, curvilinear)' are chosen. Second, formal vocabularies of 'spatial relationships' are 37 extracted from literatures related with architectural language or spatial language. Among them there are several vocabularies with similar or same meaning, they are unitized to representative one. Thereupon 20 vocabularies are finally selected as the formal vocabularies of 'spatial relationships', they are addition, subtraction, intersection, superimposition, juxtaposition, shifting, rotation, reflection, scaling, stretching, shear, bending, folding, distortion, articulation, deconstruction, symmetry, repetition, proportion, and asymmetry.

• Korean Journal of Remote Sensing
• /
• v.18 no.5
• /
• pp.243-253
• /
• 2002

Time-mean and absolute geostrophic velocities of the Kuroshio current south of Japan are derived from TOPEX/Poseidon altimeter data using a Gaussian jet model. When compared with simultaneous measurements from a shipboard acoustic Doppler current profiler (ADCP) at two intersection points, the altimetric and ADCP absolute velocities correlate well with the correlation coefficient of 0.55 to 0.74. The accuracy of time-mean velocity ranges from 1 cm s$^{-1}$ to 5 cm s$^{-1}$. The errors in the absolute and the mean velocities are similar to those reported previously for other currents. The comparable performance suggests the Gaussian jet model is a promising methodology for determining absolute geostrophic velocities, noting that in this region the Kuroshio does not meander sufficiently and thus provides unfavorable environment for the performance of the Gaussian jet model.

• International Journal of Highway Engineering
• /
• v.12 no.4
• /
• pp.79-86
• /
• 2010

This study deals with the roundabout installed at the diamond interchange. The goal is to analyze the relative effectiveness of roundabout to signalized intersection. In pursuing the above, this study gives the particular attentions to comparing the performances using VISSIM software. The main results analyzed are as follows. First, the single and double roundabouts are analyzed to have the rapid change of delay in the case of total entering vehicles of more than 2,400pcph for directional rate 10:5:5, 2,800pcph for 10:8:2, and 4,400pcph for 10:2:8. Second, the roundabouts are evaluated to be more effective than signalized intersection in the case of total entering vehicles of less than 4,000pcph for directional rate 10:5:5, and 4,400pcph for 10:2:8. In the case of directional rate 10:8:2, double roundabout of total entering vehicles less than 5,600pcph is analyzed to be more effective than signalized intersection. Finally, the performance of double roundabout is analyzed to be very similar to that of single roundabout in the case of total entering vehicles less than 4,400pcph. However, the double roundabout is evaluated to be more effective than single in the case of total entering vehicles more than 4,400pcph.

• Journal of Korean Society of Transportation
• /
• v.13 no.3
• /
• pp.67-86
• /
• 1995

It requires the detector system which can collect highly reliable traffic data in order to perform the real-time traffic signal control. This study is to decide the optimal shape of inductive loop for the real-time traffic signal control .This loop is located at the stopline in the signalized intersection for DS(Degree of Saturation) control. In order to find out the optimal shape of loop, 6types of experiments were performed . The results of the basic experiments of loops are as follows ; -the optimal number of turns for loop is 3 turns. -the impedance values of the loop detectors are similar to that of NEMA standards -the 1.8${\times}$4.5M loop is excellent for sensitivity in actual detection range of car length comparing to other shape of inductive loops. At the experimental of establishments of the optimal loop shape, it found that 1.8 4.5M loop has the highest values of $\DeltaL$ comparing to other types of loops, It means that the range of Lead-in cable length of this loop. And this loop is highly reliable in occpupancy time. Conclusivley, the 1.8${\times}$4.5M inductive loop is the optimal solution as a stop line loop detector for real -time traffic signal control.