• Korean Journal of Remote Sensing
• /
• v.39 no.6_2
• /
• pp.1615-1633
• /
• 2023

Sargassum horneri is one of the floating algae in the sea, which breeds in large quantities in the Yellow Sea and East China Sea and then flows into the coast of Republic of Korea, causing various problems such as destroying the environment and damaging fish farms. In order to effectively prevent damage and preserve the coastal environment, the development of Sargassum horneri detection algorithms using satellite-based remote sensing technology has been actively developed. However, incorrect detection information causes an increase in the moving distance of ships collecting Sargassum horneri and confusion in the response of related local governments or institutions,so it is very important to minimize false detections when producing Sargassum horneri spatial information. This study applied technology to automatically remove false detection results using the GOCI-II-based Sargassum horneri detection algorithm of the National Ocean Satellite Center (NOSC) of the Korea Hydrographic and Oceanography Agency (KHOA). Based on the results of analyzing the causes of major false detection results, it includes a process of removing linear and sporadic false detections and green algae that occurs in large quantities along the coast of China in spring and summer by considering them as false detections. The technology to automatically remove false detection was applied to the dates when Sargassum horneri occurred from February 24 to June 25, 2022. Visual assessment results were generated using mid-resolution satellite images, qualitative and quantitative evaluations were performed. Linear false detection results were completely removed, and most of the sporadic and green algae false detection results that affected the distribution were removed. Even after the automatic false detection removal process, it was possible to confirm the distribution area of Sargassum horneri compared to the visual assessment results, and the accuracy and precision calculated using the binary classification model averaged 97.73% and 95.4%, respectively. Recall value was very low at 29.03%, which is presumed to be due to the effect of Sargassum horneri movement due to the observation time discrepancy between GOCI-II and mid-resolution satellite images, differences in spatial resolution, location deviation by orthocorrection, and cloud masking. The results of this study's removal of false detections of Sargassum horneri can determine the spatial distribution status in near real-time, but there are limitations in accurately estimating biomass. Therefore, continuous research on upgrading the Sargassum horneri monitoring system must be conducted to use it as data for establishing future Sargassum horneri response plans.

• Korean Journal of Cognitive Science
• /
• v.11 no.2
• /
• pp.37-51
• /
• 2000

The present study compared computer users target-selection response patterns when the targets were varied in terms of their relative location and distance from the current position of the cursor. In Experiment 1, where the mouse was used as an input device, the effects of different directions and distances of simple target(small rectangle) on target-selection response were investigated. The results of Experiment 1 can be summarized as follows: (1) Overshooting was more frequent than either undershooting or correct movement and (2) this tendency was more prominent when the targets were presented in the oblique direction or in farther location from the current cursor position. (3) Although the overshooting and undershooting were more frequent in the oblique direction, the degree of deviation was larger in horizontal and vertical direction. (4) Time spent in moving the mouse rather than that spent in planning, calibrating or clicking was found to be the most critical factor in determining total response time. In Experiment 2, effects of the font size and line-height of the target on target-selection response were compared with regard to two types of input devices(keyboard vs. mouse). The results are as follows: (1) Mouse generally yielded shorter target-selection time than keyboard. but this tendency was reversed when the targets were presented in horizontal and vertical directions. (2) In general, target-selection time was the longest in the condition of font size of 10 and line-height of 100%, and the shortest in the condition of font size of 12 and line-height of 150%. (3) When keyboard was used as the input device, target-selection time was shortest in the 150% line-height condition, whereas in the mouse condition, target-selection time tended to be increased as the line-height increased. which resulted in the significant interaction effect between input device and line-height. Finally, several issues relating to human-computer interaction were discussed based on the results of the present study.

• Journal of Korean Ophthalmic Optics Society
• /
• v.20 no.3
• /
• pp.377-384
• /
• 2015

Purpose: On this study, we measured and compared the dynamic stereoacuity by two-rods test (Howard-Dolman Test) and three-rods test. And we analyzed the correlation between PD and refractive error with dynamic stereoacuity. Methods: Dynamic stereoacuity of two-rods test and three-rods test at 2.5 m distance for 93 adults 93 (50 males, 43 females), mean age of $21.27{\pm}2.32$ (19~32) years old, were measured 5 times for two tests. Results: The mean of dynamic stereoacuity measured by two-rods test and three-rods test were $29.91{\pm}23.03sec$ of arc and $23.75{\pm}21.65sec$ of arc for total subjects, respectively. The mean of male and female were $36{\pm}22.38sec$ of arc and $22.28{\pm}23.79$, respectively. Three-rods test showed better dynamic stereoacuity than two-rods test, but there was no statistically significant difference (p>0.05). For the average standard deviation of PD between 60.63 mm~66.19 mm, dynamic stereoacuity fo two-rod test and three-rod test were $31.48{\pm}24.87sec$ of arc and $31.48{\pm}24.87sec$ of arc, respectively. The results showed statistically significant difference (p<0.05), but the relationship between dynamic stereoacuity and PD was not great. Comparison between two tests on the basis of refractive error, dynamic stereoacuity by three-rods test was better than by two-rods test with no significant difference between both tests (p>0.05) and there was little correlation between refractive error and two dynamic stereoacuity. Conclusions: Three-rods test showing lower stereoacuity than two-rods tests could measure the lower minimum threshold of dynamic stereoacuity. It was found that both tests can be applied to dynamic stereoacuity test as a standard test, and PD and refractive error was found that little effect upon the dynamic stereoacuity. PD and refractive error was found that little effect upon the dynamic stereoacuity.

• Korean Journal of Ecology and Environment
• /
• v.53 no.3
• /
• pp.229-240
• /
• 2020

In this study, the results of fish assemblage survey for 16 reservoirs in the Yeongsan·Seomjin-River watershed were presented with fish assemblage characteristics analysis in relation to reservoir size. The survey method including number of sampling sites was followed the "Biomonitoring survey and assessment manual" of the Ministry of Environment/National Institute of Environmental Research (MOE/NIER), and the reservoirs were categorized as three size groups, small, medium or large reservoirs, based on the MOE/NIER as well. Total 13 family classified into 44 species were collected from 2018 (7 reservoirs) to 2019 (9 reservoirs), and the dominant and subdominant species were Hemiculter eigenmanni (Relative abundance, RA, 32.9%) and Lepomis macrochirus (RA, 31.4%), respectively. As a result of the analysis in relation to the reservoir size, the average (±standard deviation) number of species of the small, medium and large reservoirs were 11±2.9, 14.3±2.1, 22.7±0.6, respectively, which showed positive correlation with the reservoir size. Total 6 fish assemblage characteristics(number of species, number of individuals, richness index, herbivorous fish ratio, carnivorous fish ratio, exotic fish ratio) showed significant differences between the each reservoir size groups (P<0.05). As a result of cluster analysis, 16 reservoirs were clustered into 5 groups with 60% similarity, and the each reservoirs seems to be clustered depends on the distance from each other, watershed and their historical geology rather than size. These results are baseline information for the understanding of fish assemblage in Korean reservoirs, important for establishing management policy of reservoirs in the Yeongsan·Seomjin-River watershed.

• Journal of the Korean Society of Fisheries and Ocean Technology
• /
• v.39 no.4
• /
• pp.251-261
• /
• 2003

To measure the GPS position accuracy and its distribution according to the length of the baseline, 30 minutes to 24 hours observations at the fixed location were conducted with two GPS receivers (Ll, 12 channels) on May 29 to June 2, 2002. The GPS data received at the reference station, the rover station and the ordinary times GPS observation station operated by the National Geography Institute in Korea were processed in kinematic and static post-processing methods with a post -processing software. The results obtained are summarized as follows: 1. The number of the satellite that could be observed continuously more than six hours was 16 and most of these satellites were positioned at east-west direction on May 31, 2002. The number of the satellite observed and the geometric dilution of precision (GDOP) determined by the average of every 10 minute for the day were 8 and 3.89, respectively. 2. Both the average GPS positions before and after post-processing were shifted (standalone: 1.17 m, post -processing: 0.43m) to the south and west. The twice distance root mean square (2drms) measured with standalone was 6.65m. The 2drms could be reduced to 33.8% (standard deviation 0=17.2) and 5.3% (0=2.2) of standalone by the kinematic and the static post-processing methods, respectively. 3. The relationship between the length of the baseline x (km) and the 2drms y (m) obtained by the static post-processing method was y=0.00l6x+0.006 $(R^2=0.87)$. In the case of the positioning with the static post-processing method using the GPS receiver, it was found that a positioning within 20cm 2drms was possible when the length of the baseline was less than 100km and the receiving time of the GPS is more than 30 minutes.

• Restorative Dentistry and Endodontics
• /
• v.35 no.6
• /
• pp.453-460
• /
• 2010

Objectives: To evaluate the accuracy and consistency of two different apex locators at both the Apex and 0.5 marks. Materials and Methods: Twenty-six root canals was scheduled for extraction for periodontal or prosthodontic reasons. Thirteen canals were measured using Root ZX and the rest by i-ROOT. The root canal length was measured both the at 0.5 mark and the Apex mark. The file was then fixed to the toot, and the distance from the file tip to the major foramen of each canal was measured after removing the root dentin under the microscope so that the major foramen and the file tip were seen. Results: 1. When the Apex mark was used, 100% of both the Root ZX and i-ROOT groups were within 0.5 mm of the major foramen. 2. When 0.5 mark was used, 100% of the Root ZX group and 77% of the i-ROOT group were within 0.5 mm of the major foramen. 3. In terms of standard deviation and quartile value, the Apex mark was more consistent than 0.5 mark in the Root ZX group, and 0.5 mark was more consistent in the i-ROOT group, but there was no statistically significant difference when compared with t-test. 4. The root canal length difference between the Apex mark and 0.5 mark was 0.22 mm and 0.46 mm in the Root ZX and i-ROOT groups, respectively. Conclusions: In this study, the Apex mark was the more consistent mark. Therefore, it is recommended to subtract 0.5 mm, which is the average length between the apex and apical constriction, from the root canal length at the Apex mark to obtain the working length clinically.

• Journal of radiological science and technology
• /
• v.37 no.4
• /
• pp.341-348
• /
• 2014

Wedge filter could use to increase the dose distribution at the hot dose regions. We evaluated dose discrepancy at surface and build region in the infield and outfield that Metal Wedge (MW) and Enhance Dynamic Wedge (EDW) were interact with photon. In this paper, we used Gafchromic EBT3 film that had excellent spatial resolution, composed the water equivalent materials and changed the optical density without development. The set up conditions of linear accelerator were fixed 6 MV photon, 100 cm SSD, $10{\times}10cm^2$ field size and were irradiated 400 cGy at Dmax. The dose distribution and absorbed dose were evaluated when we compared the open field with $15^{\circ}$, $30^{\circ}$, $45^{\circ}$ metal wedge and enhanced dynamic wedge. A $15^{\circ}$ metal wedge could increase the surface and build up region dose than using a $15^{\circ}$ enhanced dynamic wedge. A $30^{\circ}$ metal wedge could decrease the surface and build up region dose than using a $30^{\circ}$ enhanced dynamic wedge. A $45^{\circ}$ metal wedge could decrease by large deviation the surface and build up region dose than using a $15^{\circ}$ enhanced dynamic wedge. The dose of penumbra region at outfield were increased on the thick side but were decreased on the thin side. It could be decrease the surface dose and build up region dose, if the metal wedge filters were properly used to make a good dose distribution and not closed the distance of surface.

• Journal of Environmental Impact Assessment
• /
• v.28 no.3
• /
• pp.183-200
• /
• 2019

In this study, the characteristics of seasonal $PM_{2.5}$ behavior in South Korea and other Northeast Asian regions were analyzed by using the $PM_{2.5}$ ground measurement data, weather data, WRF and CMAQ models. Analysis of seasonal $PM_{2.5}$ behavior in Northeast Asia showed that $PM_{2.5}$ concentration at 6 IMS sites in South Korea was increased by long-distance transport and atmospheric congestion, or decreased by clean air inflow due to seasonal weather characteristics. As a result of analysis by applying BFM to air quality model, the contribution from foreign countries dominantly influenced the $PM_{2.5}$ concentrations of Baengnyeongdo due to the low self-emission and geographical location. In the case of urban areas with high self-emissions such as Seoul and Ulsan, the $PM_{2.5}$ contribution from overseas was relatively low compared to other regions, but the standard deviation of the season was relatively high. This study is expected to improve the understanding of the air pollutant phenomenon by analyzing the characteristics of $PM_{2.5}$ behavior in Northeast Asia according to the seasonal weather condition change. At the same time, this study can be used to establish the air quality policy in the future, knowing that the contribution of $PM_{2.5}$ concentration to the domestic and overseas can be different depending on the regional emission characteristics.

• Journal of Intelligence and Information Systems
• /
• v.25 no.1
• /
• pp.163-177
• /
• 2019

As smartphones are getting widely used, human activity recognition (HAR) tasks for recognizing personal activities of smartphone users with multimodal data have been actively studied recently. The research area is expanding from the recognition of the simple body movement of an individual user to the recognition of low-level behavior and high-level behavior. However, HAR tasks for recognizing interaction behavior with other people, such as whether the user is accompanying or communicating with someone else, have gotten less attention so far. And previous research for recognizing interaction behavior has usually depended on audio, Bluetooth, and Wi-Fi sensors, which are vulnerable to privacy issues and require much time to collect enough data. Whereas physical sensors including accelerometer, magnetic field and gyroscope sensors are less vulnerable to privacy issues and can collect a large amount of data within a short time. In this paper, a method for detecting accompanying status based on deep learning model by only using multimodal physical sensor data, such as an accelerometer, magnetic field and gyroscope, was proposed. The accompanying status was defined as a redefinition of a part of the user interaction behavior, including whether the user is accompanying with an acquaintance at a close distance and the user is actively communicating with the acquaintance. A framework based on convolutional neural networks (CNN) and long short-term memory (LSTM) recurrent networks for classifying accompanying and conversation was proposed. First, a data preprocessing method which consists of time synchronization of multimodal data from different physical sensors, data normalization and sequence data generation was introduced. We applied the nearest interpolation to synchronize the time of collected data from different sensors. Normalization was performed for each x, y, z axis value of the sensor data, and the sequence data was generated according to the sliding window method. Then, the sequence data became the input for CNN, where feature maps representing local dependencies of the original sequence are extracted. The CNN consisted of 3 convolutional layers and did not have a pooling layer to maintain the temporal information of the sequence data. Next, LSTM recurrent networks received the feature maps, learned long-term dependencies from them and extracted features. The LSTM recurrent networks consisted of two layers, each with 128 cells. Finally, the extracted features were used for classification by softmax classifier. The loss function of the model was cross entropy function and the weights of the model were randomly initialized on a normal distribution with an average of 0 and a standard deviation of 0.1. The model was trained using adaptive moment estimation (ADAM) optimization algorithm and the mini batch size was set to 128. We applied dropout to input values of the LSTM recurrent networks to prevent overfitting. The initial learning rate was set to 0.001, and it decreased exponentially by 0.99 at the end of each epoch training. An Android smartphone application was developed and released to collect data. We collected smartphone data for a total of 18 subjects. Using the data, the model classified accompanying and conversation by 98.74% and 98.83% accuracy each. Both the F1 score and accuracy of the model were higher than the F1 score and accuracy of the majority vote classifier, support vector machine, and deep recurrent neural network. In the future research, we will focus on more rigorous multimodal sensor data synchronization methods that minimize the time stamp differences. In addition, we will further study transfer learning method that enables transfer of trained models tailored to the training data to the evaluation data that follows a different distribution. It is expected that a model capable of exhibiting robust recognition performance against changes in data that is not considered in the model learning stage will be obtained.