• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.10a
• /
• pp.562-565
• /
• 2022

Recently, the increasing number of overloaded vehicles on the road poses a risk to traffic safety, such as falling objects, road damage, and chain collisions due to the abnormal weight distribution, and can cause great damage once an accident occurs. However, this irregular weight distribution is not possible to be recognized with the current weight measurement system for vehicles on roads. To address this limitation, we propose to build an object detection-based AI model to identify overloaded vehicles that cause such social problems. In addition, we present a simple yet effective method to construct an object detection model for the large-scale vehicle images. In particular, we utilize the large-scale of vehicle image sets provided by open AI-Hub, which include the overloaded vehicles from the CCTV, black box, and hand-held camera point of view. We inspected the specific features of sizes of vehicles and types of image sources, and pre-processed these images to train a deep learning-based object detection model. Finally, we demonstrated that the detection performance of the overloaded vehicle was improved by about 23% compared to the one using raw data. From the result, we believe that public big data can be utilized more efficiently and applied to the development of an object detection-based overloaded vehicle detection model.

• Korean Journal of Remote Sensing
• /
• v.39 no.2
• /
• pp.157-168
• /
• 2023

Image segmentation and supervised classification techniques are widely used to monitor the ground surface using high-resolution remote sensing images. In order to classify various objects, a process of defining a class corresponding to each object and selecting samples belonging to each class is required. Existing methods for extracting class samples should select a sufficient number of samples having similar intensity characteristics for each class. This process depends on the user's visual identification and takes a lot of time. Representative samples of the class extracted are likely to vary depending on the user, and as a result, the classification performance is greatly affected by the class sample extraction result. In this study, we propose an image classification technique that minimizes user intervention when extracting class samples by applying the histogram back-projection technique and has consistent intensity characteristics of samples belonging to classes. The proposed classification technique using histogram back-projection showed improved classification accuracy in both the experiment using hue subchannels of the hue saturation value transformed image from Compact Advanced Satellite 500-1 imagery and the experiment using the original image compared to the technique that did not use histogram back-projection.

• Korean Journal of Remote Sensing
• /
• v.39 no.2
• /
• pp.223-232
• /
• 2023

A widely used method for monitoring land cover using high-resolution satellite images is to classify the images based on the colors of the objects of interest. In urban areas, not only major objects such as buildings and roads but also vegetation such as trees frequently appear in high-resolution satellite images. However, the colors of vegetation objects often resemble those of other objects such as buildings, roads, and shadows, making it difficult to accurately classify objects based solely on color information. In this study, we propose a method that can accurately classify not only objects with various colors such as buildings but also vegetation objects. The proposed method uses the normalized difference vegetation index (NDVI) image, which is useful for detecting vegetation objects, along with the RGB image and classifies objects into subclasses. The subclass classification results are fused, and the final classification result is generated by combining them with the image segmentation results. In experiments using Compact Advanced Satellite 500-1 imagery, the proposed method, which applies the NDVI and subclass classification together, showed an overall accuracy of 87.42%, while the overall accuracy of the subchannel classification technique without using the NDVI and the subclass classification technique alone were 73.18% and 81.79%, respectively.

• Korean Journal of Remote Sensing
• /
• v.39 no.6_1
• /
• pp.1467-1475
• /
• 2023

The public-interest direct payment program involves providing direct payments to agricultural producers and rural residents through public funds, premised on performing public functions such as environmental conservation, stable food supply, and maintaining rural communities via agricultural activities. Scientific estimation of crop cultivation areas and production levels is crucial for formulating agricultural policies linked to regulating food supply, which increasingly impacts the national economy. Conducting comprehensive on-site inspections for compliance monitoring of direct payment programs has shown very low efficiency in relation to budget and time. The expansion of areas subject to compliance monitoring and various challenges in on-site inspections necessitate streamlining current monitoring methods and devising effective strategies. As a solution, the application of Remote Sensing technology and spatial information utilization, allowing swift acquisition of necessary information for policies without overall on-site visits, is being discussed as an efficient compliance monitoring method. Therefore, this study evaluated the potential use of remote sensing for improving operational efficiency in monitoring compliance with public-interest direct payment programs. Using satellite images during farming seasons in Gimje and Hapcheon, vegetation indices and spatial variations were utilized to identify cultivated areas, presence of mixed crops, validated against on-site inspection data.

• Korean Journal of Remote Sensing
• /
• v.39 no.5_2
• /
• pp.755-770
• /
• 2023

Due to recent severe climate change, abnormal weather phenomena, and other factors, the frequency and magnitude of natural disasters are increasing. The need for disaster management using artificial satellites is growing, especially during large-scale disasters due to time and economic constraints. In this study, we have summarized the current status of next-generation medium-sized satellites and microsatellites in operation and under development, as well as trends in satellite imagery analysis techniques using a large volume of satellite imagery driven by the advancement of the space industry. Furthermore, by utilizing satellite imagery, particularly focusing on recent major disasters such as floods, landslides, droughts, and wildfires, we have confirmed how satellite imagery can be employed for damage analysis, thereby establishing its potential for disaster management. Through this study, we have presented satellite development and operational statuses, recent trends in satellite imagery analysis technology, and proposed disaster response strategies that utilize various types of satellite imagery. It was observed that during the stages of disaster progression, the utilization of satellite imagery is more prominent in the response and recovery stages than in the prevention and preparedness stages. In the future, with the availability of diverse imagery, we plan to research the fusion of cutting-edge technologies like artificial intelligence and deep learning, and their applicability for effective disaster management.

• The Sea:JOURNAL OF THE KOREAN SOCIETY OF OCEANOGRAPHY
• /
• v.11 no.1
• /
• pp.1-10
• /
• 2006

Previous studies on zooplankton in Asan Bay were mostly based on samples collected seasonally with three months intervals. Present study was aimed to know the temporal variation of meso-zooplankton distribution using the data collected monthly. Relationships between zooplankton abundances and environmental factors such as seawater temperatures, salinities and chlorophyll-a contents were also studied. Seawater temperature showed typical pattern of seasonal variation found in temperate waters. The fluctuations of environmental factors ranged relatively wider In the inner part of the bay than those in outer part of the bay. Salinity was very low right after the summer rainy period due to the sporadic outflow of freshwater from the adjacent artificial lakes. Sudden changes in salinity seemed to have significant impact on zooplankton assemblages. Chlorophyll-a contents were increased in general when compared with previous reports probably due to the recent human exploitations in the coastal zone, which might enhance the nutrients level . The timing and duration of spring bloom showed geographical differences. In the inner part of the bay it began earlier (February) and last longer (three months) while in the outer part of the bay it began late (April) and last just one month. Zooplankton abundance, especially most abundant taxon Acartia hongi, showed weak but significant positive correlation with chlorophyll-a contents. The difference in temporal variation found with two different sampling intervals indicated the necessity of shorter time interval samplings.

• The Journal of the Convergence on Culture Technology
• /
• v.10 no.1
• /
• pp.539-549
• /
• 2024

This study employed text mining techniques like frequency analysis, word clouds, and LDA topic modeling to assess consumer satisfaction and dissatisfaction with Hyundai Motor Company in the Chinese market, compared to brands such as Toyota, Volkswagen, Buick, and Geely. Focusing on compact vehicles from these brands between 2021 and 2023, this study analyzed customer reviews. The results indicated Hyundai Avante's positive factors, including a long wheelbase. However, it also highlighted dissatisfaction aspects like Manipulate, engine performance, trunk space, chassis and suspension, safety features, quantity and brand of audio speakers, music membership service, separation band, screen reflection, CarLife, and map services. Addressing these issues could significantly enhance Hyundai's competitiveness in the Chinese market. Previous studies mainly focused on literature research and surveys, which only revealed consumer perceptions limited to the variables set by the researchers. This study, through text mining and comparing various car brands, aims to gain a deeper understanding of market trends and consumer preferences, providing useful information for marketing strategies of Hyundai and other brands in the Chinese market.

• Korean Journal of Plant Resources
• /
• v.37 no.4
• /
• pp.423-430
• /
• 2024

Cymbidium (C.) spp. 'Wedding Day' was developed by the National Institute of Horticultural and Herbal Science, Rural Development Administration in 2021. This cultivar was derived from the artificial cross between C. 'Boksam Holiday' and C. 'Persicolor' in 2008. After the crossing, 112 seedlings were obtained through in vitro germination and transferred to the green house. Based on their vegetative and flowering characteristics, two lines were selected through the first selection. To confirm the stability and uniformity of the two lines, the first and second trials were conducted from 2017 to 2021. As a result, the final line with the code 'C0844-29' was selected as 'Wongyuo F1-78'. After evaluating consumer preference, the line was named as 'Wedding Day'. This hybrid is medium sized cultivar with long vase life of cut flowers and more than 11 white flowers per stalk. It has 63.2 cm of erect stalks which are suitable for cut flower. The leaves are 65.8 cm long and located lower than the flowers, providing appropriate shape for viewing the flowers. It starts flowering from late January. It has sufficient proliferation ability to enable mass proliferation for commercial use (Registration No. 9523).

• Ocean and Polar Research
• /
• v.34 no.4
• /
• pp.413-430
• /
• 2012

To understand the scales of the spatial distribution and temporal duration of the subsurface chlorophyll-a maximum (SCM) observed in the Ulleung Basin of the East Sea, we analyzed physical and chemical data collected during the East Asian Seas Time-series-I (EAST-I) program. The SCM layer occurred at several observation lines from the Korea Strait to $37.9^{\circ}N$ in the Ulleung Basin during August of 2008 and 2011. At each observation line, the SCM layer extended from the coast to about 200 km off the coast. The SCM layer was observed between 30 and 40 m depth in the Ulleung Basin as well as in the northwestern Japan Basin along $132.3^{\circ}E$ from $38^{\circ}N$ to $42.3^{\circ}N$ during July 2009, and was observed around 50 m depth in the northeastern Japan Basin ($135-140^{\circ}E$ and $40-45^{\circ}N$) during July 2010. From these observed features, we hypothesize that the SCM layer observed in the Ulleung Basin may exist in most of the East Sea and may last for at least half-year (from the early May to late October). The nutrient supply mechanism for prolonged the SCM layer in the East Sea was not known, but it may be closely related to the horizontal advection of the nutrient rich and low oxygen waters observed in the Korea Strait between a 50 m depth to near the bottom. The prolonged development of the SCM layer in the Ulleung Basin may result in high primary production and would also be responsible for the high organic carbon content observed in the surface sediment of the region.

• KOREAN JOURNAL OF CROP SCIENCE
• /
• v.36 no.2
• /
• pp.134-142
• /
• 1991

Since IR8, the first high-yielding rice cultivar characterised by semidwarf and high-tillering, was released in 1966, rice yields during the last two decades have apparently reached a plateau and subsequent efforts to further improve yielding ability have not resulted in visible gains. At this point of time, a new ideotype of rice plant would be necessary to increase grain yield potential. This experiment was conducted to investigate the yield contribution of different tillers within a plant in relation to an ideo type of rice plant. A low-tillering, large panicled IR25588 was compared with a high-tillering, small panicled IR58. Based on spikelet number and grain weight per panicle, the top six panicles in both low- and high-tillering cultivarswere significantly bigger than those in the other panicles. The top six panicles were M, PI, P2, P3, P4 and S1P2 in both cultivars. Their tillers had 100% probability of occurring. The top six tillers were characterised by earlier initiation and heading, longer growth duration, greater leaf area. and heavier culm and total dry weight per tiller. The top six panicles, based on grain weight was mainly due to higher spikelet number per panicle with little differences in 1,000 grain weight and percent fertility. They had also a greater number of high -density grains. The top six panicles were significantly bigger than the rest of the panicles in both low- and high-tilliering types suggesting that a new rice ideotype having six or fewer potential tillers or panicles per paint with 200 to 250 spikelets per panicle (a low-tillering, paicle weight type) may help increase grain yield potential since they have been shown to be superior physio-morphologically to the rest of the tillers.