• Journal of the Korea Convergence Society
• /
• v.11 no.3
• /
• pp.19-26
• /
• 2020

In this paper, we propose a method to analyze the loss region of the dictionary-based super resolution result learned for image quality improvement and to map the learning data according to the analyzed loss region. In the conventional learned dictionary-based method, a result different from the feature configuration of the input image may be generated according to the learning image, and an unintended artifact may occur. The proposed method estimate loss information of low resolution images by analyzing the reconstructed contents to reduce inconsistent feature composition and unintended artifacts in the example-based super resolution process. By mapping the training data according to the final interpolation feature map, which improves the noise and pixel imbalance of the estimated loss information using a Gaussian-based kernel, it generates super resolution with improved noise, artifacts, and staircase compared to the existing super resolution. For the evaluation, the results of the existing super resolution generation algorithms and the proposed method are compared with the high-definition image, which is 4% better in the PSNR (Peak Signal to Noise Ratio) and 3% in the SSIM (Structural SIMilarity Index).

• Proceedings of the Korean Institute of Navigation and Port Research Conference
• /
• v.1
• /
• pp.277-282
• /
• 2006

It is well known that Kalman filtering is an optimal real-time data fusion method for GPS/INS integration. However, it has some limitations in terms of stability, adaptability and observability. A Kalman filter can perform optimally only when its dynamic model is correctly defined and the noise statistics for the measurement and process are completely known. It is found that estimated Kalman filter states could be influenced by several factors, including vehicle dynamic variations, filter tuning results, and environment changes, etc., which are difficult to model. Neural networks can map input-output relationships without apriori knowledge about them; hence a proper designed neural network is capable of learning and extracting these complex relationships with enough training. This paper presents a GPS/INS integrated system that combines Kalman filtering and neural network algorithms to improve navigation solutions during GPS outages. An Extended Kalman filter estimates INS measurement errors, plus position, velocity and attitude errors etc. Kalman filter states, and gives precise navigation solutions while GPS signals are available. At the same time, a multi-layer neural network is trained to map the vehicle dynamics with corresponding Kalman filter states, at the same rate of measurement update. After the output of the neural network meets a similarity threshold, it can be used to correct INS measurements when no GPS measurements are available. Selecting suitable inputs and outputs of the neural network is critical for this hybrid method. Detailed analysis unveils that some Kalman filter states are highly correlated with vehicle dynamic variations. The filter states that heavily impact system navigation solutions are selected as the neural network outputs. The principle of this hybrid method and the neural network design are presented. Field test data are processed to evaluate the performance of the proposed method.

• Korean Journal of Microbiology
• /
• v.25 no.4
• /
• pp.282-289
• /
• 1987

The characteristics of the plasmid pKU10 isolated from Pseudomonas putida KU816 were investigated and its restriction map was constructed. The pKU10 plasmid was a small R plasmid carrying genes for resistance to ampicillin, tetracyclin, and chloramphenicol, and cured by treatment with mitomycin C. The molecular size of pKU10 was estimated to be 9.4Kb. Pseudomonas strains and E. coli cells could be transformed for antibiotic resistance characters specified by pKU10 plasmid DNA. By incompatibility test with other plasmids, pKU10 is grouped into IncP-1. EcoRI, XhoI, SalI, BglII, and SmaI cleaved pKU10 once, while PstI cleaved at two sites, and HindIII cleaved at six sites. The restriction map was constructed by partial and complete digestion of the purified plasmid DNA with single, double, or triple restriction enzymes. Thus, pKU10 is expected to be used for a cloning vector in Pseudomonas cells.

• Korean Journal of Remote Sensing
• /
• v.20 no.2
• /
• pp.139-152
• /
• 2004

This paper proposes to improve the results of image classification with spatial region growing segmentation by using an MRF-based classifier. The proposed approach is to re-classify the pixels in the boundary area, which have high probability of having classification error. The MRF-based classifier performs iteratively classification using the class parameters estimated from the region growing segmentation scheme. The proposed method has been evaluated using simulated data, and the experiment shows that it improve the classification results. But, conventional MRF-based techniques may yield incorrect results of classification for remotely-sensed images acquired over the ground area where has complicated types of land-use. A multistage MRF-based iterative class-modification in boundary is proposed to alleviate difficulty in classifying intricate land-cover. It has applied to remotely-sensed images collected on the Korean peninsula. The results show that the multistage scheme can produce a spatially smooth class-map with a more distinctive configuration of the classes and also preserve detailed features in the map.

• KSCI Review
• /
• v.14 no.1
• /
• pp.241-248
• /
• 2006

The recognition of a relatively big and rarely movable object. such as refrigerator and air conditioner, etc. is necessary because these objects can be crucial global stable features of Simultaneous Localization and Map building(SLAM) in the indoor environment. In this paper. we propose a novel method to recognize these big objects using a sequence of 3D scenes. The particles representing an object to be recognized are scattered to the environment and then the probability of each particles is calculated by the matching test with 3D lines of the environment. Based on the probability and degree of convergence of particles, we can recognize the object in the environment and the pose of object is also estimated. The experimental results show the feasibility of incremental object recognition based on particle filtering and the application to SLAM

• Journal of Convergence for Information Technology
• /
• v.9 no.7
• /
• pp.78-84
• /
• 2019

The purpose of this study is to introduce the concept of the Smart Sewer System and to analyze the economic ripple effect when smart sewer is built all over Korea. The research method is the input-output model based on the assumption that the smart sewerage will be constructed throughout the Korea from 2021 to 2040. Estimation results show that the production-induced effect reaches 343.73 trillion Korean won, the added value-induced effect is 155.867 trillion Korean won, and the employment-induced effect is estimated by 25,118,470, indicating that the smart sewer project leads to being considerably large in the nation-wide economy. In addition, the increase of social welfare by smart sewer is expected to be realized through the improvement of both the environment improvement and the national health. Therefore, the smart sewer project should be implemented without delay by planning a concrete road map and putting it into effect with a budget.

• Journal of the Architectural Institute of Korea Structure & Construction
• /
• v.34 no.2
• /
• pp.29-40
• /
• 2018

Strength design wind loads for the wind resistance design of structures shall be evaluated by the product of wind loads calculated based on the basic wind speed with 100 years return period and the wind load factor 1.3 specified in the provisions of load combinations in Korean Building Code (KBC) 2016. It may be sure that the wind load factor 1.3 in KBC(2016) had not been determined by probabilistic method or empirical method using meteorological wind speed data in Korea. In this paper, wind load factors were evaluated by probabilistic method and empirical method. The annual maximum 10 minutes mean wind speed data at 69 meteorological stations during past 40 years from 1973 to 2012 were selected for this evaluation. From the comparison of the results of those two method, it can be found that the mean values of wind load factors calculated both probability based method and empirical based method were similar at all meteorological stations. When target level of reliability index is set up 2.5, the mean value of wind load factors for all regions should be presented about 1.35. When target level of reliability index is set up 3.0, wind load factor should be presented about 1.46. By using the relationship between importance factor(conversion factor for return period) and wind load factor, the return periods for strength design were estimated and expected wind speeds of all regions accounting for strength design were proposed. It can be found that return period to estimate wind loads for strength design should be 500 years and 800 years in according to target level of reliability index 2.5 and 3.0, respectively. The 500 years basic wind speed map for strength design was suggested and it can be used with a wind load factor 1.0.

• Journal of Advanced Navigation Technology
• /
• v.27 no.2
• /
• pp.182-189
• /
• 2023

Depth estimation is a key technology in 3D map generation for autonomous driving of vehicles, robots, and drones. The existing sensor-based method has high accuracy but is expensive and has low resolution, while the camera-based method is more affordable with higher resolution. In this study, we propose self-attention-based unsupervised monocular depth estimation for UAV camera system. Self-Attention operation is applied to the network to improve the global feature extraction performance. In addition, we reduce the weight size of the self-attention operation for a low computational amount. The estimated depth and camera pose are transformed into point cloud. The point cloud is mapped into 3D map using the occupancy grid of Octree structure. The proposed network is evaluated using synthesized images and depth sequences from the Mid-Air dataset. Our network demonstrates a 7.69% reduction in error compared to prior studies.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.26 no.6D
• /
• pp.1011-1017
• /
• 2006

It is so sensitive that the wetland ecosystem very weak in artificial interference and environment change. wetlands are a transitional zone between aquatic and terrestrial ecosystems. This natural property is important to people and life. It is necessary to preservation and protection of the wetland with a countermeasure. we really need to Environment-GIS (E-GIS) and digital map which is included correct position, attribute data and range of the wetland. In this study, we take priority of making a database of wetland management. Moreover, we standardize a digital map production of wetland in our research and we improve accuracy of control survey using GPS surveying. The main purpose of this study is to suggest a pre-estimated wetland that have not yet been discovered. by analysing terrain, geological feature, a geographical distribution of plants and animals using GIS.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2008.05a
• /
• pp.194-198
• /
• 2008

Significant soil erosion and water quality degradation issues are occurring at highland agricultural areas of Kangwon province because of agronomic and topographical specialities of the region. Thus spatial and temporal modeling techniques are often utilized to analyze soil erosion and sediment behaviors at watershed scale. The Soil and Water Assessment Tool (SWAT) model is one of the watershed scale models that have been widely used for these ends in Korea. In most cases, the SWAT users tend to use the readily available input dataset, such as the Ministry of Environment (MOE) land cover data ignoring temporal and spatial changes in land cover. Spatial and temporal resolutions of the MOE land cover data are not good enough to reflect field condition for accurate assesment of soil erosion and sediment behaviors. Especially accelerated soil erosion is occurring from agricultural fields, which is sometimes not possible to identify with low-resolution MOD land cover data. Thus new land cover data is prepared with cadastral map and high spatial resolution images of the Doam-dam watershed. The SWAT model was calibrated and validated with this land cover data. The EI values were 0.79 and 0.85 for streamflow calibration and validation, respectively. The EI were 0.79 and 0.86 for sediment calibration and validation, respectively. These EI values were greater than those with MOE land cover data. With newly prepared land cover dataset for the Doam-dam watershed, the SWAT model better predicts hydrologic and sediment behaviors. The number of HRUs with new land cover data increased by 70.2% compared with that with the MOE land cover, indicating better representation of small-sized agricultural field boundaries. The SWAT estimated annual average sediment yield with the MOE land cover data was 61.8 ton/ha/year for the Doam-dam watershed, while 36.2 ton/ha/year (70.7% difference) of annual sediment yield with new land cover data. Especially the most significant difference in estimated sediment yield was 548.0% for the subwatershed #2 (165.9 ton/ha/year with the MOE land cover data and 25.6 ton/ha/year with new land cover data developed in this study). The results obtained in this study implies that the use of MOE land cover data in SWAT sediment simulation for the Doam-dam watershed could results in 70.7% differences in overall sediment estimation and incorrect identification of sediment hot spot areas (such as subwatershed #2) for effective sediment management. Therefore it is recommended that one needs to carefully validate land cover for the study watershed for accurate hydrologic and sediment simulation with the SWAT model.