• Journal of Nuclear Fuel Cycle and Waste Technology(JNFCWT)
• /
• v.17 no.1
• /
• pp.1-13
• /
• 2019

The accident at the Fukushima Dai-ichi Nuclear Power Plant (FDNPP) resulted in the deposition of large quantities of radionuclides over parts of eastern Japan. Radioactive contaminants have been observed over a large area including forests, cities, rivers and lakes. Due to the strong adsorption of radioactive cesium by soil particles, radioactive cesium migrates with the eroded soil, follows the surface flow paths, and is delivered downstream of population-rich regions and eventually to coastal areas. In this study, we developed a model to simulate the transport of contaminated sediment in a watershed hydrological system and this model was compared with observation data from eroded soil observation instruments located at the Korea Atomic Energy Research Institute. Two methods were applied to analyze the soil particle size distribution of the collected soil samples, including standardized sieve analysis and image analysis methods. Numerical models were developed to simulate the movement of soil along with actual rainfall considering initial saturation, rainfall infiltration, multilayer and rain splash. In the 2019 study, a numerical model will be used to add rainfall shield effect by trees, evaporation effect and shield effects of surface water. An eroded soil observation instrument has been installed near the Wolsong nuclear power plant since 2018 and observation data are being continuously collected. Based on these observations data, we will develop the numerical model to analyze long-term behavior of radionuclides on land as they move from land to rivers, lakes and coastal areas.

• Journal of the Institute of Convergence Signal Processing
• /
• v.23 no.3
• /
• pp.166-172
• /
• 2022

In this study, we photograph driving obstacle objects such as crosswalks, side spheres, manholes, braille blocks, partial ramps, temporary safety barriers, stairs, and inclined curb that hinder or cause inconvenience to the movement of the vulnerable using electric mobility aids. We develop an optimal AI model that classifies photographed objects and automatically recognizes them, and implement an algorithm that can efficiently determine obstacles in front of electric mobility aids. In order to enable object detection to be AI learning with high probability, the labeling form is labeled as a polygon form when building a dataset. It was developed using a Mask R-CNN model in Detectron2 framework that can detect objects labeled in the form of polygons. Image acquisition was conducted by dividing it into two groups: the general public and the transportation weak, and image information obtained in two areas of the test bed was secured. As for the parameter setting of the Mask R-CNN learning result, it was confirmed that the model learned with IMAGES_PER_BATCH: 2, BASE_LEARNING_RATE 0.001, MAX_ITERATION: 10,000 showed the highest performance at 68.532, so that the user can quickly and accurately recognize driving risks and obstacles.

• Journal of the Institute of Convergence Signal Processing
• /
• v.24 no.2
• /
• pp.119-125
• /
• 2023

In this study, we photograph driving obstacle objects such as crosswalks, side spheres, manholes, braille blocks, partial ramps, temporary safety barriers, stairs, and inclined curb that hinder or cause inconvenience to the movement of the vulnerable using electric mobility aids. We develop an optimal AI model that classifies photographed objects and automatically recognizes them, and implement an algorithm that can efficiently determine obstacles in front of electric mobility aids. In order to enable object detection to be AI learning with high probability, the labeling form is labeled as a polygon form when building a dataset. It was developed using a Mask R-CNN model in Detectron2 framework that can detect objects labeled in the form of polygons. Image acquisition was conducted by dividing it into two groups: the general public and the transportation weak, and image information obtained in two areas of the test bed was secured. As for the parameter setting of the Mask R-CNN learning result, it was confirmed that the model learned with IMAGES_PER_BATCH: 2, BASE_LEARNING_RATE 0.001, MAX_ITERATION: 10,000 showed the highest performance at 68.532, so that the user can quickly and accurately recognize driving risks and obstacles.

• Journal of Veterinary Clinics
• /
• v.24 no.4
• /
• pp.667-672
• /
• 2007

This report describes the use of conventional computed tomography(CT) for the diagnosis of head trauma in three canine patients. According to physical and neurologic examinations, survey radiography and computed tomography, these patients were diagnosed as traumatic brain injury. Especially, CT is the imaging modality of first choice for head trauma patients. It provides rapid acquisition of images, superior bone detail, and better visualization of acute hemorrhage than magnetic resonance imaging. It is also less expensive and more readily available. Pre-contrast computed tomography was used to image the head. Then, post-contrast CT was performed using the same technique. The Modified Glasgow Coma Scale(MGCS) score was used to predict their probability of survival rate after head trauma in these dogs. Computed tomogram showed fluid filled tympanic bulla, fracture of the left temporal bone and cerebral parenchymal hemorrhage with post contrast ring enhancement. However, in one case, computed tomographic examination didn't delineate cerebellar parenchymal hemorrhage, which was found at postmortem examination. Treatments for patients placed in intensive care were focused to maintain cerebral perfusion pressure and to normalize intracranial pressure. In these cases, diagnostic computed tomography was a useful procedure. It revealed accurate location of the hemorrhage lesion.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.9 no.5
• /
• pp.205-211
• /
• 2009

The Weightless Neural Network(WNN) has the advantage of the processing speed, less computability than weighted neural network which readjusts the weight. Especially, The behavior information such as sequential gesture has many serial correlation. So, It is required the high computability and processing time to recognize. To solve these problem, Many algorithms used that added preprocessing and hardware interface device to reduce the computability and speed. In this paper, we proposed the Ram based Sequential Cumulative Neural Network(SCNN) model which is sign language recognition system without preprocessing and hardware interface. We experimented with using compound words in continuous korean sign language which was input binary image with edge detection from camera. The recognition system of sign language without preprocessing got 93% recognition rate.

• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.46 no.8
• /
• pp.41-52
• /
• 2009

In Motion JPEG2000 encoding, huge bandwidth requirement of data memory access is the bottleneck in required system performance. For the alleviation of this bandwidth requirement, a new embedded compression(EC) algorithm with a little bit of image quality drop is devised. For both random accessibility and low latency, very simple and efficient entropy coding algorithm is proposed. We achieved significant memory bandwidth reductions (about 53${\sim}$81%) and reduced code-block memory to about half size through proposed multi-mode algorithms, without requiring any modification in JPEG2000 standard algorithm.

• Journal of the Korean Crystal Growth and Crystal Technology
• /
• v.30 no.1
• /
• pp.1-6
• /
• 2020

In this research, 50 nm thick AlN thin films were deposited on the patterned sapphire (0001) substrate by using HVPE (Hydride Vapor Phase Epitaxy) system and then epitaxial layer structure was grown by MOCVD (metal organic chemical vapor deposition). The surface morphology of the AlN buffer layer film was observed by SEM (scanning electron microscopy) and AFM (atomic force microscope), and then the crystal structure of GaN films of the epitaxial layer structure was investigated by HR-XRC (high resolution X-ray rocking curve). The XRD peak intensity of GaN thin film of epitaxial layer structure deposited on AlN buffer layer film and sapphire substrate was rather higher in case of that on PSS than normal sapphire substrate. In AFM surface image, the epitaxial layer structure formed on AlN buffer layer showed rather low pit density and less defect density. In the optical output power, the epitaxial layer structure formed on AlN buffer layer showed very high intensity compared to that of the epitaxial layer structure without AlN thin film.

• The Journal of the Convergence on Culture Technology
• /
• v.3 no.4
• /
• pp.165-169
• /
• 2017

CNN(Convolutional Nerual Network) is one of the algorithms that show superior performance in image recognition and classification among machine learning algorithms. CNN is simple, but it has a large amount of computation and it takes a lot of time. Consequently, in this paper we performed an parallel processing unit for the convolution layer, pooling layer and the fully connected layer, which consumes a lot of handling time in the process of CNN, through the SIMT(Single Instruction Multiple Thread)'s structure of GPGPU(General-Purpose computing on Graphics Processing Units).And we also expect to improve performance by reducing the number of memory accesses and directly using the output of convolution layer not storing it in pooling layer. In this paper, we use MNIST dataset to verify this experiment and confirm that the proposed CNN structure is 12.38% better than existing structure.

• Analytical Science and Technology
• /
• v.12 no.5
• /
• pp.415-420
• /
• 1999

We estimated the quantity of the scum remaining on the Black Matrix (BM) surface of color filter. To do this, histogram was analyzed which was obtained from AFM image of the BM surface. We divided the histogram to two Gaussian functions of the free BM surface (1) and the scum (2), and calculated the areas ($a_1$, $a_2$) of both the Gaussian functions. We quantified the residue as the ratio of the area ($a_2/(a_1+a_2)$). As a result of the Gaussian functions of the free BM surface, it was revealed that another kind of residue remained on the BM surface. It was difficult to quantify it. but it could relatively be estimated from the average height and the standard deviation.

• Proceedings of the Korean Society for Agricultural Machinery Conference
• /
• 2017.04a
• /
• pp.141-141
• /
• 2017

장미과(Rosaceae)에 속하는 딸기(Fragaria ananassa Duch.)는 비타민 C가 풍부하고 독특한 향기를 갖는 과채류로서 겨울에서 봄까지의 기간 동안 대부분 생식으로 소비되고 있다. 국내에서 재배되는 품종으로는 설향, 매향, 장희 등이 있으며 품종에 따라 성분과 함량이 다양하지만 일반적으로 유기산이 많아서 신맛과 단맛이 조화로운 특징이 있다. 소비자들이 딸기를 구입할 때 딸기가 포장된 상자에 모양이 일정하고 붉은 색상이 선명한 딸기에 호감을 갖게 된다. 딸기는 품종에 따라 기준이 되는 모양이 다르기 때문에 숙련된 선별사에 의해서 대부분 육안으로 선별되고 있는 실정이다. 하지만 개인적인 선별 능력의 차이와 주관적인 판단으로 인해 규격을 벗어난 딸기가 혼입되어 전체적인 품질 등급을 떨어뜨리는 경우가 종종 발생하기도 한다. 따라서 본 연구에서는 품종별로 기준이 되는 표준 형상과 비정상적인 모양의 기형 딸기를 객관적으로 판별하여 선별할 수 있는 영상 시스템을 구축하기 위해 수행되었으며 표준이 되는 딸기의 3차원 형상을 구축하기 위해 2차원 레이저 변위 센서를 이용하여 딸기의 입체 영상을 구축하고자 하였다. 실험을 위해 사용된 딸기는 시중에서 구입한 설향 품종이었으며 2차원 레이저 변위 센서는 라인 스캔 방식으로 1회 프로파일 스캔에 1,280개의 데이터 포인터를 획득할 수 있으며 분해능은 0.095~0.17 mm이었다. 상부에 부착된 2차원 레이저 변위 센서와 하부에 놓인 딸기의 거리는 100 mm였다. 획득한 딸기의 2차원 영상은 높이 차이를 이용하여 색상 농도로 표현하였으며 이 영상을 다시 3차원 영상으로 구축하였다.