• Nuclear Engineering and Technology
• /
• 제56권8호
• /
• pp.3199-3209
• /
• 2024

During the decommissioning of nuclear and particle accelerator facilities, a considerable amount of large-scale radioactive waste may be generated. Accurately defining the activation level of the waste is crucial for proper disposal. However, directly measuring the internal radioactivity distribution poses challenges. This study introduced a novel technology employing machine learning to assess the internal radioactivity distribution based on external measurements. Random radioactivity distribution within a structure were established, and the photon spectrum measured by detectors from outside the structure was simulated using the FLUKA Monte-Carlo code. Through training with spectrum data corresponding to various radioactivity distributions, an evaluation model for radioactivity using simulated data was developed by above Monte-Carlo simulation. Convolutional Neural Network and Transformer methods were utilized to establish the evaluation model. The machine learning construction involves 5425 simulation datasets, and 603 datasets, which were used to obtain the evaluated results. Preprocessing was applied to the datasets, but the evaluation model using raw spectrum data showed the best evaluation results. The estimation of the intensity and shape of the radioactivity distribution inside the structure was achieved with a relative error of 10%. Additionally, the evaluation based on the constructed model takes only a few seconds to complete the process.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 1999년도 하계종합학술대회 논문집
• /
• pp.162-167
• /
• 1999

ATM technology is well developed. Small-scale access node and edge switches are introduced in the network. Large scale ATM core switches are prepared for backbone application. Currently, Internet traffic is increasing so rapidly and we need to consider effective way of accommodating the volume of traffic. In the other hand, QoS and traffic engineering concept is required in the Internet services. Here, in this paper, two technologies are explained and suggested for integration of networks for future ATM based IP network.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2003년도 하계종합학술대회 논문집 Ⅲ
• /
• pp.1553-1556
• /
• 2003

In this paper, We evaluate the learning ability of non-monotonic DBM(Deterministic Boltzmann Machine) network through numerical simulations. The simulation results show that the proposed system has higher performance than monotonic DBM network model. Non-monotonic DBM network also show an interesting result that network itself adjusts the number of hidden layer neurons. DBM network can be realized with fewer components than other neural network models. These results enhance the utilization of non-monotonic neurons in the large scale integration of neuro-chips.

• 한국지능시스템학회:학술대회논문집
• /
• 한국퍼지및지능시스템학회 2001년도 추계학술대회 학술발표 논문집
• /
• pp.275-278
• /
• 2001

In this paper, We evaluate the learning ability of non-monotonic DBM(Deterministic Boltzmann Machine) network through numerical simulations. The simulation results show that the proposed system has higher performance than monotonic DBM network model. Non-monotonic DBM network also show an interesting result that network itself adjusts the number of hidden layer neurons. DBM network can be realized with fewer components than other neural network models. These results enhance the utilization of non-monotonic neurons in the large scale integration of neuro-chips.

• 한국산업정보학회:학술대회논문집
• /
• 한국산업정보학회 2003년도 춘계학술대회
• /
• pp.52-56
• /
• 2003

In this paper, We evaluate the learning ability of non-monotonic DBM(Deterministic Boltzmann Machine) network through numerical simulations. The simulation results show that the proposed system has higher performance than monotonic DBM network model. Non-monotonic DBM network also show an interesting result that network itself adjusts the number of hidden layer neurons. DBM network can be realized with fewer components than other neural network models. These results enhance the utilization of non-monotonic neurons in the large scale integration of neuro-chips.

• Applied Microscopy
• /
• 제46권4호
• /
• pp.171-175
• /
• 2016

A distinctive neuronal network in the brain is believed to make us unique individuals. Electron microscopy is a valuable tool for examining ultrastructural characteristics of neurons, synapses, and subcellular organelles. A recent technological breakthrough in volume electron microscopy allows large-scale circuit reconstruction of the nervous system with unprecedented detail. Serial-section electron microscopy-previously the domain of specialists-became automated with the advent of innovative systems such as the focused ion beam and serial block-face scanning electron microscopes and the automated tape-collecting ultramicrotome. Further advances in microscopic design and instrumentation are also available, which allow the reconstruction of unprecedentedly large volumes of brain tissue at high speed. The recent introduction of correlative light and electron microscopy will help to identify specific neural circuits associated with behavioral characteristics and revolutionize our understanding of how the brain works.

• 한국통신학회논문지
• /
• 제35권5A호
• /
• pp.457-465
• /
• 2010

센서네트워크가 넓은 영역에서의 환경 감시 혹은 u-City에서의 정보전송 등에 이용될 경우 노드 개수는 많이 증가하게 된다. 이 때 발생하는 문제점 중 하나는 역방향 전송 지연시간이 급격하게 늘어난다는 점이다. 이 논문 에서는 대규모 센서네트워크에서 역방향 패킷의 지연시간을 최소화할 수 있는 알고리즘을 제시하였다. 지그비 방식과 비교할 때 에너지 소비는 지그비와 거의 비슷하면서도 지연시간을 90% 이상 줄일 수 있음을 확인하였다.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2008년도 하계종합학술대회
• /
• pp.255-256
• /
• 2008

Distributed localization algorithms are required for large-scale wireless sensor network applications. In this paper, we introduce an efficient algorithm, termed weighted neighbor-node distribution localization(WNDL), which emphasizes simple refinement and low system-load for low-cost and low-rate wireless sensors. We inspect WNDL algorithm through MATLAB simulation.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2016년도 추계학술발표대회
• /
• pp.506-509
• /
• 2016

We present real-time pedestrian detection that exploit accuracy of Faster R-CNN network. Faster R-CNN has shown to success at PASCAL VOC multi-object detection tasks, and their ability to operate on raw pixel input without the need to design special features is very engaging. Therefore, in this work we apply and adjust Faster R-CNN to single object detection, which is pedestrian detection. The drawback of Faster R-CNN is its failure when object size is small. Previously, small sized object problem was solved by Scale-aware Network. We incorporate Scale-aware Network to Faster R-CNN. This made our method Scale-aware Faster R-CNN (DF R-CNN) that is both fast and very accurate. We separated Faster R-CNN networks into two sub-network, that is one for large-size objects and another one for small-size objects. The resulting approach achieves a 28.3% average miss rate on the Caltech Pedestrian detection benchmark, which is competitive with the other best reported results.

• 대한교통학회지
• /
• 제22권5호
• /
• pp.61-73
• /
• 2004

복잡한 대규모의 도로망에서 방대한 정보를 분석하여 실시간으로 최적 경로를 탐색해야 하는 경로 안내 시스템에서는 탐색 효율이 필수적이다. 리를 위하여 많은 연구들이 탐색 대상이 되는 노드와 링크의 수를 줄이려고 노력해왔다. 이 논문에서는 일부 영역만이 탐색으로 함수의 최적값을 찾는 단계별 세분화 방법(Coarse Grid Method)의 원리를 도로망에 응용한ㄴ다. 처음에는 간선 도로망, 다음에는 주요 도로망, 그 다음에는 세부 도로망 등으로 그 대상을 단계적으로 세분화함으로써 동시에 수많은 노드들간의 경로를 찾는 기존 방법에서의 탐색시간을 단축한다. 이 시스템을 우리나라 전국 규모의 충분히 세분화된 실제 도로망에 적용하여 시스템의 효율성, 실용성과 실시간 운영 가능성을 경로의 탐색 시간, 경로의 적합성 등에서 입증한다.