In this report, we introduce a new self-developed violence dataset named AICS-Violence. It contains 7576 high-resolution video clips of violent and non-violent scenarios collected by outdoor security cameras. It includes two different test sets with additional non-violent but seemingly violent actions as well. One of those test sets has data collected using a different camera angle compared to that of the training set and the remaining test set. To focus on each group of people in frames, we develop a method to automatically crop candidate boxes from detected human bounding boxes. Furthermore, two methods named 3D DenseNet Fusion OF RGB and 3D DenseNet Fusion OFnom RGB are proposed. In both methods, two 3D DenseNets are used to extract features from RGB and visualized optical flow respectively. Then, two different ways of fusion namely addition and normalization-based multiplication are applied to the two methods respectively. Evaluation results show that our methods achieve slightly better performance in the first test set and especially significantly better generation than those of some other state-of-the-art methods on the second test set.This study focuses on developing a comprehensive model to calculate the techno-economic and CO2 emissions of the ethanol plant from cassava in Vietnam. Four steps were done in this study: (1) collecting plant data of the ethanol production, (2) modeling ethanol production plant to calculate the mass and energy balances, (3) calculating fossil fuel consumption and CO2 emissions during cultivation, transportation, production and use of ethanol, (4) calculating economic feasibility of ethanol production. The total capital investment is 0.6 $/liter of ethanol/year, total production cost is 0.4 $/liter of ethanol, and annual net profit is 12 million USD. Fossil energy consumption during cultivation, transportation, production and use of ethanol is 18.36 MJ/liter of ethanol, which contributes 1252 gCO2eq/l of ethanol. It was confirmed that ethanol can be the alternative fuel based on the benefits to the economy and environment by reducing CO2 emissions in Vietnam.This research proposes a novel simple anti-bias modulation technique for single phase Dual-Active-Bridge DC/DC converters. When a pulsating load occurs, the control efforts of the controller lead to sudden changes in the phase shift. In turn, this cause a momentary unbalance in the positive and negative volt-second product of the transformer. As a consequence, DC bias in the transformer current appears and takes some time to vanish. By some small changes in the setting of the pulse-width-modulation modules of the digital-signal-processor, the bias can be completely eliminated. Theoretical analysis and experimental results show that, it takes only a half of one switching period to remove the bias caused by a pulsating load change without the need of a high sampling speed or any complicated measurements or feedback.117Cuong VO-LE 118Truong Xuan DO 119Duy-Dinh NGUYEN Violence Detection using Feature Fusion of Optical Flow and 3D CNN on AICS-Violence Dataset(Project 2021)Techno-economic and greenhouse gas emission analysis of ethanol production from Cassava in Vietnam(Project 2021)A flux balancing technique for DC/DC converters in quick charging station applications(Project 2021)80旭硝子財団 助成研究成果報告（2022）
元のページ ../index.html#84