http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Tung Lam Nguyen,Hieu Tran Doan Trung,이우석,이호철 한국광학회 2021 Current Optics and Photonics Vol.5 No.6
In this study, we propose a biomimetic optical structure design methodology for investigating microoptical mechanisms associated with the compound eyes of insects. With these compound eyes, insects can respond fast while maintaining a wide field of view. Also, considerable research attention has been focused on the insect compound eyes to utilize these benefits. However, their nano micro-structures are complex and challenging to demonstrate in real applications. An effectively integrated design methodology is required considering the manufacturing difficulty. We show that photorealistic ray-traced visualization is an effective method for designing the biomimetic of a micro-compound eye of an insect. We analyze the image formation mechanism and create a three-dimensional computer-aided design model. Then, a ray-trace visualization is applied to observe the optical image formation. Finally, the segmented images are stitched together to generate an image with a wide-angle; the image is assessed for quality. The high structural similarity index (SSIM) value (approximately 0.84 to 0.89) of the stitched image proves that the proposed MATLAB-based image stitching algorithm performs effectively and comparably to the commercial software. The results may be employed for the understanding, researching, and design of advanced optical systems based on biological eyes and for other industrial applications.
FGW-FER: Lightweight Facial Expression Recognition with Attention
Huy-Hoang Dinh,Hong-Quan Do,Trung-Tung Doan,Cuong Le,Ngo Xuan Bach,Tu Minh Phuong,Viet-Vu Vu 한국인터넷정보학회 2023 KSII Transactions on Internet and Information Syst Vol.17 No.9
The field of facial expression recognition (FER) has been actively researched to improve human-computer interaction. In recent years, deep learning techniques have gained popularity for addressing FER, with numerous studies proposing end-to-end frameworks that stack or widen significant convolutional neural network layers. While this has led to improved performance, it has also resulted in larger model sizes and longer inference times. To overcome this challenge, our work introduces a novel lightweight model architecture. The architecture incorporates three key factors: Depth-wise Separable Convolution, Residual Block, and Attention Modules. By doing so, we aim to strike a balance between model size, inference speed, and accuracy in FER tasks. Through extensive experimentation on popular benchmark FER datasets, our proposed method has demonstrated promising results. Notably, it stands out due to its substantial reduction in parameter count and faster inference time, while maintaining accuracy levels comparable to other lightweight models discussed in the existing literature.
Phuong T.M. Ha,Binh T.T. Le,Trung C. To,Son H. Doan,Tung T. Nguyen,Nam T.S. Phan 한국공업화학회 2017 Journal of Industrial and Engineering Chemistry Vol.54 No.-
Iron-organic framework Fe3O(BPDC)3 was synthesized, and subsequently utilized as an productive heterogeneous catalyst for the cyclization reaction of N,N-dialkylanilines with ketoxime carboxylates to produce aryl-substituted pyridines. This iron-organic framework catalyst demonstrated remarkably higher catalytic productivity for the synthesis of aryl-substituted pyridines as compared to numerous conventional homogeneous catalysts as well as MOF-based catalysts. It was possible to reuse the iron- framework catalyst in the cyclization transformation for numerous cycles without a noticeable decline in activity. To our best knowledge, this is the first heterogeneous catalytic approach towards the synthesis of aryl-substituted pyridines from ketoximes.