http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Longbin Qi,Yunxia Hu,Qingzhi Chai,Qun Wang 한국공업화학회 2019 Journal of Industrial and Engineering Chemistry Vol.72 No.-
The anti-biofouling performance of silver nanoparticles (Ag NPs) modified polyethersulfone (PES)membrane was evaluated during the concentration of fermentation broth. The Ag NPs containingmembrane did not exhibit biofouling mitigation performance during thefirstfiltration cycle, but couldhelp to recover waterflux upon physical cleaning. After threefiltration-clean cycles, the Ag NPscontaining membranes presented higher waterflux and slowerflux decline than the control membraneswithout Ag NPs. Ag NPs on the membrane surface facilitated the effective removal of cake layer. Moreover, the Ag NPs-containing membrane had no negative effects on the activities of bacteria infermentation broth.
장재영(JaeYoung Chang),김룡빈(Longbin Jin),김은이(Eunyi Kim) 한국HCI학회 2022 한국HCI학회 학술대회 Vol.2022 No.2
손상된 이미지를 원래 이미지와 같은 상태로 복원하는 것은 매우 어려운 작업이다. 기존의 모델들은 인코더-디코더 (encoder-decoder)나 U-Net과 같은 형태로 영상을 복원하였다. 하지만 기존 방식들은 물체의 모양을 바로잡기 위한 정보가 없어 물체의 경계가 흐릿해지는 문제점이 존재한다. 본 논문에서는 이러한 문제점을 개선시키기 위해 기존의 인코더-디코더 모델에 추가로 시멘틱 세그멘테이션(semantic segmentation) 정보를 이용하여 이미지 복원 결과를 개선하는 방법을 제안하였다. 본 논문에서 제안한 방법을 통해 기존의 인코더-디코더 모델보다 개선된 결과를 얻을 수 있었으며, 이를 통해 시멘틱 세그멘테이션 정보가 이미지 복원에 도움을 준다는 것을 확인할 수 있다.
A Tuberculosis Detection Method Using Attention and Sparse R-CNN
Xuebin Xu,Jiada Zhang,Xiaorui Cheng,Longbin Lu,Yuqing Zhao,Zongyu Xu,Zhuangzhuang Gu 한국인터넷정보학회 2022 KSII Transactions on Internet and Information Syst Vol.16 No.7
To achieve accurate detection of tuberculosis (TB) areas in chest radiographs, we design a chest X-ray TB area detection algorithm. The algorithm consists of two stages: the chest X-ray TB classification network (CXTCNet) and the chest X-ray TB area detection network (CXTDNet). CXTCNet is used to judge the presence or absence of TB areas in chest X-ray images, thereby excluding the influence of other lung diseases on the detection of TB areas. It can reduce false positives in the detection network and improve the accuracy of detection results. In CXTCNet, we propose a channel attention mechanism (CAM) module and combine it with DenseNet. This module enables the network to learn more spatial and channel features information about chest X-ray images, thereby improving network performance. CXTDNet is a design based on a sparse object detection algorithm (Sparse R-CNN). A group of fixed learnable proposal boxes and learnable proposal features are using for classification and location. The predictions of the algorithm are output directly without non-maximal suppression post-processing. Furthermore, we use CLAHE to reduce image noise and improve image quality for data preprocessing. Experiments on dataset TBX11K show that the accuracy of the proposed CXTCNet is up to 99.10%, which is better than most current TB classification algorithms. Finally, our proposed chest X-ray TB detection algorithm could achieve AP of 45.35% and AP50 of 74.20%. We also establish a chest X-ray TB dataset with 304 sheets. And experiments on this dataset showed that the accuracy of the diagnosis was comparable to that of radiologists. We hope that our proposed algorithm and established dataset will advance the field of TB detection.
알츠하이머 진단 및 중증도 예측을 위한 단일화된 딥러닝 아키텍처 연구
전효진(Hyo Jin Jon),정현택(Hyuntaek Jung),김룡빈(Longbin Jin),김은이(Eun Yi Kim) 대한전자공학회 2023 대한전자공학회 학술대회 Vol.2023 No.6
Alzheimers disease is a neurodegenerative disorder characterized by a long-term and progressive decline in cognitive abilities, which can be monitored simply through the MMSE test. In this paper, we propose a Weighted MSE-CE Loss with Bernoulli Penalty to simultaneously perform MMSE score prediction and Alzheimers disease detection. Additionally, by utilizing the VGGish feature which extracts acoustic characteristics on the ADReSSo dataset, we achieve an RMSE of 4.55 and an accuracy of 80.28% for each of the two tasks. This is the highest performance among models using only acoustic features for both tasks.
김현서(Hyunseo Kim),장재영(Jaeyoung Chang),오예림(Yelim Oh),김룡빈(Longbin Jin),신정은(Jung Eun Shin),김은이(Eun Yi Kim) 대한전자공학회 2023 대한전자공학회 학술대회 Vol.2023 No.6
Hearing loss is closely related to depression, cognitive impairment, and dementia. Therefore, early diagnosis and treatment of hearing loss can prevent these conditions. However, there has been no research on classifying hearing loss based solely on speech data. In this paper, we conducted a study on classifying hearing loss using speech data from normal individuals and those with hearing loss over the age of 50. We found that the combination of Mel-spectrogram and VGGish DNN model perform 86% accuracy in 2-class classification.