• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제14권2호
• /
• pp.771-788
• /
• 2020

Detection of abandoned objects for smart video surveillance should be robust and accurate in various situations with low computational costs. This paper presents a new algorithm for abandoned object detection based on the dual background model. Through the template registration of a candidate stationary object and presence authentication methods presented in this paper, we can handle some complex cases such as occlusions, illumination changes, long-term abandonment, and owner's re-attendance as well as general detection of abandoned objects. The proposed algorithm also analyzes video frames at specific intervals rather than consecutive video frames to reduce the computational overhead. For performance evaluation, we experimented with the algorithm using the well-known PETS2006, ABODA datasets, and our video dataset in a live streaming environment, which shows that the proposed algorithm works well in various situations.

• 한국정보통신학회논문지
• /
• 제24권10호
• /
• pp.1262-1270
• /
• 2020

최근 관심이 높아지고 있는 스마트 감시 비디오에 관한 연구는 주로 침입자 탐지 및 추적과 유기 물체 탐지에 초점이 맞춰져 왔고, 도난 물체의 실시간 탐지에 대한 연구는 중요성에 비해 상대적으로 미흡한 상황이다. 본 논문은 스마트 감시 비디오 적용 환경을 고려하여 두 가지의 서로 다른 도난 물체 탐지 알고리즘을 제시한다. 먼저 이중 배경 차감 모델(dual background subtraction model)을 사용하여 사전에 정적 및 동적으로 등록된 감시 대상 물체의 도난을 탐지하는 알고리즘을 제시한다. 그리고 이중 배경 차감 모델과 Mask R-CNN 기반의 객체 세그멘테이션 기술을 통합적으로 적용하여 일반 감시 물체의 도난을 탐지하는 알고리즘을 제시한다. 전자의 알고리즘은 등록된 감시 물체를 대상으로 계산 능력이 높지 않은 환경에서 경제적인 도난 탐지 서비스를 제공할 수 있고, 후자의 알고리즘은 충분한 계산 능력을 제공할 수 있는 환경에서 보다 광범위한 일반 감시 물체의 도난 탐지에 적용할 수 있다.

• The Korean Journal of Pain
• /
• 제35권2호
• /
• pp.173-182
• /
• 2022

Background: Neurokinin-1 (NK1) and calcitonin gene-related peptide (CGRP) play a vital role in pain pathogenesis, and these proteins' antagonists have attracted attention as promising pharmaceutical candidates. The authors investigated the anti-nociceptive effect of co-administration of the CGRP antagonist and an NK1 antagonist on pain models compared to conventional single regimens. Methods: C57Bl/6J mice underwent sciatic nerve ligation for the neuropathic pain model and were injected with 4% formalin into the hind paw for the inflammatory pain model. Each model was divided into four groups: vehicle, NK1 antagonist, CGRP antagonist, and combination treatment groups. The NK1 antagonist aprepitant (BIBN4096, 1 mg/kg) or the CGRP antagonist olcegepant (MK-0869, 10 mg/kg) was injected intraperitoneally. Mechanical allodynia, thermal hypersensitivity, and anxiety-related behaviors were assessed using the von Frey, hot plate, and elevated plus-maze tests. The flinching and licking responses were also evaluated after formalin injection. Results: Co-administration of aprepitant and olcegepant more significantly alleviated pain behaviors than administration of single agents or vehicle, increasing the mechanical threshold and improving the response latency. Anxiety-related behaviors were also markedly improved after dual treatment compared with either naive mice or the neuropathic pain model in the dual treatment group. Flinching frequency and licking response after formalin injection decreased significantly in the dual treatment group. Isobolographic analysis showed a meaningful additive effect between the two compounds. Conclusions: A combination pharmacological therapy comprised of multiple neuropeptide antagonists could be a more effective therapeutic strategy for alleviating neuropathic or inflammatory pain.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제15권10호
• /
• pp.3668-3684
• /
• 2021

Video action recognition is widely used in video surveillance, behavior detection, human-computer interaction, medically assisted diagnosis and motion analysis. However, video action recognition can be disturbed by many factors, such as background, illumination and so on. Two-stream convolutional neural network uses the video spatial and temporal models to train separately, and performs fusion at the output end. The multi segment Two-Stream convolutional neural network model trains temporal and spatial information from the video to extract their feature and fuse them, then determine the category of video action. Google Xception model and the transfer learning is adopted in this paper, and the Xception model which trained on ImageNet is used as the initial weight. It greatly overcomes the problem of model underfitting caused by insufficient video behavior dataset, and it can effectively reduce the influence of various factors in the video. This way also greatly improves the accuracy and reduces the training time. What's more, to make up for the shortage of dataset, the kinetics400 dataset was used for pre-training, which greatly improved the accuracy of the model. In this applied research, through continuous efforts, the expected goal is basically achieved, and according to the study and research, the design of the original dual-flow model is improved.

• International Journal of Computer Science & Network Security
• /
• 제21권8호
• /
• pp.238-246
• /
• 2021

The rapid rise of the Internet and social media has resulted in a large number of text-based reviews being placed on sites such as social media. In the age of social media, utilizing machine learning technologies to analyze the emotional context of comments aids in the understanding of QoS for any product or service. The classification and analysis of user reviews aids in the improvement of QoS. (Quality of Services). Machine Learning algorithms have evolved into a powerful tool for analyzing user sentiment. Unlike traditional categorization models, which are based on a set of rules. In sentiment categorization, Bidirectional Long Short-Term Memory (BiLSTM) has shown significant results, and Convolution Neural Network (CNN) has shown promising results. Using convolutions and pooling layers, CNN can successfully extract local information. BiLSTM uses dual LSTM orientations to increase the amount of background knowledge available to deep learning models. The suggested hybrid model combines the benefits of these two deep learning-based algorithms. The data source for analysis and classification was user reviews of Indian Railway Services on Twitter. The suggested hybrid model uses the Keras Embedding technique as an input source. The suggested model takes in data and generates lower-dimensional characteristics that result in a categorization result. The suggested hybrid model's performance was compared using Keras and Word2Vec, and the proposed model showed a significant improvement in response with an accuracy of 95.19 percent.

• 시큐리티연구
• /
• 제41호
• /
• pp.123-151
• /
• 2014

이 연구에서는 민간경비원을 대상으로 직무착근도가 이중몰입에 어떠한 영향을 미치는지 실증적 분석을 통해 그 결과를 살펴보았다. 먼저 독립변인으로 직무착근도를 선정하였고, 직무착근도는 적합성, 연계, 희생의 세 가지 하위요소로 구분하였다. 그리고 종속변인으로 이중몰입(조직몰입, 경력몰입)을 선정하였고, 통제변인으로 성별, 연령, 학력, 근무기간, 소득을 선정하였다. 다중회귀분석결과 직무착근도의 세 가지 구성요소인 적합성, 연계, 희생은 조직몰입과 경력몰입에 각각 유의미한 영향을 미치는 것으로 나타났다. 즉 자신의 조직 및 직무에 대해 잘 맞는다고 지각하고, 조직 내 구성원들과 연결되어 있는 정도가 높으며, 조직을 떠날 경우 잃게 되는 심리적 또는 물질적 편익에 대한 지각비용이 클수록 이중몰입이 높음을 알 수 있다. 하지만 통계적으로 유의미한 것으로 나타난 변인들을 모두 포함시켜 각 변인들을 서로 통제하도록 하였을 때, 조직몰입 모형에서는 연계의 효과는 사라졌고 적합성과 희생이 통계적으로 유의미한 효과를 나타내었다. 그리고 경력몰입 모형에서는 적합성과 연계의 효과는 사라졌고, 희생이 통계적으로 유의미한 효과를 나타내었다. 따라서 직무착근도의 구성요소 중 적합성과 희생이 가장 중요한 요소임을 알 수 있다. 이상의 결과에서 살펴보았듯이 직무착근도(적합성, 연계, 희생)는 민간경비원들의 이중몰입(조직몰입, 경력몰입)에 긍정적인 영향을 미치므로 조직차원에서 직무착근도를 증진시키기 위한 방안을 마련할 필요성이 있는데, 이를 위해 적합성과 관련하여서는 신규채용단계에서 조직에 적합한 구성원을 선발하고 채용 이후에는 질 높은 교육훈련이 이루어져야 할 것이며, 연계와 관련하여서는 조직 내 친목회, 정례회, 모임 등과 같은 제도를 활성화할 필요성 있으며, 희생과 관련하여서는 근무여건, 인센티브, 복지혜택 수준을 점진적으로 증진시킬 필요성이 있다.

• 한국실내디자인학회논문집
• /
• 제24권1호
• /
• pp.93-103
• /
• 2015

This study set out to figure out the tendencies of the staff members participating in the space design of Blade Runner and compare and analyze its set and location characteristics with its narrative based on Gilles Deleuze' Simulacrum, one of the basic theories of Post modernism, thus identifying the characteristics of postmodern space inherent in it. The findings were as follows: first, the spaces in a Late modernism tendency in Blade Runner seem to have been created by the cinematic imagination of Syd Mead and Douglas Trumbull rather than being influenced by the old Late modernism architecture. Second, the postmodern spaces of the movie were designed to depict a more realistic future by reinforcing the old ornamental elements or adding the mechanical aesthetics of Late modernism based on a prediction of future cities. Third, the characters representing Late modernism and Post modernism in the narrative of the movie embrace the tendencies of the parties objected by Model and Simulacrum in the scenes where they deny the tendencies of the spaces to which they belong, thus exhibiting a dual trend. Fourth, the dual narrative of Model and Simulacrum holds duality even in the space and architecture of the movie, which is the reason why the movie chose postmodern spaces reflecting historical contexts instead of inner spaces in the tendency of minimalism, which was in vogue when SF movies were made those days. Finally, the spaces of the movie can be categorized according to the Late modernism and Post modernism tendencies from the perspective of the 1980s and be understood to show the architecture and space of future Post modernism feasible through the layering of historicity, locality, and mechanical aesthetics from ancient Maya to a future city in Los Angeles, the background of the movie, from the perspective of 2019.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제9권9호
• /
• pp.3721-3737
• /
• 2015

3D reconstruction based on binocular data is significant to machine vision. In our method, we propose a new and high efficiency 3D reconstruction approach by using a consumer camera aiming to: 1) address the configuration problem of dual camera in the binocular reconstruction system; 2) address stereo matching can hardly be done well problem in both time computing and precision. The kernel feature is firstly proposed in calibration stage to rectify the epipolar. Then, we segment the objects in the camera into background and foreground, for which system obtains the disparity by different method: local window matching and kernel feature-based matching. Extensive experiments demonstrate our proposed algorithm represents accurate 3D model.

• 동의신경정신과학회지
• /
• 제33권2호
• /
• pp.133-141
• /
• 2022

Objectives: Brainspotting is a relatively new type of brain-body psychotherapeutic approach discovered and developed by David Grand. The objective of this study was to introduce possible clinical application of Brainspotting in Korean medicine. Methods: The background, basic tools, standard processes, and principles of Brainspotting are presented mainly in reference to "Brainspotting: The revolutionary new therapy for rapid and effective change" published by David Grand. Results: There are many similarities between Brainspotting and Korean medicine, such as Mind-Body holism, non-prejudiced attitude of therapist like Tao and wu-wei, and the importance of the eyes to the mind. They also share similarities in methods such as Iijungbyunqi and Qigong. Conclusions: Brainspotting is expected to be applied to Korean medicine in various forms. It needs to be researched more in the future.

• 한국멀티미디어학회논문지
• /
• 제16권11호
• /
• pp.1272-1280
• /
• 2013

본 논문에서는 실시간으로 위험한 소리를 인식하고 그 방향을 파악하여 이를 통해 PTZ Camera가 위험한 소리 방향으로 회전하여 해당 지역 영상을 획득하여 전송할 수 있도록 지원하는 소리 감시 시스템을 제안한다. 제안 소리 감시 시스템은 적응 혼합 가우시안 모델(AGMM)을 사용하여 일상적인 배경 소리와는 비정상적인 소리를 전경 소리로 검출하고, AGMM 모델로 미리 학습된 전경 소리들 중의 하나로 분류한다. 분류된 소리가 위험한 소리에 속하는 경우, Dual delay-line 방법에 기반을 둔 음원 방향 추정 기법을 사용하여 그 방향을 파악한다. 최종적으로 방향 정보를 사용하여 PTZ 카메라를 조절하여 그 방향 지역의 해당 영상을 획득하고 전송될 수 있도록 지원한다. 제안하는 소리 감시 시스템은 전경 위험 소리들을 안정적으로 검출하고, 79%의 정확도로 위험소리들을 분류하고, 작은 오차범위 이내 음원 방향 추정 성능을 나타냄을 실험결과를 통해 확인하였다.