• Journal of Multimedia Information System
• /
• 제5권3호
• /
• pp.189-194
• /
• 2018

In this paper, we propose the drowsiness detection method for car driver. This paper determines whether or not the driver's eyes are closed using the depth and infrared videos. The proposed method has the advantage to detect drowsiness without being affected by illumination. The proposed method detects a face in the depth picture by using the fact that the nose is closest to the camera. The driver's eyes are detected by using the extraction of harr-like feature within the detected face region. This method considers to be drowsiness if eyes are closed for a certain period of time. Simulation results show the drowsiness detection performance for the proposed method.

• Journal of Multimedia Information System
• /
• 제4권1호
• /
• pp.33-38
• /
• 2017

This article suggests a new algorithm for detecting two-wheelers on the road that have various shapes according to viewpoints. Because of complicated shapes, it is more difficult than detecting a human. In general, the Histograms of Oriented Gradients(HOG) feature is well known as a useful method of detecting a standing human. We propose a method of detecting a human on a two-wheelers using the spatial relocation of HOG (Histogram of Oriented Gradients) features. And this paper adapted the contrast method which is generally using in the image process to improve the detection rate. Our experimental results show that a two-wheelers detection system based on proposed approach leads to higher detection accuracy, less computation, and similar detection time than traditional features.

• 한국군사과학기술학회지
• /
• 제11권1호
• /
• pp.66-74
• /
• 2008

This paper presents a new small target detection method using scale invariant feature. Detecting small targets whose sizes are varying is very important to automatic target detection. Scale invariant feature using the Laplacian scale-space can detect different sizes of targets robustly compared to the conventional spatial filtering methods with fixed kernel size. Additionally, scale-reflected adaptive thresholding can reduce many false alarms. Experimental results with real IR images show the robustness of the proposed target detection in real world.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2003년도 하계종합학술대회 논문집 Ⅳ
• /
• pp.2112-2115
• /
• 2003

The system for the real time face detection is described in this paper. For face verification, support vector machine (SVM) was utilized. Although SVM performs quit well, SVM has a drawback that the computational cost is high because all pixels in a mask are used as an input feature vector of SVM. To resolve this drawback, a method to reduce the dimension of feature vectors using the integer DCT was proposed. Also for the real time face detection applications, low-complexity methods for face candidate detection in a gray image were used. As a result, the accurate face detection was performed in real time.

• Computers and Concrete
• /
• 제26권5호
• /
• pp.411-420
• /
• 2020

Deep Convolution neural network (DCNN) has been widely used in the healthy maintenance of civil infrastructure. Using DCNN to improve crack detection performance has attracted many researchers' attention. In this paper, a light-weight spatial attention network module is proposed to strengthen the representation capability of ResNet and improve the crack detection performance. It utilizes attention mechanism to strengthen the interested objects in global receptive field of ResNet convolution layers. Global average spatial information over all channels are used to construct an attention scalar. The scalar is combined with adaptive weighted sigmoid function to activate the output of each channel's feature maps. Salient objects in feature maps are refined by the attention scalar. The proposed spatial attention module is stacked in ResNet50 to detect crack. Experiments results show that the proposed module can got significant performance improvement in crack detection.

• 한국음향학회:학술대회논문집
• /
• 한국음향학회 2002년도 하계학술발표대회 논문집 제21권 1호
• /
• pp.25-28
• /
• 2002

Voice activity detection is important Problem in the speech recognition and communication. This paper introduces feature parameter which is reconstructed by the spectral entropy of information theory for the robust voice activity detection in the noise environment, analyzes and compares it with the energy method of voice activity detection and performance. In experiment, we confirmed that the spectral entropy is more feature parameter than the energy method for the robust voice activity detection in the various noise environment.

• International Journal of Contents
• /
• 제9권3호
• /
• pp.1-9
• /
• 2013

In this paper, we propose a fast and accurate system for detecting pedestrians from a static image. Histogram of Oriented Gradients (HOG) is a well-known feature for pedestrian detection systems but extracting HOG is expensive due to its high dimensional vector. It will cause long processing time and large memory consumption in case of making a pedestrian detection system on high resolution image or video. In order to deal with this problem, we use Principal Components Analysis (PCA) technique to reduce the dimensionality of HOG. The output of PCA will be input for a linear SVM classifier for learning and testing. The experiment results showed that our proposed method reduces processing time but still maintains the similar detection rate. We got twenty five times faster than original HOG feature.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2014년도 추계학술발표대회
• /
• pp.520-522
• /
• 2014

Due to the availability of easy-to-use and powerful image editing tools, the authentication of digital images cannot be taken for granted and it gives rise to non-intrusive forgery detection problem because all imaging devices do not embed watermark. Forgery detection plays an important role in this case. In this paper, an effective framework for passive-blind method for copy-move image forgery detection is proposed, based on Gabor filter which is robust to illumination, rotation invariant, robust to scale. For the detection, the suspicious image is selected and Gabor wavelet is applied from whole scale space and whole direction space. We will extract the mean and the standard deviation as the texture features and feature vectors. Finally, a distance is calculated between two textures feature vectors to determine the forgery, and the decision will be made based on that result.

• 한국정보시스템학회지:정보시스템연구
• /
• 제31권4호
• /
• pp.01-23
• /
• 2022

Purpose This study aims to propose a deep learning-based fake review detection model by combining authors' linguistic features and semantic information of reviews. Design/methodology/approach This study used 358,071 review data of Yelp to develop fake review detection model. We employed linguistic inquiry and word count (LIWC) to extract 24 linguistic features of authors. Then we used deep learning architectures such as multilayer perceptron(MLP), long short-term memory(LSTM) and transformer to learn linguistic features and semantic features for fake review detection. Findings The results of our study show that detection models using both linguistic and semantic features outperformed other models using single type of features. In addition, this study confirmed that differences in linguistic features between fake reviewer and authentic reviewer are significant. That is, we found that linguistic features complement semantic information of reviews and further enhance predictive power of fake detection model.

• 방송공학회논문지
• /
• 제23권1호
• /
• pp.86-92
• /
• 2018

본 논문에서는 디지털 음성 신호의 부호화 이력을 검출하는 방법을 제안한다. 음성 신호를 디지털 방식으로 전송 또는 저장할 때 데이터양을 줄이기 위해 부호화한다. 따라서 음성 신호 파형이 주어질 때, 해당 신호가 원본인지 부호화된 신호인지 판단하고, 만일 부호화 되었다면 부호화 횟수를 검출하는 부호화 이력 검출 과정이 필요하다. 본 논문에서는 12.2kbps 비트율의 AMR 부호화기에 대하여 원본, 단일 부호화, 이중 부호화 여부를 판단하는 부호화 이력 검출 방법을 제안한다. 제안한 방법은 입력 음성 신호에서 음성 고유의 특성 벡터를 추출하고, 해당 특성 벡터를 심층 신경망으로 모델링 하는 방법을 사용한다. 본 논문에서 제안하는 특성 벡터가 일반적인 스펙트로그램으로부터 추출한 특성 벡터보다 우수한 부호화 이력 검출 성능을 제공하는 것을 확인하였다.