• Journal of KIISE:Computer Systems and Theory
• /
• v.35 no.12
• /
• pp.560-567
• /
• 2008

In this paper, we discuss a simple and straightforward binarization procedure which can generate black/white comics from the video frame image. Generally, the region of human's skin is colored white or light gray, while the dark region is filled with the irregular but regular patterns like hatching in most of the black/white comics. Note that it is not enough for simple threshold method to perform this work. Our procedure is decoupled into four processes. First, we use bilateral filter to suppress noise color variation and reserve boundaries. Then, we perform mean-shift segmentation for each similar colored pixels to be clustered. Third, the clustered regions are merged and extended by our region extension algorithm considering each color of their regions. Finally, we decide which pixels are on or off using by our dynamic binarization method based on the HSV color model. Our novel black/white cartooning procedure was so successful to render comic cuts from a well-known cinema in a resonable time and manual intervention.

• The KIPS Transactions:PartC
• /
• v.14C no.1 s.111
• /
• pp.1-6
• /
• 2007

In order to operate a secure network, it is very important for the network to raise positive detection as well as lower negative detection for reducing the damage from network intrusion. By using SVM on the intrusion detection field, we expect to improve real-time detection of intrusion data. However, due to classification based on calculating values after having expressed input data in vector space by SVM, continuous data type can not be used as any input data. Therefore, we present the hybrid model between SVM and decision tree method to make up for the weak point. Accordingly, we see that intrusion detection rate, F-P error rate, F-N error rate are improved as 5.6%, 0.16%, 0.82%, respectively.

• Journal of the Korea Society of Computer and Information
• /
• v.15 no.12
• /
• pp.27-35
• /
• 2010

Extracting skin color is significant in adult image detection. However, conventional methods still have essential problems in extracting skin color. That is, colors of human skins are basically not the same because of individual skin difference or difference races. Moreover, skin regions of images may not have identical color due to makeup, different cameras used, etc. Therefore, most of the existing methods use predefined skin color models. To resolve these problems, in this paper, we propose a new adult image detection method that robustly segments skin areas with an input image-adapted skin color distribution model, and verifies if the segmented skin regions contain naked bodies by fusing several representative features through a neural network scheme. Experimental results show that our method outperforms others through various experiments. We expect that the suggested method will be useful in many applications such as face detection and objectionable image filtering.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• v.9 no.1
• /
• pp.1145-1148
• /
• 2005

In this paper, we introduce a probabilistic model based on Bayesian networks (BNs) for inferring human fatigue by integrating information from various visual cues and certain relevant contextual information. Visual parameters, typically characterizing the cognitive states of a person including parameters related to eyelid movement, gaze, head movement, and facial expression, serve as the sensory observations. But, an individual visual cue or contextual Information does not provide enough information to determine human fatigue. Therefore in this paper, a Bayesian network model was developed to fuse as many as possible contextual and visual cue information for monitoring human fatigue. At the experiment results, display the utility of the proposed BNs for predicting and modeling fatigue.

• Journal of Digital Convergence
• /
• v.17 no.7
• /
• pp.285-292
• /
• 2019

With the popularization of PC, SNS and IoT, a lot of data is generated and the amount is increasing exponentially. Artificial neural network learning is a topic that attracts attention in many fields in recent years by using huge amounts of data. Artificial neural network learning has shown tremendous potential in speech recognition and image recognition, and is widely applied to a variety of complex areas such as medical diagnosis, artificial intelligence games, and face recognition. The results of artificial neural networks are accurate enough to surpass real human beings. Despite these many advantages, privacy problems still exist in artificial neural network learning. Learning data for artificial neural network learning includes various information including personal sensitive information, so that privacy can be exposed due to malicious attackers. There is a privacy risk that occurs when an attacker interferes with learning and degrades learning or attacks a model that has completed learning. In this paper, we analyze the attack method of the recently proposed neural network model and its privacy protection method.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.23 no.12
• /
• pp.1492-1499
• /
• 2019

This paper propose how to track the position of the observer to control the viewing zone using an adaptive parallax barrier. The pose is estimated using a Constrained Local Model based on the shape model and Landmark for robust eye-distance measurement in the face pose. Camera's correlation converts distance and horizontal location to centimeter. The pixel pitch of the adaptive parallax barrier is adjusted according to the position of the observer's eyes, and the barrier is moved to adjust the viewing area. This paper propose a method for tracking the observer in the range of 60cm to 490cm, and measure the error, measurable range, and fps according to the resolution of the camera image. As a result, the observer can be measured within the absolute error range of 3.1642cm on average, and it was able to measure about 278cm at 320×240, about 488cm at 640×480, and about 493cm at 1280×960 depending on the resolution of the image.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.3 no.1
• /
• pp.59-68
• /
• 1999

Recently, research and developmental direction of communication system is concurrent adopting voice data and face image in speaking to provide more higher recognition rate then in the case of only voice data. Therefore, we present a method of lipreading in speech image sequence by using the 3-D facial shape model. The method use a feature information of the face image such as the opening-level of lip, the movement of jaw, and the projection height of lip. At first, we adjust the 3-D face model to speeching face image sequence. Then, to get a feature information we compute variance quantity from adjusted 3-D shape model of image sequence and use the variance quality of the adjusted 3-D model as recognition parameters. We use the intensity inclination values which obtaining from the variance in 3-D feature points as the separation of recognition units from the sequential image. After then, we use discrete HMM algorithm at recognition process, depending on multiple observation sequence which considers the variance of 3-D feature point fully. As a result of recognition experiment with the 8 Korean vowels and 2 Korean consonants, we have about 80% of recognition rate for the plosives and vowels. We propose that usability with visual distinguishing factor that using feature vector because as a result of recognition experiment for recognition parameter with the 10 korean vowels, obtaining high recognition rate.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.44 no.1
• /
• pp.56-63
• /
• 2007

In this paper, an eye tracking method is presented using a neural network (NN) and mean-shift algorithm that can accurately detect and track user's eyes under the cluttered background. In the proposed method, to deal with the rigid head motion, the facial region is first obtained using skin-color model and con-nected-component analysis. Thereafter the eye regions are localized using neural network (NN)-based tex-ture classifier that discriminates the facial region into eye class and non-eye class, which enables our method to accurately detect users' eyes even if they put on glasses. Once the eye region is localized, they are continuously and correctly tracking by mean-shift algorithm. To assess the validity of the proposed method, it is applied to the interface system using eye movement and is tested with a group of 25 users through playing a 'aligns games.' The results show that the system process more than 30 frames/sec on PC for the $320{\times}240$ size input image and supply a user-friendly and convenient access to a computer in real-time operation.

• Journal of Broadcast Engineering
• /
• v.23 no.3
• /
• pp.351-360
• /
• 2018

Human emotion recognition is a research topic that is receiving continuous attention in computer vision and artificial intelligence domains. This paper proposes a method for classifying human emotions through multiple neural networks based on multi-modal signals which consist of image, landmark, and audio in a wild environment. The proposed method has the following features. First, the learning performance of the image-based network is greatly improved by employing both multi-task learning and semi-supervised learning using the spatio-temporal characteristic of videos. Second, a model for converting 1-dimensional (1D) landmark information of face into two-dimensional (2D) images, is newly proposed, and a CNN-LSTM network based on the model is proposed for better emotion recognition. Third, based on an observation that audio signals are often very effective for specific emotions, we propose an audio deep learning mechanism robust to the specific emotions. Finally, so-called emotion adaptive fusion is applied to enable synergy of multiple networks. The proposed network improves emotion classification performance by appropriately integrating existing supervised learning and semi-supervised learning networks. In the fifth attempt on the given test set in the EmotiW2017 challenge, the proposed method achieved a classification accuracy of 57.12%.

• Journal of Broadcast Engineering
• /
• v.17 no.3
• /
• pp.519-528
• /
• 2012

Voice activity detection (VAD) is generally conducted by extracting features from the acoustic signal and a decision rule. The performance of such VAD algorithms driven by the input acoustic signal highly depends on the acoustic noise. When video signals are available as well, the performance of VAD can be enhanced by using the visual information which is not affected by the acoustic noise. Previous visual VAD algorithms usually use single visual feature to detect the lip activity, such as active appearance models, optical flow or intensity variation. Based on the analysis of the weakness of each feature, we propose to combine intensity change measure and the optical flow in the mouth region, which can compensate for each other's weakness. In order to minimize the computational complexity, we develop simple measures that avoid statistical estimation or modeling. Specifically, the optical flow is the averaged motion vector of some grid regions and the intensity variation is detected by simple thresholding. To extract the mouth region, we propose a simple algorithm which first detects two eyes and uses the profile of intensity to detect the center of mouth. Experiments show that the proposed combination of two simple measures show higher detection rates for the given false positive rate than the methods that use a single feature.