• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2015.05a
• /
• pp.407-409
• /
• 2015

In this paper, a hardware implementation of MOD(Moving Object Detection) algorithm is described, which is based GMM(Gaussian Mixture Model) and background subtraction. The EGML(Effective Gaussian Mixture Learning) is used to model and update background. Some approximations of EGML calculations are applied to reduce hardware complexity, and pipelining technique is used to improve operating speed. Gaussian parameters are adjustable according to various environment conditions to achieve better MOD performance. MOD processor is verified by using FPGA-in-the-loop verification, and it can operate with 109 MHz clock frequency on XC5VSX95T FPGA device.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.16 no.12
• /
• pp.1286-1299
• /
• 1991

In this paper we present a multiple capture model for common spreading code CDMA packet radio systems with star topology. Basic equations for the collision free, header detection. and multiple capture probabilities are derived at the central receiver. Link performances. including the average number of packet captures, allowable number of simultaneous transmission, and system throughput are theoretically evaluated for a hybrid system. combining envelope header detection and differential data detection, Using the Block Oriented Systems Simulator(BOSS), simulations were carried out for the central receivers with envelope or differential geader detection, It is shown that for a threshold approx-imation to the probability of data packet success, the mulyiple capture model significantly improves system throughput.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.31 no.6
• /
• pp.1171-1179
• /
• 2021

Image forensic is performed to check image limpidness. In this paper, a robust scheme is discussed to detect median filtering in low quality images. Detection of median filtering assists in overall image forensic. Improved spatial statistical features are extracted from the image to classify pristine and median filtered images. Image array data is rescaled to enhance the spatial statistical information. Features are extracted using Markov model on enhanced spatial statistics. Multiple difference arrays are considered in different directions for robust feature set. Further, texture operator features are combined to increase the detection accuracy and SVM binary classifier is applied to train the classification model. Experimental results are promising for images of low quality JPEG compression.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2012.05a
• /
• pp.646-648
• /
• 2012

In this paper, an image detection technique is proposed to extract image features in a mobile terminal environment. To detect objects, the HSI color model of the image is used. The object's corner points are detected using the Harris corner detection method. Finally we detect the object of interest using region growing The experiment results show that the proposed method improves detection performance and reduces the amount of computation.

• Journal of Internet Computing and Services
• /
• v.7 no.2
• /
• pp.23-35
• /
• 2006

This work presents o novel method which extract facial region und features from motion picture automatically and controls the 3D facial expression in real time. To txtract facial region and facial feature points from each color frame of motion pictures a new nonparametric skin color model is proposed rather than using parametric skin color model. Conventionally used parametric skin color models, which presents facial distribution as gaussian-type, have lack of robustness for varying lighting conditions. Thus it needs additional work to extract exact facial region from face images. To resolve the limitation of current skin color model, we exploit the Hue-Tint chrominance components and represent the skin chrominance distribution as a linear function, which can reduce error for detecting facial region. Moreover, the minimal facial feature positions detected by the proposed skin model are adjusted by using edge information of the detected facial region along with the proportions of the face. To produce the realistic facial expression, we adopt Water's linear muscle model and apply the extended version of Water's muscles to variation of the facial features of the 3D face. The experiments show that the proposed approach efficiently detects facial feature points and naturally controls the facial expression of the 3D face model.

• Journal of the Institute of Convergence Signal Processing
• /
• v.12 no.2
• /
• pp.76-82
• /
• 2011

In this paper, we propose the technique to detect the location and size of the small target in multi-resolution image using cubic facet model. The input image is reduced by the multi-resolution and we obtain the multi-resolution images. We apply the facet model and the local maxima conditions to the multi-resolution images of each level. And then, we detect the location of the small target. We estimate that the location at the maximum of the $D_2$ which means the local maxima value of the facet model in the multi-resolution images is the location of the small target. We can detect the small target of the various size about the multi-resolution images of each level. In this paper, we experimented in the various infrared images with the small target. The method using the typical facet model applies a mask. However, the proposed method applies a mask in the multi-resolution images. We verified to vary the mask size and differ the size of the small target. The proposed algorithm can detect the location and size of the small target.

• Annual Conference on Human and Language Technology
• /
• 2022.10a
• /
• pp.517-520
• /
• 2022

지능형 대화 시스템은 줄곧 서비스의 목표와 무관한 사용자 입력을 전달받아, 그 처리 성능을 의심받는다. 특히 종단간 대화 이해 생성 모델이나, 기계학습 기반 대화 이해 모델은 학습 시간대에 한정된 범위의 도메인 입력에만 노출됨으로, 사용자 발화를 자신이 처리 가능한 도메인으로 과신하는 경향이 있다. 본 연구에서는 대화 생성 모델이 처리할 수 없는 입력과 신뢰도가 낮은 생성 결과를 배제하기 위해 불확실성 정량화 기법을 대화 의도 분류 모델에 적용한다. 여러 번의 추론 샘플링이 필요 없는 실용적인 예측 신뢰도 획득 방법과 함께, 평가 시간대와 또다른 도메인으로 구성된 분포 외 입력 데이터를 학습에 노출시키는 것이 분포 외 입력을 구분하는데 도움이 되는지를 실험으로 확인한다.

• Annual Conference of KIPS
• /
• 2022.05a
• /
• pp.320-321
• /
• 2022

부두 내 컨테이너를 적재하는 과정에서 정렬 상태가 부정확한 경우 강풍으로 인한 안전사고가 발생할 가능성이 있다. 본 논문에서는 컨테이너 안전사고를 예방하기 위한 딥러닝 기반의 컨테이너 정렬 상태 분류 알고리즘을 제안한다. 제안하는 알고리즘은 정렬을 분류하는 기준을 제시하고 YOLO 기반의 모델을 구현했다. 추론 속도, 검출 정확도, 분류 정확도를 기준으로 각 모델의 성능을 평가했으며 성능 결과는 YOLOv4모델이 YOLOv3모델에 비해서 추론 속도는 느리지만, 검출 정확도와 분류 정확도는 높음을 보인다.

• Journal of KIISE
• /
• v.44 no.1
• /
• pp.44-50
• /
• 2017

Pedestrian detection, an important component of autonomous driving and driving assistant system, has been extensively studied for many years. In particular, image based pedestrian detection methods such as Hierarchical classifier or HOG and, deep models such as ConvNet are well studied. The evaluation score has increased by the various methods. However, pedestrian detection requires high sensitivity to errors, since small error can lead to life or death problems. Consequently, further reduction in pedestrian detection error rate of autonomous systems is required. We proposed a new method to detect pedestrians and reduce the error rate by using the Faster R-CNN with new developed pedestrian training data sets. Finally, we compared the proposed method with the previous models, in order to show the improvement of our method.

• The Journal of the Acoustical Society of Korea
• /
• v.40 no.5
• /
• pp.466-472
• /
• 2021

Speaker Change Detection (SCD) refers to finding the moment when the main speaker changes from one person to the next in a speech conversation. In speaker change detection, difficulties arise due to overlapping speakers, inaccuracy in the information labeling, and data imbalance. To solve these problems, TIMIT corpus widely used in speech recognition have been concatenated artificially to obtain a sufficient amount of training data, and the detection of changing speaker has performed after identifying overlapping speakers. In this paper, we propose an speaker change detection system that considers the speaker overlapping. We evaluated and verified the performance using various approaches. As a result, a detection system similar to the X-Vector structure was proposed to remove the speaker overlapping region, while the Bi-LSTM method was selected to model the speaker change system. The experimental results show a relative performance improvement of 4.6 % and 13.8 % respectively, compared to the baseline system. Additionally, we determined that a robust speaker change detection system can be built by conducting related studies based on the experimental results, taking into consideration text and speaker information.