• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.38 no.4
• /
• pp.422-433
• /
• 2001

In this paper, we propose a flew edge detection method based on water flow model, in which gradient image surface is considered as a 3-dimensional (3-D) geographical feature. The edges of the objects in the background can be detected by the large gradient magnitude areas and to make the edges immersed it is required to invert the gradient image. The proposed edge detector uses a water flow model based enhancement and locally adaptive thresholding technique applied to the inverted gradient image resulting in better noise performance. Computer simulations with a few synthetic and real images show that the Proposed method can extract edge contour effectively.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.38 no.4
• /
• pp.325-334
• /
• 2001

It is shown that the multiscale prewhitening matched filter for detecting Gaussian objects in Markov noise can be implemented by the undecimated wavelet transform with a biorthogonal spline wavelet. If the object to be detected is Gaussian shaped and its scale coincides with one of those computed by the wavelet transform, and if the background noise is truly Markov, then optimum detection is realized by thresholding the appropriate details image. Our detection algorithm is applied to the digitized mammograms for detecting microcalcifications. However, microcalcifications are not exactly Gaussian shaped and its background noise may not be Markov. In order to campensate for these discrepancy, Hotelling observer is employed, which is applied to feature vectors comprised of 3-octave wavelet coefficients.

• Smart Media Journal
• /
• v.4 no.4
• /
• pp.9-17
• /
• 2015

Recently the need for advanced security technologies are increasing as the occurrence of intelligent crime is growing fastly. Previous liveness detection and fake face detection methods are required for the improvement of accuracy in order to be put to practical use. In this paper, we propose a new liveness detection method using pupil reflection, and new fake image detection using Adaboost detector. The proposed system detects eyes based on multi-scale Gabor feature vector in the first stage, The template matching plays a role in determining the allowed eye area. And then, the reflected image in the pupil is used to decide whether or not the captured image is live or not. Experimental results indicate that the proposed method is superior to the previous methods in the detection accuracy of fake images.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.10 no.12
• /
• pp.5643-5656
• /
• 2016

The detection accuracy of steganalysis depends on many factors, including the embedding algorithm, the payload size, the steganalysis feature space and the properties of the cover source. In practice, the cover source mismatch (CSM) problem has been recognized as the single most important factor negatively affecting the performance. To address this problem, we propose a new framework for blind, universal steganalysis which uses traditional steganalyst features. Firstly, cover images with the same statistical properties are searched from a reference image database as aided samples. The test image and its aided samples form a whole test set. Then, by assuming that most of the aided samples are innocent, we conduct outlier detection on the test set to judge the test image as cover or stego. In this way, the framework has removed the need for training. Hence, it does not suffer from cover source mismatch. Because it performs anomaly detection rather than classification, this method is totally unsupervised. The results in our study show that this framework works superior than one-class support vector machine and the outlier detector without considering the image retrieval process.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2016.05a
• /
• pp.830-833
• /
• 2016

자동차 번호판 인식은 대중적인 감시 기술 중의 한 종류로서, 주어진 비디오나 영상 내 광학문자 인식을 수반한다. 고속도로나 국도 상에 과속 단속 시스템, 재형 건물이나 유통센서 및 주차장 등에서 주차 정산 시스템, 고속도로 톨 게이트에서 hi-pass 에러 및 불법 도주 차량 잔속 시스템, 전국 주요 도로 불법 주 정차 단속 시스템, 공공기관, 기업 출퇴근 시간 확인 및 외부 차양 안내 시스템 등의 지능형 교통 시스템(ITS)이나 국도 상에 범위 차량 검거 시스템, 사건 발생 시 주요 도로상에 설치된 CCTV를 통해 용의 차량 이동 추적 시스템, 이동식 범죄 차량 조회, 버스에 탑재된 버스 전용차선 위반 단속들의 지능형 방범 시스템 등에 활용하고 있다. 번호판 인식은 자동차 번호판 국부화, 번호판의 크기, 차원, 명암대비, 밝기를 조정하는 정규화, 개별문자를 얻어내는 문자 분할, 문자를 인식하는 광학 문자 인식, 번호판의 형태, 크기, 위치 들이 연도별, 지역별로 차이가 있는 번호판들의 데이터베이스를 비교하여 구문 분석을 하는 절차를 거친다. 본 논문에서는 EmguCV를 이용하여 구현한 번호판 감지를 수행하여 위치를 찾아내고, 오픈 소스 광학 문자 인식 엔진으로 잘 알려져 있는 테서렉트 OCR을 이용하여 번호판의 문자를 인식하는 자동 인식 프로그램을 구현하고 기술하였다.

• The Journal of Korea Robotics Society
• /
• v.1 no.1
• /
• pp.25-35
• /
• 2006

Improving practicality of SLAM requires various sensors to be fused effectively in order to cope with uncertainty induced from both environment and sensors. In this case, combining sonar and vision sensors possesses numerous advantages of economical efficiency and complementary cooperation. Especially, it can remedy false data association and divergence problem of sonar sensors, and overcome low frequency SLAM update caused by computational burden and weakness in illumination changes of vision sensors. In this paper, we propose a SLAM method to join sonar sensors and stereo camera together. It consists of two schemes, extracting robust point and line features from sonar data and recognizing planar visual objects using multi-scale Harris corner detector and its SIFT descriptor from pre-constructed object database. And fusing sonar features and visual objects through EKF-SLAM can give correct data association via object recognition and high frequency update via sonar features. As a result, it can increase robustness and accuracy of SLAM in indoor environment. The performance of the proposed algorithm was verified by experiments in home -like environment.

• Journal of Korea Multimedia Society
• /
• v.11 no.4
• /
• pp.454-461
• /
• 2008

This paper is regarding a new input interface based on displacement of mobile devices having a camera. The mobile device can capture consecutive images by the camera, the displacement of the device is estimated by computing the displacement between consecutive images in real-time. The proposed system extracts feature points based on SUSAN comer detector which has low computational complexity. We generate Voronoi domain by using the two-pass algorithm to match extracted features. Finally, the displacement of a mobile device is estimated by calculating SAD values between two consecutive images. We evaluated the performance of the proposed algorithm with 1500 images. True matching accuracy of the proposed algorithm is 90% and the computation for each image is conducted in 5m sec.

• Smart Structures and Systems
• /
• v.24 no.6
• /
• pp.745-757
• /
• 2019

Road surface deteriorations such as potholes have caused motorists heavy monetary damages every year. However, effective road condition monitoring has been a continuing challenge to road owners. Depth cameras have a small field of view and can be easily affected by vehicle bouncing. Traditional image processing methods based on algorithms such as segmentation cannot adapt to varying environmental and camera scenarios. In recent years, novel object detection methods based on deep learning algorithms have produced good results in detecting typical objects, such as faces, vehicles, structures and more, even in scenarios with changing object distances, camera angles, lighting conditions, etc. Therefore, in this study, a Deep Learning Pothole Detector (DLPD) based on the deep region-based convolutional neural network is proposed for autonomous detection of potholes from images. About 900 images with potholes and road surface conditions are collected and divided into training and testing data. Parameters of the network in the DLPD are calibrated based on sensitivity tests. Then, the calibrated DLPD is trained by the training data and applied to the 215 testing images to evaluate its performance. It is demonstrated that potholes can be automatically detected with high average precision over 93%. Potholes can be differentiated from manholes by training and applying a manhole-pothole classifier which is constructed using the convolutional neural network layers in DLPD. Repeated detection of the same potholes can be prevented through feature matching of the newly detected pothole with previously detected potholes within a small region.

• The Korean Journal of Physiology and Pharmacology
• /
• v.23 no.2
• /
• pp.131-139
• /
• 2019

Manually reviewing electroencephalograms (EEGs) is labor-intensive and demands automated seizure detection systems. To construct an efficient and robust event detector for experimental seizures from continuous EEG monitoring, we combined spectral analysis and deep neural networks. A deep neural network was trained to discriminate periodograms of 5-sec EEG segments from annotated convulsive seizures and the pre- and post-EEG segments. To use the entire EEG for training, a second network was trained with non-seizure EEGs that were misclassified as seizures by the first network. By sequentially applying the dual deep neural networks and simple pre- and post-processing, our autodetector identified all seizure events in 4,272 h of test EEG traces, with only 6 false positive events, corresponding to 100% sensitivity and 98% positive predictive value. Moreover, with pre-processing to reduce the computational burden, scanning and classifying 8,977 h of training and test EEG datasets took only 2.28 h with a personal computer. These results demonstrate that combining a basic feature extractor with dual deep neural networks and rule-based pre- and post-processing can detect convulsive seizures with great accuracy and low computational burden, highlighting the feasibility of our automated seizure detection algorithm.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.5
• /
• pp.1671-1686
• /
• 2022

Face manipulation tools represented by Deepfake have threatened the security of people's biological identity information. Particularly, manipulation tools with deep learning technology have brought great challenges to Deepfake detection. There are many solutions for Deepfake detection based on traditional machine learning and advanced deep learning. However, those solutions of detectors almost have problems of poor performance when evaluated on different quality datasets. In this paper, for the sake of making high-quality Deepfake datasets, we provide a preprocessing method based on the image pixel matrix feature to eliminate similar images and the residual channel attention network (RCAN) to resize the scale of images. Significantly, we also describe a Deepfake detector named Cascaded-Hop which is based on the PixelHop++ system and the successive subspace learning (SSL) model. By feeding the preprocessed datasets, Cascaded-Hop achieves a good classification result on different manipulation types and multiple quality datasets. According to the experiment on FaceForensics++ and Celeb-DF, the AUC (area under curve) results of our proposed methods are comparable to the state-of-the-art models.