• Journal of the Korea Institute of Military Science and Technology
• /
• v.27 no.3
• /
• pp.338-345
• /
• 2024

Magnetic anomaly signals from the ferromagnetic targets such as ships in the sea are measured by drone-based magnetic anomaly detection. A quantum magnetometer is suspended from the drone by 4 strings. Flight altitude and speed of drone are 100 m and 5 m/s, respectively. We obtain magnetic anomaly signals of few nT from the ships clearly. We analyze the signal characteristics by the ferromagnetic target through simulation using COMSOL multiphysics.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.10 no.3
• /
• pp.47-63
• /
• 2011

A traffic detection system can be deemed as a traffic data and information collection system to serve traffic policies, traffic management, and user services. The system plays a crucial role in verifying whether or not the current traffic system has issues or problems by checking out traffic data. In addition, the system does so in finding out a point or a section where an issue or a problem has occurred, if any, and in examining the causes of the issue or problem, the extent of its impact that has occurred and spread, and a method for resolving it. However, the existing point detection system of Korea has too many flaws. In order to fix the flaws, in this paper, the theoretical characteristics of the section detection system were researched in relation to the calculation of travel time. In addition, the travel time of probe cars was obtained by field survey, and it was compared to that of spot and section detection data. Then, simulation was performed to determine the optimal section detection interval. In conclusion, introduction of optimal section detection system was examined in order to achieve the advanced road management including traffic policy, traffic management, and user services.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.29 no.4
• /
• pp.785-794
• /
• 2019

Recently Deep Learning technology, one of the fourth industrial revolution technologies, is used to identify the hidden meaning of network data that is difficult to detect in the security arena and to predict attacks. Property and quality analysis of data sources are required before selecting the deep learning algorithm to be used for intrusion detection. This is because it affects the detection method depending on the contamination of the data used for learning. Therefore, the characteristics of the data should be identified and the characteristics selected. In this paper, the characteristics of malware were analyzed using network data set and the effect of each feature on performance was analyzed when the deep learning model was applied. The traffic classification experiment was conducted on the comparison of characteristics according to network characteristics and 96.52% accuracy was classified based on the selected characteristics.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.15 no.11
• /
• pp.2365-2373
• /
• 2011

This paper propose the hardware IP of real-time face detection system for mobile devices and digital cameras required for high speed, smaller size and lower power. The proposed face detection system is robust against illumination changes, face size, and various face angles as the main cause of the face detection performance. Input image is transformed to LBP(Local Binary Pattern) image to obtain face characteristics robust against illumination changes, and detected the face using face feature data that was adopted to learn and generate in the various face angles using the Adaboost algorithm. The proposed face detection system can be detected maximum 36 faces at the input image size of QVGA($320{\times}240$), and designed by Verilog-HDL. Also, it was verified hardware implementation by using Virtex5 XC5VLX330 FPGA board and HD CMOS image sensor(CIS) for FPGA verification.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.18 no.1
• /
• pp.210-216
• /
• 2014

The edge in the image is a part which the brightness changes rapidly between the object and the object or objects and background, and includes information of the features such as size, position, orientation, and texture of the object. The edge detection is the technique that acquires these information of the images, and now the researches to detect edges are making steady progress. Typical conventional edge detection methods are Sobel, Prewitt, Roberts using the first derivative operator and Laplacian method using the second derivative operator and so on. These methods is more or less insufficient that the characteristics of the edge detection in the image added salt and pepper noise. therefore, in this paper, an edge detection algorithm using modified mask that applies different size mask according to noise density of local mask is proposed.

• Journal of IKEEE
• /
• v.13 no.2
• /
• pp.232-240
• /
• 2009

This paper presents a robust vehicle detection and tracking algorithm for supervision of illegal parking. The proposed algorithm is composed of four parts. First, a vehicle detection algorithm is proposed using the improved codebook object detection algorithm to segment moving vehicles from the input sequence. Second, a preprocessing technique using the geometric characteristics of vehicles is employed to exclude non-vehicle objects. Then, the detected vehicles are tracked by an object tracker which incorporates histogram tracking method with Kalman filter. To make the tracking results more accurate, histogram tracking results are used as measurement data for Kalman filter. Finally, Real Stop Counter (RSC) is introduced for trustworthy and accurate performance of the stopped vehicle detection. Experimental results show that the proposed algorithm can track multiple vehicles simultaneously and detect stopped vehicles successfully in the complicated street environment.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.15 no.1
• /
• pp.179-186
• /
• 2015

This paper proposes the target path detection algorithm using statistical characteristics of an activated time lag along a moving path of target from a neighboring sensor in PDR(Pulse Doppler Radar) sensor node environment based on USN(Ubiquitous Sensor Network) with a limitation detecting only an existence of moving target. In the proposed algorithm, detection and non-detection time lag obtained from the experimental data are used. The experimental data are through repetitive action of each 500 times about three path scenarios such as passing in between two sensors, moving parallel to two sensors, and turning through two sensors. From this experiments, error detection percentages of three path scenarios are 5.67%, 5.83%, and 7.17%, respectively. They show that the proposed algorithm can exactly detect a target path using the limited PDR sensor nodes.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.43 no.2 s.344
• /
• pp.56-65
• /
• 2006

In this paper, we introduce modified difference type symbol timing detection method of simple structure and show the relations between S/N ratio and timing detection performance which less influenced by multipath channel delay profile and added noise level and it show very exact GI detection performance characteristics. In the computer simulations, 4 symbol time duration of short and long training of IEEE802.11a standard OFDM frame are used for symbol synchronization timing detection. The computer simulation results show the very exact symbol timing detection performance characteristic within 1 sample error of OFDM signal regardless channel delay profile from minimn phase channels of phase rotation ${\pi}/2$ to non-minimum phase channels of phase rotation ${\pi}/2$ of received OFDM signal and added noise level in channel.

• Journal of Broadcast Engineering
• /
• v.18 no.3
• /
• pp.401-408
• /
• 2013

In this paper, we propose a method that automatically detects organs having similar appearances in medical images by learning both context and appearance features. Since only the appearance feature is used to learn the classifier in most existing detection methods, detection errors occur when the medical images include multiple organs having similar appearances. In the proposed method, based on the probabilities acquired by the appearance-based classifier, new classifier containing the context feature is created by iteratively learning the characteristics of probability distribution around the interest voxel. Furthermore, both the efficiency and the accuracy are improved through 'region based voting scheme' in test stage. To evaluate the performance of the proposed method, we detect femur and tibia which have similar appearance from SKI10 knee joint dataset. The proposed method outperformed the detection method only using appearance feature in aspect of overall detection performance.

• Proceedings of the KSRS Conference
• /
• 2003.11a
• /
• pp.1460-1463
• /
• 2003

In this study, plastic detection technique was developed, applying remote sensing technology as a method to extract plastic wastes, which is one of the big causes of concern contributing to environmental destruction. It is possible to extract areas where plastic (including polypropylene and polyethylene) wastes are prominent, using ASTER data by taking advantage of its absorptive characteristics of ASTER/SWIR bands. The algorithm is applicable to define large industrial wastes disposal sites and areas where plastic greenhouses are concentrated. However, the detection technique with ASTER/SWIR data has some research tasks to be tackled, which includes a partial secretion of reference spectral, depending on some conditions of plastic wastes and a detection error in a region mixed with vegetations and waters. Following results were obtained after making comparisons between several detection methods and plastic wastes in different conditions; (a)'spectral extraction method' was suitable for areas where plastic wastes exist separated from other objects, such as coastal areas where plastic wastes drifted ashore. (single plastic spectral was used as a reference for the 'spectral extraction method') (b)On the other hand, the 'spectral extraction method' was not suitable for sites where plastic wastes are mixed with vegetation and soil. After making comparison of the processing results of a mixed area, it was found that applying both 'separation method' using un-mixing and ‘spectral extraction method’ with NDVI masked is the most appropriate method to extract plastic wastes. Also, we have investigated the possibility of reducing the influence of vegetation and water, using ASTER/TIR, and successfully extracted some places with plastics. As a conclusion, we have summarized the relationship between detection techniques and conditions of plastic wastes and propose the practical application of remote sensing technology to the extraction of plastic wastes.