• Journal of the Korea Convergence Society
• /
• v.7 no.5
• /
• pp.29-34
• /
• 2016

Recently, with the development of high-tech industry, the use of the drones in various aspects of daily life is rapidly advancing. With technical and functional advancements, drones have an advantage of being easy to be utilized in the areas of use according to various lifestyles. In addition, through the diversification of the drone service converged with image processing medium such as camera and CCTV, an automated security system that can replace humans is expected to be introduced. By designing these unmanned security technology, a new convergence security drone service techniques that can strengthen the previous drone application technology will be proposed. In the proposed techniques, a biometric authentication technology will be designed as additional authentication methods that can determine the safety incorporated with security by selecting the search and areas of an object focusing on the objects in the initial windows and search windows through OpenCV technology and CAM-Shift algorithm which are an object tracking algorithm. Through such, a highly efficient security drone convergence service model will be proposed for performing unmanned security by using the drones that can continuously increase the analysis of technology on the mobility and real-time image processing.

• Korean Chemical Engineering Research
• /
• v.57 no.2
• /
• pp.274-282
• /
• 2019

In recent years, big data analysis has been expanded to include automatic control through reinforcement learning as well as prediction through modeling. Research on the utilization of image data is actively carried out in various industrial fields such as chemical, manufacturing, agriculture, and bio-industry. In this paper, we applied NASNet, which is an AutoML reinforced learning algorithm, to DeepU-Net neural network that modified U-Net to improve image semantic segmentation performance. We used BRATS2015 MRI data for performance verification. Simulation results show that DeepU-Net has more performance than the U-Net neural network. In order to improve the image segmentation performance, remove dropouts that are typically applied to neural networks, when the number of kernels and filters obtained through reinforcement learning in DeepU-Net was selected as a hyperparameter of neural network. The results show that the training accuracy is 0.5% and the verification accuracy is 0.3% better than DeepU-Net. The results of this study can be applied to various fields such as MRI brain imaging diagnosis, thermal imaging camera abnormality diagnosis, Nondestructive inspection diagnosis, chemical leakage monitoring, and monitoring forest fire through CCTV.

• Journal of the Korea Society of Computer and Information
• /
• v.26 no.1
• /
• pp.45-55
• /
• 2021

This paper has proposed an algorithm that detecting for dense small vehicle in large image efficiently. It is consisted of two Ensemble Deep-Learning Network algorithms based on Coarse to Fine method. The system can detect vehicle exactly on selected sub image. In the Coarse step, it can make Voting Space using the result of various Deep-Learning Network individually. To select sub-region, it makes Voting Map by to combine each Voting Space. In the Fine step, the sub-region selected in the Coarse step is transferred to final Deep-Learning Network. The sub-region can be defined by using dynamic windows. In this paper, pre-defined mapping table has used to define dynamic windows for perspective road image. Identity judgment of vehicle moving on each sub-region is determined by closest center point of bottom of the detected vehicle's box information. And it is tracked by vehicle's box information on the continuous images. The proposed algorithm has evaluated for performance of detection and cost in real time using day and night images captured by CCTV on the road.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.21 no.6
• /
• pp.183-190
• /
• 2011

In this paper, we propose a prototype system for abnormal sound detection and identification which detects and recognizes the abnormal situations by means of analyzing audio information coming in real time from CCTV cameras under surveillance environment. The proposed system is composed of two layers: The first layer is an one-class support vector machine, i.e., support vector data description (SVDD) that performs rapid detection of abnormal situations and alerts to the manager. The second layer classifies the detected abnormal sound into predefined class such as 'gun', 'scream', 'siren', 'crash', 'bomb' via a sparse representation classifier (SRC) to cope with emergency situations. The proposed system is designed in a hierarchical manner via a mixture of SVDD and SRC, which has desired characteristics as follows: 1) By fast detecting abnormal sound using SVDD trained with only normal sound, it does not perform the unnecessary classification for normal sound. 2) It ensures a reliable system performance via a SRC that has been successfully applied in the field of face recognition. 3) With the intrinsic incremental learning capability of SRC, it can actively adapt itself to the change of a sound database. The experimental results with the qualitative analysis illustrate the efficiency of the proposed method.

• Journal of IKEEE
• /
• v.21 no.2
• /
• pp.142-145
• /
• 2017

Currently, a trend in image processing is high-quality and high-resolution. The size and amount of image data are increasing exponentially because of the development of information and communication technology. Thus, license plate recognition with a single processor cannot handle the increasing data. This paper proposes a number plate recognition system using a distributed processing framework, Hadoop. Using SequenceFile format in Hadoop, each mapper performs a license plate recognition with a number of image data in a data block Experimental results show that license plate recognition performance with 16 data nodes accomplishes speedup of maximum 14.7 times comparing with one data node. In large dataset, the recognition performance is robust even if the number of data nodes increases gradually.

• Journal of Broadcast Engineering
• /
• v.25 no.1
• /
• pp.105-108
• /
• 2020

This paper presents a real-time low-resolution face recognition method that uses a super-resolution technique. Conventional face recognition methods are limited by low accuracy resulting from the distance between the camera and objects. Although super-resolution methods have been developed to resolve this issue, they are not suitable for integrated face recognition systems. The proposed method recognizes faces with low resolution using key frame selection, super resolution, face detection, and recognition on real-time processing. Experiments involving several databases indicated that the proposed algorithm is superior to conventional methods in terms of face recognition accuracy.

• Journal of Environmental Science International
• /
• v.24 no.4
• /
• pp.403-414
• /
• 2015

This study is daytime and nighttime runoff image data caused by heavy rain on May 27, 2013 at Oedo Water Treatment Plant of Oedo-Stream, Jeju to compute runoff by applying Surface image velocimeter (SIV) and analyzing correlation according to current. At the same time, current was comparatively analyzed using ADCP observation data and fixed electromagnetic surface current meter (Kalesto) observed at the runoff site. As a result of comparison on resolutions of daytime and nighttime runoff images collected, correlation coefficient corresponding to the range of 0.6~0.7 was 6.8% higher for nighttime runoff image compared to daytime runoff image. On the contrary, correlation coefficient corresponding to the range of 0.9~1.0 was 17% lower. This result implies that nighttime runoff image has lower image quality than daytime runoff image. In the process of computing current using SIV, a rational filtering process for correlation coefficient is needed according to images obtained.

• Journal of Environmental Science International
• /
• v.24 no.4
• /
• pp.359-369
• /
• 2015

This study analyzed the velocimetry of runoff and measured the flood discharge by applying the SIV (Surface Image Velocimetrer) to the daytime and nighttime flow image data with special reference to Seong-eup Bridge at Cheonmi stream of Jeju during the flow by the severe rainstorm on May 27, 2013. A 1000W lighting apparatus with more than 150 lux was installed in order to collect proper nighttime flow image applied to the SIV. Its value was compared and analyzed with the velocity value of the fixed electromagnetic wave surface velocimetry (Kalesto) at the same point to check the accuracy and applicability of the measured velocity of flow. As a result, determination coefficient $R^2$ values were 0.891 and 0.848 respectively in line with the velocity distribution of the daytime and nighttime image and the flow volume measured with Kalesto was approximately 18.2% larger than the value measured with the SIV.

• Annual Conference of KIPS
• /
• 2016.04a
• /
• pp.86-87
• /
• 2016

본 논문에서는 optical flow sensor를 이용하여 실내의 바닥 영상인식를 통한 영상처리기법을 이용해 움직임 없는 hovering을 할 수 있는 방법을 제안한다. 또한 optical flow와 BLE finger printing 기법을 혼합해 위치 인식 정밀도를 높일 수 있다. 본 고에서는 optical flow sensor와 BLE finger printing의 두 기술을 혼합하면 드론 스스로 실내에서 정밀도 높은 위치인식이 가능 하며 실외에서만 사용할 수 있는 GPS 비행모드를 대신 할 수 있어 실내에서 자동 경로 비행이 가능하게 하고 위치 안내, 실내 방송촬영, 이동식 CCTV등 질 높은 서비스를 제공하고자 한다.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.8 no.11
• /
• pp.1647-1654
• /
• 2013

Detection performance of the vehicle on the road depends on weather conditions, the shadow by the movement of the sun, or illumination changes, etc. In this paper, a vehicle detection system in conjunction with a robust background estimate algorithm to environment change on the road in daytime is proposed. Gaussian Mixture Model is applied as background estimation algorithm, and also, Adaboost algorithm is applied to detect the vehicle for candidate region. Through the experiments with input videos obtained from a various weather conditions at the same actual road, the proposed algorithm were useful to detect vehicles in the road.