• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.39 no.6
• /
• pp.409-418
• /
• 2021

As the amount of construction for aerial photography increases, the need for automation of quality inspection is emerging. In this study, an experiment was performed to classify or detect clouds in aerial photos using deep learning techniques. Also, classification and detection were performed by including satellite images in the learning data. As algorithms used in the experiment, GoogLeNet, VGG16, Faster R-CNN and YOLOv3 were applied and the results were compared. In addition, considering the practical limitations of securing erroneous images including clouds in aerial images, we also analyzed whether additional learning of satellite images affects classification and detection accuracy in comparison a training dataset that only contains aerial images. As results, the GoogLeNet and YOLOv3 algorithms showed relatively superior accuracy in cloud classification and detection of aerial images, respectively. GoogLeNet showed producer's accuracy of 83.8% for cloud and YOLOv3 showed producer's accuracy of 84.0% for cloud. And, the addition of satellite image learning data showed that it can be applied as an alternative when there is a lack of aerial image data.

• Journal of The Korean Astronomical Society
• /
• v.22 no.1
• /
• pp.25-30
• /
• 1989

We report a null detection of $^{12}CO$ emission from a sub-condensation in a High Velocity Cloud (HVC). As a consequence of this, an upper limit of $n(H_2)\frac{X(CO)}{DV/DR}{\leq}2{\times}10^{-5}$ was set. This implies that $^{12}CO$ abundance is deficient by at least a factor of 10 if the HVC is predominantly molecular, otherwise the CO abundance of the HVC might be normal.

• Advances in robotics research
• /
• v.1 no.2
• /
• pp.127-140
• /
• 2014

This work presents an IR-based system for parking assistance and obstacle detection in the automotive field that employs the Microsoft Kinect camera for fast 3D point cloud reconstruction. In contrast to previous research that attempts to explicitly identify obstacles, the proposed system aims to detect "reachable regions" of the environment, i.e., those regions where the vehicle can drive to from its current position. A user-friendly 2D traversability grid of cells is generated and used as a visual aid for parking assistance. Given a raw 3D point cloud, first each point is mapped into individual cells, then, the elevation information is used within a graph-based algorithm to label a given cell as traversable or non-traversable. Following this rationale, positive and negative obstacles, as well as unknown regions can be implicitly detected. Additionally, no flat-world assumption is required. Experimental results, obtained from the system in typical parking scenarios, are presented showing its effectiveness for scene interpretation and detection of several types of obstacle.

• IEIE Transactions on Smart Processing and Computing
• /
• v.4 no.6
• /
• pp.403-406
• /
• 2015

In this paper, we propose a supervoxel clustering-based staircase extraction algorithm to obtain poses and dimensions of staircases from a point cloud. In order to effectively reduce the candidate points and accelerate supervoxel clustering, large planes in the scene, such as walls, floors, and ceilings, are eliminated while scanning the environment. Next, staircase candidates with small planes are initially estimated using supervoxel clustering. Then, parameter values for the staircases are refined, and higher staircases that remain undetected due to occlusion are predicted and generated virtually. Experimental results show that staircases are detected accurately and predicted successfully.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.5 no.6
• /
• pp.672-677
• /
• 2010

Recently, the efficient management system for large-scale sensor data has been required due to the increasing deployment of large-scale sensor networks. In this paper, we propose a cloud-based sensor data management system with low cast, high scalability, and efficiency. Sensor data in sensor networks are transmitted to the cloud through a cloud-gateway. At this point, outlier detection and event processing is performed. Transmitted sensor data are stored in the Hadoop HBase, distributed column-oriented database, and processed in parallel by query processing module designed as the MapReduce model. The proposed system can be work with the application of a variety of platforms, because processed results are provided through REST-based web service.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.8 no.11
• /
• pp.4226-4241
• /
• 2014

Cloud storage provides an easy, cost-effective and reliable way of data management for users without the burden of local data storage and maintenance. Whereas, this new paradigm poses many challenges on integrity and privacy of users' data, since users losing grip on their data after outsourcing the data to the cloud server. In order to address these problems, recently, Worku et al. have proposed an efficient privacy-preserving public auditing scheme for cloud storage. However, in this paper, we point out the security flaw existing in the scheme. An adversary, who is on-line and active, is capable of modifying the outsourced data arbitrarily and avoiding the detection by exploiting the security flaw. To fix this security flaw, we further propose a secure and efficient privacy-preserving public auditing scheme, which makes up the security flaw of Worku et al.'s scheme while retaining all the features. Finally, we give a formal security proof and the performance analysis, they show the proposed scheme has much more advantages over the Worku et al.'s scheme.

• Tunnel and Underground Space
• /
• v.19 no.1
• /
• pp.1-10
• /
• 2009

Site characterization for structural stability in rock masses mainly involves the collection of joint property data, and in the current practice, much of this data is collected by hand directly at exposed slopes and outcrops. There are many issues with the collection of this data in the field, including issues of safety, slope access, field time, lack of data quantity, reusability of data and human bias. It is shown that information on joint orientation, spacing and roughness in rock masses, can be automatically extracted from LIDAR (light detection and ranging) point floods using the currently available Split-FX point cloud processing software, thereby reducing processing time, safety and human bias issues.

• Bulletin of the Korean Chemical Society
• /
• v.32 no.8
• /
• pp.2637-2642
• /
• 2011

A new ultrasonic-assisted micellar extraction and cloud-point pre-concentration method was developed for the determination of major saikosaponins, namely saikosaponins -A, -C and -D, in Radix Bupleuri by high performance liquid chromatography with evaporative light scattering detection (HPLC-ELSD). The non-ionic surfactant Genapol X-080 (oligoethylene glycol monoalkyl ether) was chosen as the extraction additive and parameters affecting the extraction efficiency were optimized. The highest yield was obtained with 10% (w/v) Genapol X-080, a liquid/solid ratio of 200:1 (mL/g) and ultrasonic-assisted extraction for 40 min. In addition, the optimum cloud-point pre-concentration was reached with 10% sodium sulfate and equilibration at $60^{\circ}C$ for 30 min. Separation was achieved on an Ascentis Express C18 column (100 ${\times}$ 4.6 mm i.d., 2.7 ${\mu}M$) using a binary mobile phase composed of 0.1% acetic acid and acetonitrile. Saikosaponins were detected by ELSD, which was operated at a $50^{\circ}C$ drift tube temperature and 3.0 bar nebulizer gas ($N_2$) pressure. The water-based solvent modified with Genapol X-080 showed better extraction efficiency compared to that of the conventional solvent methanol. Recovery of saikosaponins ranged from 93.1 to 101.9%. An environmentally-friendly extraction method was successfully applied to extract and enrich major saikosaponins in Radix Bupleuri.

• Journal of Korea Multimedia Society
• /
• v.13 no.5
• /
• pp.754-762
• /
• 2010

Nowadays, email-based targeted attacks using malcode-bearing documents have been steadily increased. To improve the success rate of the attack and avoid anti-viruses, attackers mainly employ zero-day exploits and relevant social engineering techniques. In this paper, we propose an architecture of the email vaccine cloud system to prevent targeted attacks using malcode-bearing documents. The system extracts attached document files from email messages, performs behavior analysis as well as signature-based detection in the virtual machine environment, and completely removes malicious documents from the messages. In the process of behavior analysis, the documents are regarded as malicious ones in cases of creating executable files, launching new processes, accessing critical registry entries, connecting to the Internet. The email vaccine cloud system will help prevent various cyber terrors such as information leakages by preventing email based targeted attacks.

• Journal of Broadcast Engineering
• /
• v.28 no.3
• /
• pp.329-332
• /
• 2023

3D Object Detection using deep neural network has been developed a lot for obstacle detection in autonomous vehicles because it can recognize not only the class of target object but also the distance from the object. But in the case of 3D Object Detection models, the detection performance for distant objects is lower than that for nearby objects, which is a critical issue for autonomous vehicles. In this paper, we introduce a technique that increases the performance of 3D object detection models, particularly in recognizing distant objects, by generating virtual 3D vehicle data and adding it to the dataset used for model training. We used a spherical point tracing method that leverages the characteristics of 3D LiDAR sensor data to create virtual vehicles that closely resemble real ones, and we demonstrated the validity of the virtual data by using it to improve recognition performance for objects at all distances in model training.