• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.32 no.2D
• /
• pp.129-136
• /
• 2012

In the US, Japan and Germany the Aerial-Based Vehicle Detection System, which collects real-time traffic data using the Unmanned Aerial Vehicle (UAV), helicopters or fixed-wing aircraft has been developed for the last several years. Therefore, this study was done to find out whether the Aerial-Based Vehicle Detection System could be used for real-time traffic data collection. For this purpose the study was divided into two parts. In the first part the possibility of retrieving real-time traffic data such as travel speed from the aerial photographic image using the image processing technique was examined. In the second part the quality of the retrieved real-time traffic data was examined to find out whether the data are good enough to be used as traffic information source. Based on the results of examinations we could conclude that it would not be easy for the Aerial- Based Vehicle Detection System to replace the present Vehicle Detection System due to technological difficulties and high cost. However, the system could be effectively used to make the emergency traffic management plan in case of incidents such as abrupt heavy rain, heavy snow, multiple pile-up, etc.

• Korean Journal of Remote Sensing
• /
• v.37 no.4
• /
• pp.747-761
• /
• 2021

With the increasing application of UAVs in intelligent transportation systems, vehicle detection for aerial images has become an essential engineering technology and has academic research significance. In this paper, a vehicle detection method for aerial images based on the YOLOv4 deep learning algorithm is presented. At present, the most known datasets are VOC (The PASCAL Visual Object Classes Challenge), ImageNet, and COCO (Microsoft Common Objects in Context), which comply with the vehicle detection from UAV. An integrated dataset not only reflects its quantity and photo quality but also its diversity which affects the detection accuracy. The method integrates three public aerial image datasets VAID, UAVD, DOTA suitable for YOLOv4. The training model presents good test results especially for small objects, rotating objects, as well as compact and dense objects, and meets the real-time detection requirements. For future work, we will integrate one more aerial image dataset acquired by our lab to increase the number and diversity of training samples, at the same time, while meeting the real-time requirements.

• Korean Journal of Remote Sensing
• /
• v.24 no.5
• /
• pp.473-481
• /
• 2008

This paper presents an approach to Back-propagation and Radial Basis Function neural network method with various training set for automatic vehicle detection from aerial images. The initial extraction of candidate object is based on Mean-shift algorithm with symmetric property of a vehicle structure. By fusing the density and the symmetry, the method can remove the ambiguous objects and reduce the cost of processing in the next stage. To extract features from the detected object, we describe the object as a log-polar shape histogram using edge strengths of object and represent the orientation and distance from its center. The spatial histogram is used for calculating the momentum of object and compensating the direction of object. BPNN and RBFNN are applied to verify the object as a vehicle using a variety of non-car training sets. The proposed algorithm shows the results which are according to the training data. By comparing the training sets, advantages and disadvantages of them have been discussed.

• Proceedings of the KSRS Conference
• /
• v.1
• /
• pp.260-262
• /
• 2006

One of future remote sensing techniques for transportation application is vehicle detection from the space, which could be the basis of measuring traffic volume and recognizing traffic condition in the future. This paper introduces an approach to vehicle detection using image object segmentation approach. The object-oriented image processing is particularly beneficial to high-resolution image classification of urban area, which suffers from noisy components in general. The project site was Dae-Jeon metropolitan area and a set of true color aerial images at 10cm resolution was used for the test. Authors investigated a variety of parameters such as scale, color, and shape and produced a customized solution for vehicle detection, which is based on a knowledge-based hierarchical model in the environment of eCognition. The highest tumbling block of the vehicle detection in the given data sets was to discriminate vehicles in dark color from new black asphalt pavement. Except for the cases, the overall accuracy was over 90%.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.32 no.1
• /
• pp.39-48
• /
• 2022

In this paper, we propose an anomaly detection system (ADS) to detect anomalies of the in-vehicle network for unmanned aerial vehicle (UAV). The proposed ADS detects the anomalies by measuring the similarity of status messages sequences periodically sent by the UAV to the ground control system. We defined three types of malicious message injection attacks that can be performed on the in-vehicle network of UAV and simulated those attack techniques in the Pixhawk4 quadcopter. The proposed ADS can detect abnormal sequences with accuracy of higher than 96%.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.4
• /
• pp.1989-2011
• /
• 2019

Vehicle detection based on aerial images is an interesting and challenging research topic. Most of the traditional vehicle detection methods are based on the sliding window search algorithm, but these methods are not sufficient for the extraction of object features, and accompanied with heavy computational costs. Recent studies have shown that convolutional neural network algorithm has made a significant progress in computer vision, especially Faster R-CNN. However, this algorithm mainly detects objects in natural scenes, it is not suitable for detecting small object in aerial view. In this paper, an accurate and effective vehicle detection algorithm based on Faster R-CNN is proposed. Our method fuse a hyperactive feature map network with Eltwise model and Concat model, which is more conducive to the extraction of small object features. Moreover, setting suitable anchor boxes based on the size of the object is used in our model, which also effectively improves the performance of the detection. We evaluate the detection performance of our method on the Munich dataset and our collected dataset, with improvements in accuracy and effectivity compared with other methods. Our model achieves 82.2% in recall rate and 90.2% accuracy rate on Munich dataset, which has increased by 2.5 and 1.3 percentage points respectively over the state-of-the-art methods.

• Journal of the Architectural Institute of Korea Structure & Construction
• /
• v.35 no.1
• /
• pp.11-19
• /
• 2019

Recently, with the development of the fourth industrial revolution that has been achieved through the fusion of information and communication technology (ICT), the technologies of AI, IOT, BIG-DATA, it is increasing utilization rate by industry and research and development of application technologies are being actively carried out. Especially, in the case of unmanned aerial vehicles, the construction market is expected to be one of the most commercialized areas in the world for the next decade. However, research on utilization of unmanned aerial vehicles in the construction field in Korea is insufficient. In this study, We have developed a quantitative building inspection method using the unmanned aerial vehicle and presented the protocol for it. The proposed protocol was verified by applying it to existing old buildings, and defect information could be quantified by calculating length, width, and area for each defect. Through this technical research, the final goal is to contribute to the development of safety diagnosis technology using unmanned aerial vehicle and risk assessment technology of buildings in case of disaster such as earthquake.

• Journal of the Society of Disaster Information
• /
• v.17 no.3
• /
• pp.556-567
• /
• 2021

Purpose: In this study, various methods of deep learning-based automatic damage analysis technology were reviewed based on images taken through Unmanned Aerial Vehicle to more efficiently and reliably inspect the exterior inspection and inspection of railway bridges using Unmanned Aerial Vehicle. Method: A deep learning analysis model was created by defining damage items based on the acquired images and extracting deep learning data. In addition, the model that learned the damage images for cracks, concrete and paint scaling·spalling, leakage, and Reinforcement exposure among damage of railway bridges was applied and tested with the results of automatic damage analysis. Result: As a result of the analysis, a method with an average detection recall of 95% or more was confirmed. This analysis technology enables more objective and accurate damage detection compared to the existing visual inspection results. Conclusion: through the developed technology in this study, it is expected that it will be possible to analysis more accurate results, shorter time and reduce costs by using the automatic damage analysis technology using Unmanned Aerial Vehicle in railway maintenance.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.39 no.6
• /
• pp.555-562
• /
• 2021

Recently, in the field of geospatial information construction, unmanned aerial vehicles have been increasingly used because they enable rapid data acquisition and utilization. In this study, photogrammetry was performed using fixed-wing, rotary-wing, and VTOL (Vertical Take-Off and Landing) unmanned aerial vehicles, and geospatial information was constructed using two types of unmanned aerial vehicle LiDAR (Light Detection And Ranging) sensors. In addition, the accuracy was evaluated to present the utility of spatial information constructed through unmanned aerial photogrammetry and LiDAR. As a result of the accuracy evaluation, the orthographic image constructed through unmanned aerial photogrammetry showed accuracy within 2 cm. Considering that the GSD (Ground Sample Distance) of the constructed orthographic image is about 2 cm, the accuracy of the unmanned aerial photogrammetry results is judged to be within the GSD. The spatial information constructed through the unmanned aerial vehicle LiDAR showed accuracy within 6 cm in the height direction, and data on the ground was obtained in the vegetation area. DEM (Digital Elevation Model) using LiDAR data will be able to be used in various ways, such as construction work, urban planning, disaster prevention, and topographic analysis.

• The Journal of Korea Robotics Society
• /
• v.9 no.3
• /
• pp.154-159
• /
• 2014

The Global Positioning System (GPS) is widely used to aid the navigation of aerial vehicles. However, the GPS cannot be used indoors, so alternative navigation methods are needed to be developed for micro aerial vehicles (MAVs) flying in GPS-denied environments. In this paper, a real-time three-dimensional (3-D) indoor navigation system and closed-loop control of a quad-rotor aerial vehicle equipped with an inertial measurement unit (IMU) and a low-cost light detection and ranging (LIDAR) is presented. In order to estimate the pose of the vehicle equipped with the two-dimensional LIDAR, an octree-based grid map and Monte-Carlo Localization (MCL) are adopted. The navigation results using the MCL are then evaluated by making a comparison with a motion capture system. Finally, the results are used for closed-loop control in order to validate its positioning accuracy during procedures for stable hovering and waypoint-following.