• Journal of the Korea Institute of Military Science and Technology
• /
• v.26 no.5
• /
• pp.392-399
• /
• 2023

This research paper investigates the effectiveness of using computer graphics(CG) based synthetic data for deep learning in military vehicle detection. In particular, we explore the use of synthetic image generation techniques to train deep neural networks for object detection tasks. Our approach involves the generation of a large dataset of synthetic images of military vehicles, which is then used to train a deep learning model. The resulting model is then evaluated on real-world images to measure its effectiveness. Our experimental results show that synthetic training data alone can achieve effective results in object detection. Our findings demonstrate the potential of CG-based synthetic data for deep learning and suggest its value as a tool for training models in a variety of applications, including military vehicle detection.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.25 no.4
• /
• pp.381-389
• /
• 2022

Detecting partially occluded objects is difficult due to the appearances and shapes of occluders are highly variable. These variabilities lead to challenges of localizing accurate bounding box or classifying objects with visible object parts. To address these problems, we propose a two-stage part-based attention approach for robust object detection under partial occlusion. First, our part attention network(PAN) captures the important object parts and then it is used to generate weighted object features. Based on the weighted features, the re-weighted object features are produced by our reinforced PAN(RPAN). Experiments are performed on our collected military vehicle dataset and synthetic occlusion dataset. Our method outperforms the baselines and demonstrates the robustness of detecting objects under partial occlusion.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.25 no.3
• /
• pp.300-310
• /
• 2022

A vast amount of labeled data is required for deep neural network training. A typical strategy to improve the performance of a neural network given a training data set is to use data augmentation technique. The goal of this work is to offer a novel image augmentation method for improving object detection accuracy. An object in an image is removed, and a similar object from the training data set is placed in its area. An in-painting algorithm fills the space that is eliminated but not filled by a similar object. Our technique shows at most 2.32 percent improvements on mAP in our testing on a military vehicle dataset using the YOLOv4 object detector.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.10 no.2
• /
• pp.12-20
• /
• 2007

The aiming ability is a key to improve the accuracy performance of the gun pointing system in the ground military vehicle. This paper describes the new detection method of chatter vibration using disturbance acceleration in the pointing structure. In order to analysis the vibration trends of the pointing system occurred while the vehicle driving, acceleration data obtained from vehicle was processed by using data processing algorithm with moving average and Hilbert transform. The specific mode constants of acceleration were obtained from various disturbances. Vehicle velocity, road condition and property of pointing structure were considered as factors which make the change of vibration trend in vehicle dynamics. Finally, back propagation neural networks have been applied to the pattern recognition of the classification of vibration signal in various driving conditions. Results of signal processing were compared with other condition result and analysed.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.15 no.1
• /
• pp.9-15
• /
• 2012

This paper shows an effective scanning trajectory for a mine detection device that is one of the mission equipments of unmanned ground vehicle. The mine detection device is composed of a mine-detection sensor, and a 4 DOF manipulator enabling sensor position control. There are three modes that manage the mine detection device: passive, semi-automatic, and automatic. The automatic mode is used the most. This paper suggests a scanning method that makes shape of 8. This method prevents missing target area and enhances scanning speed when the mine detection device scans the ground surface in automatic mode. The suggested method is verified by simulations and experiments.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.20 no.4
• /
• pp.558-565
• /
• 2017

The vehicle-mounted mine detection system with large detection sensor modules can search wide areas with a fast detection speed. To mount the heavy mine detectors on a manned or unmanned vehicle, it is necessary to design the detector driving mechanism and control system based on the considerations driven from the characteristic analysis and the operation requirements of the detection system. Furthermore, while operating the mine detector mounted on a mobile vehicle, it is significant to keep the height from the ground to sensors within a certain distance in order to get a qualified detection performance. As the mine detection sensor, we used ground penetrating radar widely used to geotechnical exploration, mine detection and etc. In this paper, we introduce a driving mechanism through analyzing the characteristics of the vehicle-mounted mine detection system. We also suggest a method to automatically control the distance between the ground and GPR by utilizing the GPR output values, used to detect mines at the same time.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.25 no.1
• /
• pp.30-44
• /
• 2022

The advent of deep learning-based algorithms has facilitated researches on target detection from synthetic aperture radar(SAR) imagery. While most of them concentrate on detection tasks for ships with open SAR ship datasets and for aircraft from SAR scenes of airports, there is relatively scarce researches on the detection of SAR ground vehicle targets where several adverse factors such as high false alarm rates, low signal-to-clutter ratios, and multiple targets in close proximity are predicted to degrade the performances. In this paper, a dataset of ground vehicle targets acquired from TerraSAR-X(TSX) satellite SAR images is presented. Then, both detection and instance segmentation are simultaneously carried out on this dataset based on the deep learning-based Mask R-CNN. Finally, this paper shows the future research directions to further improve the performances of detecting the SAR ground vehicle targets.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.10 no.3
• /
• pp.139-147
• /
• 2007

In military transportation, the use of wide trailer for transporting the large and heavy weight equipments such as tank, armoured vehicle, and mobile gunnery is quite common. So, the vulnerability of causing traffic accidents for these wide military trailer to bump or collide with another car in adjacent lane is very high due to its broad width in excess of its own lane's width. Also, the possibility of these strayed accidents can be increased especially by the careless driver. In this paper, the recognition system of lane and vehicle headway direction is developed to detect the possible collision and warn the driver to prevent the fatal accident. In the system development, Kalman filtering is used first to extract the border of driving lane from the video images supplied by the CCD camera attached to the vehicle and the driving lane detection is completed with regression analysis. Next, the vehicle headway direction is recognized by using neural network scheme with the extracted parameters of the detected driving lane feature. The practical experiments for the developed system are also carried out in the real traffic road of Seoul city area and the results show us the more than 90% accuracy in recognizing the driving lane and vehicle headway direction.

• Korean Journal of Computational Design and Engineering
• /
• v.18 no.3
• /
• pp.155-166
• /
• 2013

We present recent human-vehicle interaction (HVI) research conducted in the area of defense and military application. Research topics discussed in this paper include: training simulation for overland navigation tasks; expertise effects in overland navigation performance and scan patterns; pilot's perception and confidence on an overland navigation task; effects of UAV (Unmanned Aerial Vehicle) supervisory control on F-18 formation flight performance in a simulator environment; autonomy balancing in a manned-unmanned teaming (MUT) swarm attack, enabling visual detection of IED (Improvised Explosive Device) indicators through Perceptual Learning Assessment and Training; usability test on DaViTo (Data Visualization Tool); and modeling peripheral vision for moving target search and detection. Diverse and leading HVI study in the defense domain suggests future research direction in other HVI emerging areas such as automotive industry and aviation domain.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.20 no.5
• /
• pp.708-716
• /
• 2017

We deal with the target selection in seeker-detection image through network, using the detection information from aerial vehicle and the target information from surveillance and reconnaissance system. Especially, we constrain the sea battle environment, where it is difficult to perform scene-matching rather than land. In this paper, we suggest the target selection algorithm based on the confidence estimation with respect to distance and size. In detail, we propose the generation method of reference point for distance evaluation, and we investigate the effect of pixel margin and target course for size evaluation. Finally, the proposed algorithm is simulated and analyzed through several scenarios.