• 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.20 no.3
• /
• pp.362-368
• /
• 2017

The endurance test is important to evaluate and predict defects of the vehicles. In particular, in order to improve the reliability of the endurance test for the powertrain system of military vehicles, the driving load is measured and analyzed at test courses and maneuvering roads. As a result, new test courses for powertrain system is additionally needed to improve test reliability. In this paper, it shows that new test courses for powertrain system is designed and constructed using the measuring and anlayzing result between CPG test courses and army maneuvering roads.

• Journal of the military operations research society of Korea
• /
• v.29 no.2
• /
• pp.81-99
• /
• 2003

This paper propose a method determining life cycle for military vehicle using availability analysis. Many studies determining life cycle for military equipments have been done recently However, those studies focused on economic life such as average system cost method, equivalent annual cost method and cumulative operations cost method. In many case, those results are not appropriate in deciding replacement in the field situation, we consider an effective life cycle method using availability concept. In order to determine an equipment life cycle. Two kinds of availability is considered. One is equipment yearly availability, the other is operational availability with operating distance per year. The life cycle is determined by achieving unit target availability level. The result using this concept for K­511 military vehicle life cycle is about 19 years, which is longer than previous studies.

• Journal of Applied Reliability
• /
• v.8 no.3
• /
• pp.135-144
• /
• 2008

The economic life for three types of military wheeled vehicles with load capacities of 1/4, $1{\cdot}1/4$, and $2{\cdot}1/2$ tones has been evaluated on the basis of the equivalent acquisition and operating costs. The economic life of wheeled vehicles were calculated from 12 to 18 years by using the annual equivalent cost method. The equivalent cost was decided at the lowest point of the total amount of equivalent acquisition cost and operating cost. The operating cost were collected from the field data. The evaluated economic life can be very useful for deciding the total life cycle of these three types of military vehicles. The annual equivalent cost method may be also applied to other military equipments such as communication electronics, weapon systems, and other type of vehicles.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.21 no.5
• /
• pp.443-450
• /
• 2020

This paper reports a design improvement study to solve the stoppage phenomenon caused by the launch-support device applied to K105A1. The K105A1 is a weapon system equipped with an old 105 mm towed howitzer in a wheeled vehicle, which provides superior maneuverability compared to track equipment. The launch support device serves to withstand fire impact and load. In this way, this device is fixed firmly to the ground in preparation for the shooting mission and is responsible for the critical performance, such as fixing the position of the vehicle. On the other hand, during the field test, a temporary stoppage of the launch support occurred, which caused a problem of not being fixed to the ground. To solve this problem, the cause of failure was analyzed by a replay test and parts inspection. In addition, the operating concept, method, and design were analyzed to derive the cause and solve the problem by changing the parts design. Finally, the performance and firing missions were performed normally by applying the changed design to K105A1. The performance stability and reliability of the launch support device were confirmed, which are expected to be of great assistance in the development of military equipment in the future.