• IEMEK Journal of Embedded Systems and Applications
• /
• v.17 no.1
• /
• pp.41-49
• /
• 2022

This paper proposes a novel lane detection algorithm based on inverse perspective transformation and machine learning in lightweight embedded system. The inverse perspective transformation method is presented for obtaining a bird's-eye view of the scene from a perspective image to remove perspective effects. This method requires only the internal and external parameters of the camera without a homography matrix with 8 degrees of freedom (DoF) that maps the points in one image to the corresponding points in the other image. To improve the accuracy and speed of lane detection in complex road environments, machine learning algorithm that has passed the first classifier is used. Before using machine learning, we apply a meaningful first classifier to the lane detection to improve the detection speed. The first classifier is applied in the bird's-eye view image to determine lane regions. A lane region passed the first classifier is detected more accurately through machine learning. The system has been tested through the driving video of the vehicle in embedded system. The experimental results show that the proposed method works well in various road environments and meet the real-time requirements. As a result, its lane detection speed is about 3.85 times faster than edge-based lane detection, and its detection accuracy is better than edge-based lane detection.

• Journal of IKEEE
• /
• v.27 no.4
• /
• pp.531-536
• /
• 2023

This paper proposes a sliding window method with frame feature insertion for immediate attack detection on in-vehicle networks. This method guarantees real-time attack detection by labeling based on the attack status of the current frame. Experiments show that the proposed method improves detection performance by giving more weight to the current frame in CNN computation. The proposed model was designed based on a lightweight LeNet-5 architecture and it achieves 100% detection for DoS attacks. Additionally, by comparing the complexity with conventional models, the proposed model has been proven to be more suitable for resource-constrained devices like ECUs.

• Journal of Information Processing Systems
• /
• v.19 no.5
• /
• pp.614-630
• /
• 2023

Dead trees significantly impact forest production and the ecological environment and pose constraints to the sustainable development of forests. A lightweight YOLOv4 dead tree detection algorithm based on unmanned aerial vehicle images is proposed to address current limitations in dead tree detection that rely mainly on inefficient, unsafe and easy-to-miss manual inspections. An improved logarithmic transformation method was developed in data pre-processing to display tree features in the shadows. For the model structure, the original CSPDarkNet-53 backbone feature extraction network was replaced by MobileNetV3. Some of the standard convolutional blocks in the original extraction network were replaced by depthwise separable convolution blocks. The new ReLU6 activation function replaced the original LeakyReLU activation function to make the network more robust for low-precision computations. The K-means++ clustering method was also integrated to generate anchor boxes that are more suitable for the dataset. The experimental results show that the improved algorithm achieved an accuracy of 97.33%, higher than other methods. The detection speed of the proposed approach is higher than that of YOLOv4, improving the efficiency and accuracy of the detection process.

• Spatial Information Research
• /
• v.19 no.4
• /
• pp.73-79
• /
• 2011

With the recent increase in demand for geo-registered imagery, Mobile Mapping Systems(MMS), which can quickly construct geographic information, has become important. The main part of MMS is the high-precision observation system, which collects geographic information at a certain speed. MMS has a complex data generation process and requires a standard-specific vehicle for its use, limiting its application range. In this paper, lightweight MMS is proposed to overcome its complexity by replacing the time synchronizer with a high-speed camera and by stabilizing motion with MEMS IMU/GPS. The proposed low-cost, portable method is expected to produce of geo-registered imagery efficiently.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.20 no.12
• /
• pp.758-764
• /
• 2019

The chassis of the Korean light tactical vehicle adopts a frame-on-body structure that uses lightweight design technology in terms of equipment operating characteristics. Military vehicles are operated in much harsher conditions compared to civilian vehicles, including mountainous terrain, especially steep slopes. Due to this characteristic, frame-welded cracks were found on some military vehicles. Therefore, in this paper, road damage analysis was conducted by identifying various roads including the military unit's road. The result confirmed that the operating environment of some military units' tactical road was much harsher than the endurance road test condition. A solution was derived through defect analysis, design review, and actual vehicle driving test. This study can be used as a reference by suggesting the development direction for the durability test of a new vehicle.

• Journal of Broadcast Engineering
• /
• v.25 no.5
• /
• pp.776-788
• /
• 2020

In this paper, a vehicle type recognition system using deep learning and a license plate recognition system are proposed. In the existing system, the number plate area extraction through image processing and the character recognition method using DNN were used. These systems have the problem of declining recognition rates as the environment changes. Therefore, the proposed system used the one-stage object detection method YOLO v3, focusing on real-time detection and decreasing accuracy due to environmental changes, enabling real-time vehicle type and license plate character recognition with one RGB camera. Training data consists of actual data for vehicle type recognition and license plate area detection, and synthetic data for license plate character recognition. The accuracy of each module was 96.39% for detection of car model, 99.94% for detection of license plates, and 79.06% for recognition of license plates. In addition, accuracy was measured using YOLO v3 tiny, a lightweight network of YOLO v3.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.38 no.5
• /
• pp.513-520
• /
• 2014

An unmanned aerial vehicle (UAV) should be designed to be as small and lightweight as possible to optimize the efficiency of changing the blade shape to enhance the aerodynamic performance, such as the thrust and power. In this study, a computational fluid dynamics (CFD) simulation of an unmanned multi-rotor aerial vehicle in hover mode was performed to explore the thrust performance in terms of the blade rotational speed and blade shape parameters (i.e., taper ratio and twist angle). The commercial ADINA-CFD program was used to generate the CFD data, and the results were compared with those obtained from blade element theory (BET). The results showed that changes in the blade shape clearly affect the aerodynamic thrust of a UAV rotor blade.

• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.22 no.2
• /
• pp.115-122
• /
• 2012

Sound packages and damping materials have been widely applied on the floor to decrease the interior noise of a vehicle. Based on the previous researches on the low-noise vehicles, weight optimization through minimization of damping material usage is required while decreasing mid and high frequency range noise by application of sound packages. This paper describes the analysis process of robust design of vehicle body structure before applying damping materials and focuses on the analysis and test process of the location optimization at the stage of damping material application. A vibration experiment for the analysis of floor panel velocity with respect to the excitation of suspension attachment parts at the underfloor of a vehicle is performed. And through the improvement correlation between FEA and TEST, a design guide to optimize damping materials application in the early design stage is proposed. A research on vibration damping steel sheets and liquid acoustic spray on deadener(LASD) is performed to minimize manufacturing time and to minimize the space for pre-existing asphalt damping materials. As results of this study, panel stiffness is achieved through curved surface panel and bead optimization. And test baseline of optimum design is suggested through damping material optimization. And finally, through re-establishing the analysis process for vibration reduction of vehicle floors and lightweight design of damping materials, it is possible to design damping materials efficiently in the preceding stage of design.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.32 no.4
• /
• pp.327-332
• /
• 2008

The effectiveness of vehicle parts made through extrusion is in the limelight because of the advantages of high strength stiffness materials can be produced and the number of processes can be drastically reduced. Therefore, the parts should have sufficient stiffness and be lightweight enough to improve fuel efficiency. However, the application of extruded aluminum requires pre-bending technologies that can manufacture the complex designs profiles demanded by vehicle parts. The aim of this research is that the development of the variable curvature extrusion technology that can produce a variety of curvature. In order to produce a variable curvature, the guide transfer speed and transfer time should be controlled properly. The guide transfer speed and transfer time were examined by the theoretical analysis. A model was developed to simulate the deformation behaviors of extrusion and bending process from the symmetric bumper with range of radii from 1863mm to 2163mm. The theoretical analysis and FE analysis were verified through experimental method.

• Journal of the Korean Society of Manufacturing Technology Engineers
• /
• v.26 no.1
• /
• pp.59-65
• /
• 2017

A seat track product is a car seat part that provides a base for vehicle seats. An ultra-high strength steel sheet is used to reduce the weight of vehicle body parts. However, the formability of an ultra-high strength steel sheet is poor because of its very low elongation and very high elastic deformation. For this reason, a new forming technology of an ultra-high strength steel sheet is required. The influence of spring-back of seat track parts on the shape freeze in forming processes was investigated to be solved by adjusting the appropriate tool design such as minus clearance between punch and die, and punch angle. This paper describes how to apply the spring-back prevention technique for improving shape freeze by using the ultra-high strength steel sheet with 980MPa to develop lightweight seat tract parts.