• Korean Journal of Artificial Intelligence
• /
• v.12 no.3
• /
• pp.1-7
• /
• 2024

In this paper, we propose an innovative super-resolution technique to address the issue of reduced accuracy in license plate recognition caused by low-resolution images. Conventional vehicle license plate recognition systems have relied on images obtained from fixed surveillance cameras for traffic detection to perform vehicle detection, tracking, and license plate recognition. However, during this process, image quality degradation occurred due to the physical distance between the camera and the vehicle, vehicle movement, and external environmental factors such as weather and lighting conditions. In particular, the acquisition of low-resolution images due to camera performance limitations has been a major cause of significantly reduced accuracy in license plate recognition. To solve this problem, we propose a Single Image Super-Resolution (SISR) model with a parallel structure that combines Multi-Scale and Attention Mechanism. This model is capable of effectively extracting features at various scales and focusing on important areas. Specifically, it generates feature maps of various sizes through a multi-branch structure and emphasizes the key features of license plates using an Attention Mechanism. Experimental results show that the proposed model demonstrates significantly improved recognition accuracy compared to existing vehicle license plate super-resolution methods using Bicubic Interpolation.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.64 no.10
• /
• pp.1528-1534
• /
• 2015

This paper designed the railway vehicle running device with a proto-type for the railway vehicle maintenance training and developed a propulsion control device simulator equipped the imitation steering wheel. In addition, this paper applied a multi-thread technology in order to implement the staged fault and the propulsion control device protected operation test and an implementation of the failure that occur in actual rail vehicle and confirm the validity as the propulsion control device simulator for the maintenance training.

• Journal of Auto-vehicle Safety Association
• /
• v.14 no.3
• /
• pp.71-76
• /
• 2022

In autonomous driving cars, 3D pose estimation can be one of the effective methods to enhance safety control for OOP (Out of Position) passengers. There have been many studies on human pose estimation using a camera. Previous methods, however, have limitations in automotive applications. Due to unexplainable failures, CNN methods are unreliable, and other methods perform poorly. This paper proposes robust real-time multi-human 3D pose estimation architecture in vehicle using monocular RGB camera. Using particle filter, our approach integrates CNN 2D/3D pose measurements with available information in vehicle. Computer simulations were performed to confirm the accuracy and robustness of the proposed algorithm.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2007.11a
• /
• pp.409-414
• /
• 2007

Nowadays, with the advancement of computational mechanics, and vehicle dynamics simulation linked up with virtual testing laboratory(VTL) and virtual proving ground(VPG) technologies has become a useful method for analyzing numerous driving performances and diverse noise/vibration characteristics. In this paper, the analytical vehicle model based on multi-body dynamics theory was developed to investigate the vibration characteristics according to various road conditions. For the purpose, the whole vehicle parameters, each vehicle's part parameter, and part connecting elements such as spring, damper, and bush were measured by an experiment. Also, the vehicle dynamics model, which includes the front suspension, rear suspension, steering, front wheel, rear wheel, and body subsystems has been constructed for computer simulation. With the developed vehicle dynamics model, three forces and three moments measured at each wheel center were applied to evaluate and analyze dynamics and vibration characteristics for miscellaneous road conditions.

• Proceedings of the Korean Society for Technology of Plasticity Conference
• /
• 2008.10a
• /
• pp.277-280
• /
• 2008

Multi layer bellows are being manufactured for commercial vehicle because of the characteristic of high durability compared with single iaγor bellows used to passenger vehicle. Finite Element Method (FEM) study and optimization about single layer bellows are actively progressed, but FEM study about multi layer bellows which have gap between layer is rarely processed. Therefore, this article presents finite element modeling of multi layer bellows for the improvement of simulation reliability. For the shape optimization of multi layer bellows, design of experiment and Taguchi method are used.

• Computers and Concrete
• /
• v.19 no.6
• /
• pp.725-736
• /
• 2017

This paper presents the two-, three-, and four-lane transverse reduction factor based on FEA method, probability theory, and the recently actual traffic flow data. A total of 72 composite girder bridges with various spans, number of lanes, loading mode, and bridge type are analyzed with time-varying static load FEA method by ANSYS, and the probability models of vehicle load effects at arbitrary-time point are developed. Based on these probability models, in accordance to the principle of the same exceeding probability, the multi-lane transverse reduction factor of these composite girder bridges and the relationship between the multi-lane transverse reduction factor and the span of bridge are determined. Finally, the multi-lane transverse reduction factor obtained is compared with those from AASHTO LRFD, BS5400, JTG D60 or Eurocode. The results show that the vehicle load effect at arbitrary-time point follows lognormal distribution. The two-, three-, and four-lane transverse reduction factors calculated by using FEA method and probability respectively range between 0.781 and 1.027, 0.616 and 0.795, 0.468 and 0.645. Furthermore, a correlation between the FEA and AASHTO LRFD, BS5400, JTG D60 or Eurocode transverse reduction factors is made for composite girder bridges. For the two-, three-, and four-lane bridge cases, the Eurocode code underestimated the FEA transverse reduction factors by 27%, 25% and 13%, respectively. This underestimation is more pronounced in short-span bridges. The AASHTO LRFD, BS5400 and JTG D60 codes overestimated the FEA transverse reduction factors. The FEA results highlight the importance of considering span length in determining the multi-lane transverse reduction factors when designing two-lane or more composite girder bridges. This paper will assist bridge engineers in quantifying the adjustment factors used in analyzing and designing multi-lane composite girder bridges.

• The Journal of Korea Robotics Society
• /
• v.12 no.2
• /
• pp.152-160
• /
• 2017

Recent studies on automatic parking have actively adopted the technology developed for mobile robots. Among them, the path planning scheme plans a route for a vehicle to reach a target parking position while satisfying the kinematic constraints of the vehicle. However, previous methods require a large amount of computation and/or cannot be easily applied to different environmental conditions. Therefore, there is a need for a path planning scheme that is fast, efficient, and versatile. In this study, we use a multi-dimensional path grid map to solve the above problem. This multi-dimensional path grid map contains a route which has taken a vehicle's kinematic constraints into account; it can be used with the $A^*$ algorithm to plan an efficient path. The proposed method was verified using Prescan which is a simulation program based on MATLAB. It is shown that the proposed scheme can successfully be applied to both parallel and vertical parking in an efficient manner.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.22 no.11
• /
• pp.2432-2438
• /
• 1997

This paper proposes recognition system using multi-backpropagation neural networks rather than single backpropagation neural network to enhance the rate of character recognition resultsing from extracting the region of velhicle number in that the image of vehicle number plate from CCD camera has a distinguish feature, that is, illumination of a pattern. The experiment in this paper shows an output that the method using multi-backpropagation neural networks rather than signal backpropagation neural network takes less training time for computation and also has higher recognition rage of vehicle number.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.61 no.5
• /
• pp.735-743
• /
• 2012

In this paper, we present model predictive control (MPC) applied to lane keeping system (LKS) based on a vision module. Due to a slow sampling rate of the vision system, the conventional LKS using single rate control may result in uncomfortable steering control rate in a high vehicle speed. By applying MPC using multi-rate Kalman filter to active steering control, the proposed MPC-based active steering control system prevents undesirable saturated steering control command. The effectiveness of the MPC is validated by simulations for the LKS equipped with a camera module having a slow sampling rate on the curved lane with the minimum radius of 250[m] at a vehicle speed of 30[m/s].

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 1997.10a
• /
• pp.535-538
• /
• 1997

Studied in this paper was the lateral displacement of railway vehicle using the multi-body dynamic simulation program (VAMPIRE) and the BASS 501. The lateral displacement of railway vehicle is occurred by thc clearance between wheel flange and rail, the track irregularity, the property of each suspension of vehicle and the cant etc. The results of analysis shown that Vehicle is not interfere with subway platform in any conditions namely the tare and full load condition, the wheel. wear condition and the stationary and running of vehicle.