• Journal of Korea Multimedia Society
• /
• v.19 no.12
• /
• pp.1909-1918
• /
• 2016

This paper presents a unified framework for joint Convolutional Neural Network (CNN) based vehicle detection by leveraging multi-spectral image pairs. With the observation that under challenging environments such as night vision and limited light source, vehicle detection in a single color image can be more tractable by using additional far-infrared (FIR) image, we design joint CNN architecture for both RGB and FIR image pairs. We assume that a score map from joint CNN applied to overall image can be considered as confidence of vehicle existence. To deal with various scale ratios of vehicle candidates, multi-scale images are first generated scaling an image according to possible scale ratio of vehicles. The vehicle candidates are then detected on local maximal on each score maps. The generation of overlapped candidates is prevented with non-maximal suppression on multi-scale score maps. The experimental results show that our framework have superior performance than conventional methods with a joint framework of multi-spectral image pairs reducing false positive generated by conventional vehicle detection framework using only single color image.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.12 no.4
• /
• pp.191-197
• /
• 2004

This paper presents experimental results for human drivers' driving patterns and an Adaptive Cruise Control(ACC) strategy. Analyses have shown that female drivers' driving characteristic values such as time-gap and minimum clearance are larger than those of male drivers'. Human drivers tend to have more clearance margins at high speed than at low speed. At low speed, drivers are much more sensitive to the desired clearance than at high speed. A multi-vehicle detection method is presented to improve ride quality of an ACC. Simulation results have shown that the proposed ACC can provide superior performance compared to the ACC strategy which uses a single-vehicle detection method.

• 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].

• Journal of Broadcast Engineering
• /
• v.24 no.4
• /
• pp.573-579
• /
• 2019

To achieve fully autonomous driving, the perceptual skills of the surrounding environment must be superior to those of humans. The $60^{\circ}$ angle, $120^{\circ}$ wide angle cameras, which are used primarily in autonomous driving, have their disadvantages depending on the viewing angle. This paper uses a multi-angle object recognition system to overcome each of the disadvantages of wide and narrow-angle cameras. Also, the aspect ratio of data acquired with wide and narrow-angle cameras was analyzed to modify the SSD(Single Shot Detector) algorithm, and the acquired data was learned to achieve higher performance than when using only monocular cameras.

• Journal of Information Processing Systems
• /
• v.17 no.2
• /
• pp.411-425
• /
• 2021

In this paper, a vehicle recognition algorithm based on deep convolutional neural network and compression dictionary is proposed. Firstly, the network structure of fine vehicle recognition based on convolutional neural network is introduced. Then, a vehicle recognition system based on multi-scale pyramid convolutional neural network is constructed. The contribution of different networks to the recognition results is adjusted by the adaptive fusion method that adjusts the network according to the recognition accuracy of a single network. The proportion of output in the network output of the entire multiscale network. Then, the compressed dictionary learning and the data dimension reduction are carried out using the effective block structure method combined with very sparse random projection matrix, which solves the computational complexity caused by high-dimensional features and shortens the dictionary learning time. Finally, the sparse representation classification method is used to realize vehicle type recognition. The experimental results show that the detection effect of the proposed algorithm is stable in sunny, cloudy and rainy weather, and it has strong adaptability to typical application scenarios such as occlusion and blurring, with an average recognition rate of more than 95%.

• International Journal of Highway Engineering
• /
• v.10 no.3
• /
• pp.19-31
• /
• 2008

Video image processing systems(VIPS) offer numerous benefits to transportation models and applications, due to their ability to monitor traffic in real time. VIPS based on wide-area detection, i.e., multi-lane surveillance algorithm provide traffic parameters with single camera such as flow and velocity, as well as occupancy and density. However, most current commercial VIPS utilize a tripwire detection algorithm that examines image intensity changes in the detection regions to indicate vehicle presence and passage, i.e., they do not identify individual vehicles as unique targets. If VIPS are developed to track individual vehicles and thus trace vehicle trajectories, many existing transportation models will benefit from more detailed information of individual vehicles. Furthermore, additional information obtained from the vehicle trajectories will improve incident detection by identifying lane change maneuvers and acceleration/deceleration patterns. The objective of this research was to relate traffic safety to VIPS tracking and this paper has developed a computer vision system of monitoring individual vehicle trajectories based on image processing, and offer the detailed information, for example, volumes, speed, and occupancy rate as well as traffic information via tripwire image detectors. Also the developed system has been verified by comparing with commercial VIP detectors.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.10
• /
• pp.5095-5111
• /
• 2019

A multi-rotor Unmanned Aerial Vehicle (UAV) system is developed to solve the manhole cover detection problem for the infrastructure maintenance in the suburbs of big city. The visible light sensor is employed to collect the ground image data and a series of image processing and machine learning methods are used to detect the manhole cover. First, the image enhancement technique is employed to improve the imaging effect of visible light camera. An imaging environment perception method is used to increase the computation robustness: the blind Image Quality Evaluation Metrics (IQEMs) are used to percept the imaging environment and select the images which have a high imaging definition for the following computation. Because of its excellent processing effect the adaptive Multiple Scale Retinex (MSR) is used to enhance the imaging quality. Second, the Single Shot multi-box Detector (SSD) method is utilized to identify the manhole cover for its stable processing effect. Third, the spatial coordinate of manhole cover is also estimated from the ground image. The practical applications have verified the outdoor environment adaptability of proposed algorithm and the target detection correctness of proposed system. The detection accuracy can reach 99% and the positioning accuracy is about 0.7 meters.

• ETRI Journal
• /
• v.45 no.4
• /
• pp.650-665
• /
• 2023

A novel smart metering technique capable of anomaly detection was proposed for real-time home power management system. Smart meter data generated in real-time were obtained from 900 households of single apartments. To detect outliers and missing values in smart meter data, a deep learning model, the autoencoder, consisting of a graph convolutional network and bidirectional long short-term memory network, was applied to the smart metering technique. Power management based on the smart metering technique was executed by multi-objective optimization in the presence of a battery storage system and an electric vehicle. The results of the power management employing the proposed smart metering technique indicate a reduction in electricity cost and amount of power supplied by the grid compared to the results of power management without anomaly detection.

• International Journal of Highway Engineering
• /
• v.10 no.2
• /
• pp.159-166
• /
• 2008

Video image processing systems (VIPS) offer numerous benefits to transportation models and applications, due to their ability to monitor traffic in real time. VIPS based on a wide-area detection algorithm provide traffic parameters such as flow and velocity as well as occupancy and density. However, most current commercial VIPS utilize a tripwire detection algorithm that examines image intensity changes in the detection regions to indicate vehicle presence and passage, i.e., they do not identify individual vehicles as unique targets. If VIPS are developed to track individual vehicles and thus trace vehicle trajectories, many existing transportation models will benefit from more detailed information of individual vehicles. Furthermore, additional information obtained from the vehicle trajectories will improve incident detection by identifying lane change maneuvers and acceleration/deceleration patterns. However, unlike human vision, VIPS cameras have difficulty in recognizing vehicle movements over a detection zone longer than 100 meters. Over such a distance, the camera operators need to zoom in to recognize objects. As a result, vehicle tracking with a single camera is limited to detection zones under 100m. This paper develops a methodology capable of monitoring individual vehicle trajectories based on image processing. To improve traffic flow surveillance, a long distance tracking algorithm for use over 200m is developed with multi-closed circuit television (CCTV) cameras. The algorithm is capable of recognizing individual vehicle maneuvers and increasing the effectiveness of incident detection.

• Journal of Positioning, Navigation, and Timing
• /
• v.11 no.3
• /
• pp.173-179
• /
• 2022

The satellite navigation deception technology disturbs the navigation solution of the receiver by generating a deceptive signal simulating the actual satellite for the satellite navigation receiver mounted on the unmanned aerial vehicle, which is the target of deception. A single spoofing technique that creates a single deceptive position and velocity can be divided into a synchronized spoofing signal that matches the code delay, Doppler frequency, and navigation message with the real satellite and an unsynchronized spoofing signal that does not match. In order to generate a signal synchronized with a satellite signal, a very sophisticated and high precision signal generation technology is required. In addition, the current position and speed of the UAV equipped with the receiver must be accurately detected in real time. Considering the detection accuracy of the current radar technology that detects small UAVs, it is difficult to detect UAVs with an accuracy of less than one chip. In this paper, we assume the asynchrony of a single spoofing signal and propose a region defense technique using multiple spoofing signals.