• Journal of Broadcast Engineering
• /
• v.25 no.6
• /
• pp.944-953
• /
• 2020

Object detection plays a crucial role in a self-driving system. With the advances of image recognition based on deep convolutional neural networks, researches on object detection have been actively explored. In this paper, we proposed a lightweight model of the mask R-CNN, which has been most widely used for object detection, to efficiently predict location and shape of various objects on the road environment. Furthermore, feature maps are adaptively re-calibrated to improve the detection performance by applying an attention module to the neural network layer that plays different roles within the mask R-CNN. Various experimental results for real driving scenes demonstrate that the proposed method is able to maintain the high detection performance with significantly reduced network parameters.

• Journal of the Optical Society of Korea
• /
• v.17 no.2
• /
• pp.188-199
• /
• 2013

This paper presents a novel algorithm for nighttime detection of the lane markers painted on a road at night. First of all, the proposed algorithm uses neighborhood average filtering, 8-directional Sobel operator and thresholding segmentation based on OTSU's to handle raw lane images taken from a digital CCD camera. Secondly, combining intensity map and gradient map, we analyze the distribution features of pixels on boundaries of lanes in the nighttime and construct 4 feature sets for these points, which are helpful to supply with sufficient data related to lane boundaries to detect lane markers much more robustly. Then, the searching method in multiple directions- horizontal, vertical and diagonal directions, is conducted to eliminate the noise points on lane boundaries. Adapted Hough transformation is utilized to obtain the feature parameters related to the lane edge. The proposed algorithm can not only significantly improve detection performance for the lane marker, but it requires less computational power. Finally, the algorithm is proved to be reliable and robust in lane detection in a nighttime scenario.

• 제어로봇시스템학회:학술대회논문집
• /
• 1997.10a
• /
• pp.589-592
• /
• 1997

In this paper, we propose a simple algorithm to calculate the number of passing cars at night by using an image processing sensor for digital black and white images with 256 tone levels. To recognize cars, we capture their head lamps. The reflection of the head lamps is one of the most troublesome factors in recognizing cars. The main problem in this paper is how to recognize cars under the influence of the reflection of the head lamps especially in rainy days. In general, the image of a head lamp is nearly circular and the reflection is long and narrow. On the difference of these forms, we can exclude the reflection in our proposed algorithms For real-time operation and simple calculation, we recognize the existence of cars using fifteen lines with 256 tone levels. In the experimental application on a road, the recognition rate of a real-time operation is more than 90%. Moreover, we will also explain briefly how to recognize passing cars for 24 hours.

• 제어로봇시스템학회:학술대회논문집
• /
• 2003.10a
• /
• pp.1700-1705
• /
• 2003

In this paper, the automatic steering system for unmanned vehicle was developed. The vision system is used for the lane detection system. This paper defines two modes for detecting lanes on a road. First is searching mode and the other is recognition mode. We use inverse perspective transform and a linear approximation filter for accurate lane detections. The PD control theory is used for the design of the controller to compare with $H_{\infty}$ control theory. The $H_{\infty}$ control theory is used for the design of the controller to reduce the disturbance. The performance of the PD controller and $H_{\infty}$ controller is compared in simulations and tests. The PD controller is easy to tune in the test site. The $H_{\infty}$ controller is robust for the disturbances in the test results.

• Korean Journal of Computational Design and Engineering
• /
• v.21 no.1
• /
• pp.42-50
• /
• 2016

Traditional tactile controls that include push buttons and rotary switches may cause significant visual and biomechanical distractions if they are located away from the driver's line of sight and hand position, for example, on the central console. Gestural controls, as an alternative to traditional controls, are natural and can reduce visual distractions; however, their types and numbers are limited and have no feedback. To overcome the problems, a driver interface combining gestures and visual feedback with a head-up display has been proposed recently. In this paper, we investigated the effect of this type of interface in terms of driving performance measures. Human-in-the-loop experiments were conducted using a driving simulator with the traditional tactile and the new gesture-based interfaces. The experimental results showed that the new interface caused less visual distractions, better gap control between ego and target vehicles, and better recognition of road conditions comparing to the traditional one.

• Journal of the Korean Society of Systems Engineering
• /
• v.15 no.2
• /
• pp.1-8
• /
• 2019

Autonomous vehicles operated in the rough terrain environment must satisfy various technical requirements in order to improve the speed. Therefore, in order to design and implement a technical architecture that satisfies the requirements for speed improvement of autonomous vehicles, it is necessary to consider the overall technology of hardware and software to be mounted. In this study, the technical architecture of the autonomous vehicle operating in the rough terrain environment is presented. In order to realize high speed driving in pavement driving environment and other environment, it should be designed to improve the fast and accurate recognition performance and collect high quality database. and it should be determined the correct running speed from the running ability analysis and the frictional force estimation on the running road. We also improved synchronization performance by providing precise navigation information(time) to each hardware and software.

• Journal of Korea Multimedia Society
• /
• v.20 no.8
• /
• pp.1166-1174
• /
• 2017

In general, the detection of the vehicle license plate is a previous step of license plate recognition and has been actively studied for several decades. In this paper, we propose an algorithm to detect a license plate area of a moving vehicle from a video captured by a fixed camera installed on the road using the Convolution Neural Network (CNN) technology. First, license plate images and non-license plate images are applied to a previously learned CNN model (AlexNet) to extract and classify features. Then, after detecting the moving vehicle in the video, CNN detects the license plate area by comparing the features of the license plate region with the features of the license plate area. Experimental result shows relatively good performance in various environments such as incomplete lighting, noise due to rain, and low resolution. In addition, to protect personal information this proposed system can also be used independently to detect the license plate area and hide that area to secure the public's personal information.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.11 no.6
• /
• pp.3086-3103
• /
• 2017

Video images captured by vehicle cameras often contain blurry or dithering frames due to inadvertent motion from bumps in the road or by insufficient illumination during the morning or evening, which greatly reduces the perception of objects expression and recognition from the records. Therefore, a real-time electronic stabilization method to correct fuzzy video from driving recorders has been proposed. In the first stage of feature detection, a coarse-to-fine inspection policy and a scale nonlinear diffusion filter are proposed to provide more accurate keypoints. Second, a new antiblurry binary descriptor and a feature point selection strategy for unintentional estimation are proposed, which brought more discriminative power. In addition, a new evaluation criterion for affine region detectors is presented based on the percentage interval of repeatability. The experiments show that the proposed method exhibits improvement in detecting blurry corner points. Moreover, it improves the performance of the algorithm and guarantees high processing speed at the same time.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.10 no.6
• /
• pp.2483-2503
• /
• 2016

Traffic congestion is a severe problem in many modern cities around the world. Real-time and accurate traffic congestion identification can provide the advanced traffic management systems with a reliable basis to take measurements. The most used data sources for traffic congestion are loop detector, GPS data, and video surveillance. Video based traffic monitoring systems have gained much attention due to their enormous advantages, such as low cost, flexibility to redesign the system and providing a rich information source for human understanding. In general, most existing video based systems for monitoring road traffic rely on stationary cameras and multiple vehicle tracking method. However, most commonly used multiple vehicle tracking methods are lack of effective track initiation schemes. Based on the motion of the vehicle usually obeys constant velocity model, a novel vehicle recognition method is proposed. The state of recognized vehicle is sent to the GM-PHD filter as birth target. In this way, we relieve the insensitive of GM-PHD filter for new entering vehicle. Combining with the advanced vehicle detection and data association techniques, this multiple vehicle tracking method is used to identify traffic congestion. It can be implemented in real-time with high accuracy and robustness. The advantages of our proposed method are validated on four real traffic data.

• Journal of Environmental Science International
• /
• v.13 no.2
• /
• pp.117-127
• /
• 2004

The purpose of this study is to present the conditions and methods for keeping the sustainability of the landscape resources of the East Coast of Gangwon province, the Republic of Korea by investigating the landscape resources management in the area and pointing out its problems. The unit of analysis in this study is four cities and two counties located along the national road route seven in Gangwon province. The classification and characteristics of the landscape resources in this area was analyzed by a literature review, and we surveyed the tourists visiting the area and statistically analyzed the data in order to examine their satisfaction with the landscape resources management and make recommendations. The problems of the landscape resources management are: 1) the disturbance of the persistence of life by reclamation, the population reduction in the ecosystem due to the overload in environmental capacity, and the severance of space between land and water. 2) the reduction of the benefits from indirect experience by interfering with the conservation of fluxes --- the manipulation of horizontal arrangement of the landscape resource, the visual disturbance by the construction of high-storied buildings, and the disharmony between the color/image and the environment. The means for keeping sustainability of the landscape resources include the regulations of development and use, the change in the recognition of the value of landscape resources and the moral system, and the improvement of resource management skills.