• Journal of Korean Society of Transportation
• /
• v.27 no.6
• /
• pp.79-88
• /
• 2009

The objective of this study is to develop the new vehicle classification algorithm and minimize classification errors. The existing vehicle classification algorithm collects data from loop and piezo sensors according to the specification("Vehicle classification guide for traffic volume survey" 2006) given by the Ministry of Land, Transport and Maritime Affairs. The new vehicle classification system collects the vehicle length, distance between axles, axle type, wheel-base and tire type to minimize classification error. The main difference of new system is the "Wandering" sensor which is capable of measuring the wheel-base and tire type(single or dual). The wandering sensor obtains the wheel-base and tire type by detecting both left and right tire imprint. Verification tests were completed with the total traffic volume of 762,420 vehicles in a month for the new vehicle classification algorithm. Among them, 47 vehicles(0.006%) were not classified within 12 vehicle types. This results proves very high level of classification accuracy for the new system. Using the new vehicle classification algorithm will improve the accuracy and it can be broadly applicable to the road planning, design, and management. It can also upgrade the level of traffic research for the road and transportation infrastructure.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.14 no.6
• /
• pp.120-126
• /
• 2006

Vehicle dynamic simulation has been carried out using the coherence of road roughness between left and right wheels. The generated twin tracks with the coherence of road roughness between left and right wheels are in good agreements with the measured coherence relation of left and right wheels. And these tracks reflect well on the roughness characteristics of real roads. Using the generated roads and multibody dynamic simulation program, vehicle dynamic simulation is performed. The vertical and roll motion analysis of a vehicle are carried out using the realistic road profiles with the coherence between left and right wheels and the results are in good agreements with the dynamic characteristics of a vehicle.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.65 no.10
• /
• pp.1731-1737
• /
• 2016

In this paper, a method is proposed to recognize human actions as nonverbal expression; the proposed method is composed of two steps which are action representation and action recognition. First, MHI(Motion History Image) is used in the action representation step. This method includes segmentation based on depth information and generates spatio-temporal templates to describe actions. Second, CNN(Convolution Neural Network) which includes feature extraction and classification is employed in the action recognition step. It extracts convolution feature vectors and then uses a classifier to recognize actions. The recognition performance of the proposed method is demonstrated by comparing other action recognition methods in experimental results.

• The Transactions of the Korean Institute of Electrical Engineers
• /
• v.37 no.8
• /
• pp.563-569
• /
• 1988

This research presents several velocity modification methods for collision avoidance of two manipulators in a common workspace. Due to the distinct nature of collision avoidance between the two manipulators, a new classification of collision situations is presented and utilized in planning a collision-free path. Concepts of a collision map and velocity modification are applied for realizing collision-free motion planning. An example is shown for velocity modification of a trajectory, which shows the significance of the proposed approaches in collision-free motion planneng of two moving robots.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.65 no.1
• /
• pp.194-199
• /
• 2016

EMG has specific information which is related to movements according to the activities of muscles. Therefore, users can intuitively control a prosthesis. For this reason, biosignals are very useful and convenient in this kind of application. Bioimpednace also provides specific information about movements like EMG. In this study, we used both EMG and bioimpedance to classify the typical hand gestures such as hand open, hand close, no motion (rest), supination, and pronation. Nine able-bodied subjects and one amputee were used as experimental data set. The accuracy was $98{\pm}1.9%$ when 2 bio-impedance and 8 EMG channels were used together for normal subjects. The number of EMG channels affected the accuracy, but it was stable when more than 5 channels were used. For the amputee, the accuracy is higher when we use both of them than when using only EMG. Therefore, accurate and stable hand motion estimation is possible by adding bioimepedance which shows structural information and EMG together.

• Journal of Broadcast Engineering
• /
• v.7 no.4
• /
• pp.291-299
• /
• 2002

This paper proposes an efficient motion compensation algorithm for video sequences with brightness variations. In the proposed algorithm, the brightness variation parameters are estimated and local motions are compensated. To detect the frame with large brightness variations. we employ the frame classification based on the cross entropy between histograms of two successive frames, which can reduce the computational redundancy. Simulation results show that the proposed method yields a higher peak signal to noise ratio (PSNR) than the conventional methods, with a low computational load, when the video scene contains large brightness changes.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 1993.06a
• /
• pp.1254-1257
• /
• 1993

Presented in this paper is a newly developed motion planning method of an autonomous mobile robot(MAR) which can be applied to flexible manufacturing systems(FMS). The mobile robot is designed for transporting tools and workpieces between a set-up station and machines according to production schedules of the whole FMS. The proposed method is implemented based on an earlier developed real-time obstacle avoidance method which employs Kohonen network for pattern classification of sonar readings and fuzzy logic for local path planning. Particulary, a novel obstacle avoidance method for moving objects using a collision index, collision possibility measure, is described. Our method has been tested on the SNU mobile robot. The experimental results show that the robot successfully navigates to its target while avoiding moving objects.

• IE interfaces
• /
• v.16 no.1
• /
• pp.85-93
• /
• 2003

The purposes of this study are to investigate perceived discomfort for static joint motions, and to propose rankings for the joint motions based on the perceived discomfort. The perceived discomfort was measured through an experiment using the free modulus method of the magnitude estimation, in which ten healthy college-age female students participated. The results showed that joints, joint motions and their levels significantly affected the perceived discomfort at $\alpha$=0.01, and that the interaction of joints and joint motion levels was also significant. Based on the experimental results, three rankings were proposed by joint and joint motions, by joints and by joint motions, which were very different from the existing ones. Especially, the proposed rankings were different from the males' published before in their order and magnitude. These rankings can be used as a valuable tool for better understanding potentially adverse effects of poor working postures in industrial sites, and as basic data for developing the postural classification scheme.

• Proceedings of the Earthquake Engineering Society of Korea Conference
• /
• 2003.03a
• /
• pp.61-68
• /
• 2003

The H/V ratio (Horizontal to Vertical spectral ratio) has been used to infer site amplification without previous knowledge of near-surface geology and in fact may provide useful general site condition information. This method is used to classify the site characteristics of seismic stations in Korea by comparison with known H/V ratios representative of various sites all over the world. In addition, differences between horizontal and vertical kappa values were evaluated for each seismic stations by comparing WV ratio and Weak Motion amplification derived from inversion of stochastic ground motion parameters and were used as index to quantitatively classify the site characteristics.

• Journal of the Korea Society of Computer and Information
• /
• v.24 no.6
• /
• pp.21-28
• /
• 2019

Recent years with the development of artificial intelligence and the success of the deep model, they have been deployed in all fields of computer vision. Action recognition, as an important branch of human perception and computer vision system research, has attracted more and more attention. Action recognition is a challenging task due to the special complexity of human movement, the same movement may exist between multiple individuals. The human action exists as a continuous image frame in the video, so action recognition requires more computational power than processing static images. And the simple use of the CNN network cannot achieve the desired results. Recently, the attention model has achieved good results in computer vision and natural language processing. In particular, for video action classification, after adding the attention model, it is more effective to focus on motion features and improve performance. It intuitively explains which part the model attends to when making a particular decision, which is very helpful in real applications. In this paper, we proposed a 3D dense convolutional network based on attention mechanism(ADD-Net), recognition of human motion behavior in the video.