• Journal of Auto-vehicle Safety Association
• /
• 제12권4호
• /
• pp.23-30
• /
• 2020

Recently, self-driving research has been actively studied in various institutions. Accurate recognition is important because information about surrounding objects is needed for safe autonomous driving. This study mainly deals with the signal processing of LiDAR among sensors for object recognition. LiDAR is a sensor that is widely used for high recognition accuracy. First, we clustered and tracked objects by predicting relative position and speed of objects. The characteristic points of all objects were extracted using point cloud data of each objects through proposed algorithm. The Classification between vehicle and pedestrians is estimated using number of characteristic points and distances among characteristic points. The algorithm for classifying cars and pedestrians was implemented and verified using test vehicle equipped with LiDAR sensors. The accuracy of proposed object classification algorithm was about 97%. The classification accuracy was improved by about 13.5% compared with deep learning based algorithm.

• Journal of Korean Clinical Health Science
• /
• 제5권4호
• /
• pp.1015-1020
• /
• 2017

Purpose. Walking depends on the speed and type of shoe to be worn, and the degree of impact varies with the muscle used. In addition, the speed can be changed by moving objects and using objects when walking. This study analyzed the change of walking speed by applying various factors influencing walking. Methods. A total of 60 patients who had not undergone musculoskeletal diseases during the last 1 year were included. Shoe type was divided into slippers and shoe heels. Behavioral types were divided into bagging, books, and cell phone use. The walking speed was measured by the general walking, the middle walking, and the fast walking. The time was measured using a 10M linear distance test. The collected data were analyzed with SPSS program for independent samples t-test, one-way ANOVA. Results. There was a statistically significant difference according to the type of shoes when walking. Walking speed was slow in shoe heel. In addition, There was statistically significant difference according to type of behavior task at walking. Walking speed was slow in task type using mobile phone during walking. Conclusions. The walking speed were appeared difference in each type of shoe heel, using mobile phone.

• Journal of Korea Multimedia Society
• /
• 제7권11호
• /
• pp.1550-1555
• /
• 2004

Shadows increase the visual realism of computer-generated images and they are good hint for spatial relationships between objects. Previous methods to produce a shadow texture for an object are to render all objects between the object and light source. Consequently entire time for generating shadow textures between all objects is Ο(Ν$^2$), where Ν is the number of objects. We propose a novel shadow texture generation method with constant processing time for each object using shadow depth buffet. In addition, we also present method to achieve further speed-up using temporal coherence. If the transition between dynamic and static state is not frequent, depth values of static objects does not vary significantly. So we can reuse the depth value for static objects and render only dynamic objects.

• IEMEK Journal of Embedded Systems and Applications
• /
• 제15권5호
• /
• pp.215-225
• /
• 2020

Light detection and ranging (LiDAR) sensors detect the distance of the surrounding environment and objects. Conventional LiDAR sensors require a certain amount of a power because they detect objects by transmitting lasers at a regular interval depending on a constant resolution. The constant power consumption from operating multiple LiDAR sensors is detrimental to autonomous and electric vehicles using battery power. In this paper, we propose two algorithms that improve the inefficient power consumption during the constant operation of LiDAR sensors. LiDAR sensors with algorithms efficiently reduce the power consumption in two ways: (a) controlling the resolution to vary the laser transmission period (T_P) of a laser diode (LD) depending on the vehicle's speed and (b) reducing the static power consumption using a sleep mode depending on the surrounding environment. A proposed LiDAR sensor with a resolution control algorithm reduces the power consumption of the LD by 6.92% to 32.43% depending on the vehicle's speed, compared to the maximum number of laser transmissions (N_x·max). The sleep mode with a surrounding environment-sensing algorithm reduces the power consumption by 61.09%. The proposed LiDAR sensor has a risk factor for 4-cycles that does not detect objects in the sleep mode, but we consider it to be negligible because it immediately switches to an active mode when a change in surrounding conditions occurs. The proposed LiDAR sensor was tested on a commercial processor chip with the algorithm controlling the resolution according to the vehicle's speed and the surrounding environment.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 한국방송공학회 2009년도 IWAIT
• /
• pp.71-75
• /
• 2009

In this paper, we present a real-time method to detect moving objects in a rotating and zooming camera. It is useful for camera surveillance of fixed but rotating camera, camera on moving car, and so on. We first compensate the global motion, and then exploit the displaced frame difference (DFD) to find the block-wise boundary. For robust detection, we propose a kind of image to combine the detections from consecutive frames. We use the block-wise detection to achieve the real-time speed, except the pixel-wise DFD. In addition, a fast block-matching algorithm is proposed to obtain local motions and then global affine motion. In the experimental results, we demonstrate that our proposed algorithm can handle the real-time detection of common object, small object, multiple objects, the objects in low-contrast environment, and the object in zooming camera.

• Journal of Biosystems Engineering
• /
• 제24권2호
• /
• pp.145-152
• /
• 1999

Methods of discharging each graded agricultural product are divided into two according to the type of feeding. One is based on feeding objects using a series of specially designed holders mounted with an equal interval. The other is randomly feeding objects while being isolated without a specific interval. In this paper, a real time simultaneous discharge algorithm, which is good for objects randomly fed individually and for multi-objects located along a series of discharge buckets, was developed. And the developed algorithm was implemented to the controller and the performance was verified using the system developed for dried mushrooms. The discharge system used for the experiment was composed of a variable speed conveyor, a series of double channel bucket mounted along both sides of the conveyor, and a series of air nozzles and optic sensors. Developed algorithm worked perfectly and could be directly used for automatic discharge system for randomly feeding agricultural products.

• Journal of the Korea Safety Management & Science
• /
• 제4권2호
• /
• pp.155-168
• /
• 2002

As the market and the organizations of an enterprise expand globally, the rapid and accurate communication gets more important for product manufacturing. The manufacturing information flow among the designers, the process planners and the shop floors is characterized and modelled. Its methods are constructed using distributed objects. Their introduction to the network-based CAPP system offers speed and safety for the system and makes the reconstruction and distribution of the application programs easy. The manufacturing processes are generated based on the feature information of a designed part, then the manufacturing resources are selected using the process planning logic which is implemented by distributed objects. The databases and distributed objects are integrated under the recent internet environments. The developed system makes it possible to manipulate and to transfer the process planning and manufacturing data everywhere in the world.

• Proceedings of the KIEE Conference
• /
• 대한전기학회 2005년도 학술대회 논문집 정보 및 제어부문
• /
• pp.346-348
• /
• 2005

In this paper, we designed a vehicle active suspension controller. Vehicle suspensions have various design objects with tradeoff among them and these objects cannot be satisfied under all driving conditions. We need to design a controller adapted to variable driving conditions changing the objects of vehicle suspensions. To design such a controller, we must be able to detect the current driving conditions and focus on the road frequencies giving us useful and important information about driving conditions. Detecting the road frequencies, we use the Fourier Transform. A unexpected driving change like a speed bump was also included to items the new designed controller must consider.

• Journal of Mechanical Science and Technology
• /
• 제16권4호
• /
• pp.427-435
• /
• 2002

This paper presents a development result for image processed tracking system of multiple moving objects based on Kalman filter and a simple window tracking method. The proposed algorithm of foreground detection and background adaptation (FDBA) is composed of three modules: a block checking module(BCM), an object movement prediction module(OMPM), and an adaptive background estimation module (ABEM). The BCM is processed for checking the existence of objects. To speed up the image processing time and to precisely track multiple objects under the object's mergence, a concept of a simple window tracking method is adopted in the OMPM. The ABEM separates the foreground from the background in the reset simple tracking window in the OMPM. It is shown through experimental results that the proposed FDBA algorithm is robustly adaptable to the background variation in a short processing time. Furthermore, it is shown that the proposed method can solve the problems of mergence, cross and split that are brought up in the case of tracking multiple moving objects.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• 제27권5호
• /
• pp.697-704
• /
• 2003

In this paper, the visual tracking system for arbitrary shaped moving object is proposed. The established tracking system can be divided into model based method that needs previous model for target object and image based method that uses image feature. In the model based method, the reliable tracking is possible, but simplification of the shape is necessary and the application is restricted to definite target mod el. On the other hand, in the image based method, the process speed can be increased, but the shape information is lost and the tracking system is sensitive to image noise. The proposed tracking system is composed of the extraction process that recognizes the existence of moving object and tracking process that extracts dynamic characteristics and shape information of the target objects. Specially, active contour model is used to effectively track the object that is undergoing shape change. In initializatio n process of the contour model, the semi-automatic operation can be avoided and the convergence speed of the contour can be increased by the proposed effective initialization method. Also, for the efficient solution of the correspondence problem in multiple objects tracking, the variation function that uses the variation of position structure in image frame and snake energy level is proposed. In order to verify the validity and effectiveness of the proposed tracking system, real time tracking experiment for multiple moving objects is implemented.