• 제어로봇시스템학회:학술대회논문집
• /
• 2005.06a
• /
• pp.2337-2341
• /
• 2005

This paper presents the target detection method using Support Vector Machines(SVMs) and the navigation system using behavior-based fuzzy controller. SVM is a machine-learning method based on the principle of structural risk minimization, which performs well when applied to data outside the training set. We formulate detection of target objects as a supervised-learning problem and apply SVM to detect at each location in the image whether a target object is present or not. The behavior-based fuzzy controller is implemented as an individual priority behavior: the highest level behavior is target-seeking, the middle level behavior is obstacle-avoidance, the lowest level is an emergency behavior. We have implemented and tested the proposed method in our mobile robot "Pioneer2-AT". Comparing with a neural-network based detection method, a SVM illustrate the excellence of the proposed method.

• The Transactions of the Korean Institute of Electrical Engineers D
• /
• v.51 no.9
• /
• pp.402-412
• /
• 2002

The navigation algorithms enable autonomous mobile robots to reach given target points without collision against obstacles. To achieve safe navigations in unknown environments, this paper presents an effective navigation algorithm for the autonomous mobile robots with ultrasonic sensors. The proposed navigation algorithm consists of an obstacle-avoidance behavior, a target-reaching behavior and a fuzzy-based decision maker. In the obstacle-avoidance behavior and the target-reaching behavior, artificial immune networks are used to select a proper steering angle, make the autonomous mobile robot avoid obstacles and approach a given target point. The fuzzy-based decision maker combines the steering angles from the target-reaching behavior and the obstacle-avoidance behavior in order to steer the autonomous mobile robot appropriately. Simulational and experimental results show that the proposed navigation algorithm is very effective in unknown environments.

• International Journal of Internet, Broadcasting and Communication
• /
• v.16 no.3
• /
• pp.305-312
• /
• 2024

This field study explores how varying target amounts influence donation behavior using real-world data from the online fundraising platform GoFundMe. We analyzed donation data across four different target amounts and found significant differences in donation patterns. Lower target amounts were found to encourage higher individual donations, while excessively high targets were less effective. The data revealed that donors tend to be more responsive to campaigns with achievable goals, possibly due to a perceived higher impact of their contributions. Conversely, campaigns with unrealistically high targets often struggled to gain traction, suggesting a potential donor deterrent effect. We believe these findings provide practical insights for nonprofits on setting realistic and achievable target amounts to maximize donations. Our study underscores the importance of strategic target setting in enhancing fundraising outcomes. We conclude that this insight has significant implications for how non-profit organizations approach their fundraising strategies, potentially improving the effectiveness of online charitable campaigns.

• Journal of Korea Multimedia Society
• /
• v.25 no.6
• /
• pp.807-818
• /
• 2022

In order to further improve the accuracy and time efficiency of behavior recognition in intelligent monitoring scenarios, a human behavior recognition algorithm based on YOLO combined with LSTM and CNN is proposed. Using the real-time nature of YOLO target detection, firstly, the specific behavior in the surveillance video is detected in real time, and the depth feature extraction is performed after obtaining the target size, location and other information; Then, remove noise data from irrelevant areas in the image; Finally, combined with LSTM modeling and processing time series, the final behavior discrimination is made for the behavior action sequence in the surveillance video. Experiments in the MSR and KTH datasets show that the average recognition rate of each behavior reaches 98.42% and 96.6%, and the average recognition speed reaches 210ms and 220ms. The method in this paper has a good effect on the intelligence behavior recognition.

• 제어로봇시스템학회:학술대회논문집
• /
• 2001.10a
• /
• pp.134.6-134
• /
• 2001

The purpose of navigation algorithm is to reach a given target point without collision with obstacles while an autonomous mobile robot is navigating. To achieve a safe navigation, this paper presents an effective navigation algorithm for the autonomous mobile robot equipped with ultrasonic sensors in unknown environments. The proposed navigation algorithm consists of an obstacle-avoidance behavior, a target-reaching behavior and a fuzzy-based decision maker. In the obstacle-avoidance behavior and the target-reaching behavior, artificial immune networks are used to select a proper steering angle, make the autonomous mobile robot avoid obstacles and approach a given target point. The decision maker using fuzzy inference systems weights the steering angles selected ...

• The Research Journal of the Costume Culture
• /
• v.12 no.1
• /
• pp.182-198
• /
• 2004

The purpose of this study was to analyze the previous make-up studies. A number of publications and journals were reviewed and analyzed carefully. The results of review and analysis were as follows: There were many different subjects in make-up studies and They can be divided into ten types : cosmetics purchase behavior, change of make-up culture and comparison, make-up trend by era, cosmetics industry's standing of today and strategy, art trend in make-up, brand preference of cosmetics, make up attitude, recognition about imported cosmetics and purchase behavior, color preference of cosmetics, the relationship between self-concept and make-up. In general, the cosmetic purchase behavior studies are conducted most actively. According to result that analyze existent study, special duality of cosmetics purchase action appears very variously according to standard of classification of study target and study target. But, study target and method of study are not various, and purchase behavior study collected with make-un and clothes is yew lacking. Therefore, in this study, wished to discover problem of virtue study because analyzes studies about previous make-up and present forward study direction.

• The Journal of Information Technology
• /
• v.3 no.3
• /
• pp.67-76
• /
• 2000

Internet user are very important target market in off-line. Also as being activated e-business, they become core target market in on-line. It comes the present question to the business that marketer has to develop the most suitable marketing method by analysis of consumer's purchasing behavior In this study, we analysis internet user's purchasing behavior and consider the way which is the most suitable internet advertisement using an analysis of census.

• Journal of Korea Multimedia Society
• /
• v.24 no.5
• /
• pp.659-666
• /
• 2021

With the continuous development of deep learning, human behavior recognition algorithms have achieved good results. However, in a multi-person recognition environment, the complex behavior environment poses a great challenge to the efficiency of recognition. To this end, this paper proposes a multi-person pose estimation model. First of all, the human detectors in the top-down framework mostly use the two-stage target detection model, which runs slow down. The single-stage YOLOv3 target detection model is used to effectively improve the running speed and the generalization of the model. Depth separable convolution, which further improves the speed of target detection and improves the model's ability to extract target proposed regions; Secondly, based on the feature pyramid network combined with context semantic information in the pose estimation model, the OHEM algorithm is used to solve difficult key point detection problems, and the accuracy of multi-person pose estimation is improved; Finally, the Euclidean distance is used to calculate the spatial distance between key points, to determine the similarity of postures in the frame, and to eliminate redundant postures.

• Earthquakes and Structures
• /
• v.23 no.2
• /
• pp.115-128
• /
• 2022

One common method to select input ground motions to predict dynamic behavior of structures subjected to seismic excitation requires spectral acceleration (S_a) match target mean response spectrum. However, dispersion of ground motions, which explicitly affects the structural response, is rarely discussed in this method. Generally, selecting ground motions matching target mean and variance has been utilized as an appropriate method to predict reliable seismic response. The goal of this paper is to investigate the impact of target spectra variance of ground motions on structural seismic response. Two sets of ground motions with different target variances (zero variance and minimum variance larger than inherent variance of the target spectrum) are selected as input to two different structures. Structural responses at different heights are compared, in terms of peak, mean and dispersion. Results show that increase of target spectra variance tends to increase peak floor acceleration, peak deformation and dispersions of response of interest remarkably. To short-period structures, dispersion increase ratios of seismic response are close to that of S_a of input ground motions at the first period. To long-period structures, dispersions of floor acceleration and floor response spectra increase more significantly at the bottom, while dispersion increase ratios of IDR and deformation are close to that of S_a of input ground motions at the first period. This study could further provide useful information on selecting appropriate ground motion to predict seismic behavior of different types of structures.

• Journal of the Korean Society of Fisheries and Ocean Technology
• /
• v.51 no.2
• /
• pp.272-278
• /
• 2015

The aim of this study is to find out the behavior and acoustic backscattering of the large jellyfish Nemopliema nomurai using hydroacoustics in situ. N. nomurai was distributed at depths ranging from 10~15 m during the day. Regarding the behavior of N. nomurai, there was no significant change in depth, and 3D tortuosity was not high. The vertical direction was ${\pm}10^{\circ}$ from the horizontal, and moving speed was $0.9{\sim}1.5\;m\;s^{-1}$. With regard to hydro-acoustical characteristics, the mean TS of N. nomurai ranged from -69.6~-56.0 dB at 38 kHz and -69.4~-54.5 dB at 120 kHz. TS variation (Max TS-Min TS) at 38 and 120 kHz was 0~10.2 dB and 0.2~16.0 dB, respectively. Mean TS and TS variation (Max TS-Min TS) of N. nomurai were higher at 120 kHz than at 38 kHz. The results showed that the use of hydroacoustics was effective in estimating the distribution depth, behavior, and acoustic characteristics of the target.