• Journal of Korea Multimedia Society
• /
• v.21 no.2
• /
• pp.77-86
• /
• 2018

Although separation of touching pigs in real-time is an important issue for a 24-h pig monitoring system, it is challenging to separate accurately the touching pigs in a crowded pig room. In this study, we propose a separation method for touching pigs using the information generated from Convolutional Neural Network(CNN). Especially, we apply one of the CNN-based object detection methods(i.e., You Look Only Once, YOLO) to solve the touching objects separation problem in an active manner. First, we evaluate and select the bounding boxes generated from YOLO, and then separate touching pigs by analyzing the relations between the selected bounding boxes. Our experimental results show that the proposed method is more effective than widely-used methods for separating touching pigs, in terms of both accuracy and execution time.

• Journal of Korea Multimedia Society
• /
• v.19 no.8
• /
• pp.1361-1370
• /
• 2016

To reduce huge losses in pig farms, weaning pigs with weak immune systems are required to be carefully supervised. Even if various researches have been performed for pig monitoring environment, segmenting each pig from touching-pigs is still entrenched as a difficult problem. In this paper, we propose a segmentation method for touching-pigs by using concave-points and edge information in a video surveillance system. Especially, we interpret the segmentation problem as a time-series analysis problem in order to identify the concave-points generated by touching-pigs. Based on the experimental results with the videos obtained from a domestic pig farm, we believe that the proposed method can accurately segment the touching-pigs.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.4 no.3
• /
• pp.248-251
• /
• 2003

In this paper, a new segmentation method of machine printed character string with arbitrary length is proposed. Character recognition requires character segmentation as a previous step. However character segmentation itself requires a character recognition capability for less error segmentation. It is necessary to attack both these problem simultaneously. It is proposed that a new recognition-based segmentation method, which recognizes a character in touching characters with help of defined side-profiles. The match of ‘side-profiles of touching characters' with ‘side-profiles of prototypes' gives single character candidates in touching characters. It segments touching characters according to cutting costs.

• Child Health Nursing Research
• /
• v.13 no.4
• /
• pp.486-494
• /
• 2007

Purpose: This study was executed to investigate the route of the sexual knowledge acquisition and the sexual problem behavior of primary school students in a region. Method: The subjects of this research who were primary school 5, 6 grade students of 990 people were chosen at random in big cities, small towns and rural areas in Jeonbuk province. Result: The route of the sexual knowledge acquisition was through taking lesson & teacher in the school(40.6%), the peer group or senior member(26.1%), internet(15.3%), parents(10.0%), and multimedia(8.0%). The domain of the sexual problem behavior showed sexual impulse 3.8%, sexual violence 3.7%, masturbation 2.7%, voyeur 1.8%, prostitution 0.8%, transvestism 0.5%, exhibitionism 0.1%. The sexual problem behavior experienced by 5.0% or more of the subjects showed the experience of another person's compulsory touching on their body, the embrace desire and the kiss desire with opposite sexual friend. And the biggest sexual problem behavior of the subjects was sexual violence and sexual impulse to opposite sexual friend. The experience rate of the 6 grade was higher than that of the 5 grade (p<.05) and the male students' experience rate of another person's compulsory touching on their body showed higher than that of the female students(p<.05). Conclusion: The author through these findings suggests that systematic programs and curriculum should be developed for sexual problem behavior prevention.

• Journal of the Korean Institute of Landscape Architecture
• /
• v.19 no.2
• /
• pp.32-40
• /
• 1991

Damage to natural resources from inappropriate visitor behavior is a problem faced by National Park management. Based on the data gathered by personal interview and mail questionnaire, this study examines the effects of communication strategies to reduce depreciative behavior in Carlsbad Caverns National Park. One-quarter of Carlsbad Caverns visitors indicates that they have noticed damage to cave formation during their tour. Almost 38% of visitors say that they have witnessed someone touch a formation inside the Caverns. This is a relatively high percentage compared with other depreciative behavior research to refrain from "tattling" on fellow visitors. Two-thirds of the respondents recalls touching formation replicas before entering the Cavern, while almost all visitors (97.9%) say that they have heard the ranger's talk about not touching formation or leaving the trail. Visitors feel that the ranger's talk is more effective in reducing depreciative behavior than the formation replicas, although at least three-fourths of all respondents believes both techniques are effective.

• Journal of the Korean Society for Precision Engineering
• /
• v.16 no.3 s.96
• /
• pp.164-173
• /
• 1999

Robot calibration needs accurate measurements of robot end-effector position at a number of different robot configurations. One of the efficient ways of the measurement is "Touching on Jig" method suggested in [7], which utilizes a touch sensor and a fixture consisting of various sizes of blocks. By moving the end-effector to touch the surface of a block whose position relative to the other is known, the end-effector position relative to the fixture coordinate system can be obtained at the instant of touching. However, the global size of fixture is too small to cover the various configurations of the robot. Because of the manufacturing difficulties, the fixture cannot be manufactured large enough for well distributed position measurement. It results in the improvement of robot accuracy only in the limited space near to the fixture rather than over the whole space of the robot working volume. The paper proposes a method to resolve the above problem by measuring the end-effector positions with respect to several different coordinate system using the same measurement devices. It is found that the proposed method leads the improvements of robot position accuracy over the large space of working volume. The experimental studies are performed to show the validity of the method and their results are discussed.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2003.09a
• /
• pp.280-283
• /
• 2003

A purpose of our research is an acquisition of cooperative behaviors in inhomogeneous multi-agent system. In this research, we used the fire panic problem as an experiment environment. In Fire panic problem a fire exists in the environment, and follows in each steps of agent's behavior, and this fire spreads within the constant law. The purpose of the agent is to reach the goal established without touching the fire, which exists in the environment. The fire heat up by a few steps, which exists in the environment. The fire has unsureness to the agent. The agent has to avoid a fire, which is spreading in environment. The acquisition of the behavior to reach it to the goal is required. In this paper, we observe how agents escape from the fire cooperating with other agents. For this problem, we propose a unique CMAC based Q-learning system for inhomogeneous multi-agent system.

• KIPS Transactions on Computer and Communication Systems
• /
• v.5 no.10
• /
• pp.319-326
• /
• 2016

Abnormal situation caused by aggressive behavior of pigs adversely affects the growth of pigs, and comes with an economic loss in intensive pigsties. Therefore, IT-based video surveillance system is needed to monitor the abnormal situations in pigsty continuously in order to minimize the economic demage. In this paper, we propose a new Kinect camera-based monitoring system for the detection of the individual pigs. The proposed system is characterized as follows. 1) The background subtraction method and depth-threshold are used to detect only standing-pigs in the Kinect-depth image. 2) The moving-pigs are labeled as regions of interest. 3) A contour method is proposed and applied to solve the touching-pigs problem in the Kinect-depth image. The experimental results with the depth videos obtained from a pig farm located in Sejong illustrate the efficiency of the proposed method.

• 제어로봇시스템학회:학술대회논문집
• /
• 1996.10b
• /
• pp.1157-1160
• /
• 1996

Control of a biped robot which has compliant ankle joints is dealt in this paper. Simulated version of a human ankle joint is built using springs and mechanical constraints, which gives a flexibility of joint and compliance against the touching ground. The biped robot with compliant ankle joints proposed here gives a good contact between its sole and the ground and makes foot landing soft. As a result, installing force sensors for measuring the center of gravity of the biped becomes easier. A motor to drive an ankle joint is not needed which makes legs light. However, the control problem becomes more difficult because the torque of the ankle joint to put the biped in a desired walking gait cannot be provided from the compliant ankle joint. To solve this problem, we proposed a dynamic gait modification method by adjusting the position of a hip joint. Simulation results for the mathematical model of the SD-2 biped in the Ohio State University are given to show the validity of the proposed controller.

• The Journal of Korean Academy of Prosthodontics
• /
• v.42 no.5
• /
• pp.524-534
• /
• 2004

Statement of problem: The meaning of the beauty has a little different nature according to a time, culture, and nation. Purpose : This study was undertaken to estimate the geometric esthetic criteria for Korean woman by analysis of facial photograph in a smiling. Material and methods: The facial photographs of 678 Korean beautiful woman were collected from Miss Korea candidates(from 1971 to 2003 year). The dento-facial composition was measured and analyzed geometrically on computer monitor using $Adobe^{(R)}$ $Photoshop^{(R)}$ 7.0. The statistical significance of the differences among the 4 decade groups was estimated by a $x^2$ test. Result : The Korean beautiful woman skewed the geometric characteristics such as the 'coincidence' (83.0%) between facial midline and dental midline, the 'normal' (96.9%) alignment and arrangement of upper anterior teeth. the 'average' (79.1%) position and 'upward' (53.4%) curvature of upper lip, the 'parallel' (85.7%) relationship between the lower lip and upper anterior incisal curvature, the 'touching' (49.9%) and 'not-touching' (44.8%) relationship between upper anterior incisor and towel lip, the 'first premolar' (54.1%) and 'second premolar' (38.9%) displayed in a smile, and the 'symmetry' (89.4%) of smile. There was statistical significance among the 4 decades in the respect of the upper lip curvature, the relationship between upper anterior incisor and lower lip, and the most posterior teeth displayed in a smile (p<0.05). Conclusion : The Korean beautiful woman showed similar results to the western esthetic criteria in the some respects and the tendency that the Korean woman had more active smiling recently.