• Journal of Korea Multimedia Society
• /
• v.6 no.6
• /
• pp.1000-1007
• /
• 2003

In this paper, we propose an efficient method for locating lip. The lip shape is represented as a set of points based on Point Distribution Model. We use the Isodata clustering algorithm to find clusters for all training data. For each cluster, a lip shape model is calculated using principle component analysis. For all training data, a lip boundary model is calculated based on the pixel values around the lip boundary. To decide whether a recognition result is correct, we use a cost function based on the lip boundary model. Because of using different models according to the lip shapes, our method can localize correctly the flu far from the mean shape. The experiments have been performed for many images, and show correct recognition rate of 92%.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.19 no.4
• /
• pp.478-484
• /
• 2009

Automatic speech recognition(ASR) is attractive technique in trend these day that seek convenient life. Although many approaches have been proposed for ASR but the performance is still not good in noisy environment. Now-a-days in the state of art in speech recognition, ASR uses not only the audio information but also the visual information. In this paper, We present a novel lip detection method for visual speech recognition in mobile environment. In order to apply visual information to speech recognition, we need to extract exact lip regions. Because eye-detection is more easy than lip-detection, we firstly detect positions of left and right eyes, then locate lip region roughly. After that we apply K-means clustering technique to devide that region into groups, than two lip corners and lip center are detected by choosing biggest one among clustered groups. Finally, we have shown the effectiveness of the proposed method through the experiments based on samsung AVSR database.

• Journal of Korea Multimedia Society
• /
• v.7 no.11
• /
• pp.1531-1539
• /
• 2004

In this paper, we propose an efficient method for locating human lips. Based on Point Distribution Model and Principle Component Analysis, a lip shape model is built. Lip boundary model is represented based on the concatenated gray level distribution model. We calculate the distribution of shape parameters using Gaussian mixture. The problem to locate lip is simplified as the minimization problem of matching object function. The Down Hill Simplex Algorithm is used for the minimization with Gaussian Mixture for setting initial condition and refining estimate of lip shape parameter, which can refrain iteration from converging to local minima. The experiments have been performed for many images, and show very encouraging result.

• The KIPS Transactions:PartC
• /
• v.16C no.1
• /
• pp.91-100
• /
• 2009

Given the increased interest in ubiquitous computing, wireless sensor network has been researched widely. The localization service which provides the location information of mobile user, is one of important service provided by sensor network. Many methods to obtain the location information of mobile user have been proposed. However, these methods were developed for only one mobile user so that it is hard to extend for multiple mobile users. If multiple mobile users start the localization process concurrently, there could be interference of beacon or ultrasound that each mobile user transmits. In the paper, we propose IAL, the localization technique with interference avoidance for multiple mobile nodes in mobile wireless sensor networks. In IAL, interference is prevented by forcing the mobile node to get the permission of localization from anchor nodes. For this, we define LIP packet type for localization initiation by mobile node and LGP packet type for localization grant by anchor node. LRP packet type is used to reject localization by anchor node for interference avoidance. The experimental result shows that the number of interference between nodes are increased in proportion to the number of mobile nodes and IAL provides efficient localization.

• Korean Journal of Veterinary Research
• /
• v.33 no.4
• /
• pp.617-622
• /
• 1993

The preliminary studies on the localization of adrenoceptors were performed on smooth muscle strips of bovine esophageal groove. The mechanical activity of the muscle strip was recorded isometrically in vitro.w In the bottom circular muscle strips. the excitatory ${\alpha}-adrenergic$ responses were not blocked by tetrodotoxin$(2.1{\times}10^{-6}M)$ and denervation which was carried by cold storage of strips for 48 hrs in Tyrode's solution at $5-6{^{\circ}C}$ without oxygen supply. These excitatory ${\alpha}-adrenergic$ responses were partially blocked by atropine. In the lip longitudinal muscle strips, the inhibitory${\beta}-adrenergic$ responses were not blocked by pretreatment of tetrodotoxin and atropine. The results suggest that ${\beta}-adrenergic$ receptors mediating relaxations are located on the postsynaptic smooth muscle cells, whereas ${\beta}-adrenergic$ receptors mediating contractions are located both in the smooth muscle cells and in the cholinergic neurones.

• Korean Journal of Cleft Lip And Palate
• /
• v.6 no.2
• /
• pp.91-105
• /
• 2003

There are significant limitations in the precision of mandibular distraction in setting a desired occlusal and facial esthetic outcome. The purpose of this study is to present the simulation method for the distraction osteogenesis treatment planning. 3-D surgery simulation software programs V-works and V-Surgery(Cybermed, Seoul, Korea) were used from the 3D CT data in addition to the conventional data facial photography, panorama and cephalogram, dental cast model. We have utilized already for the various surgical procedures to get information preoperatively for the maxillofacial surgery like cancer localization and reconstructive surgery, orthognathic surgery and implant surgery in the department of Oral and Maxillofacial surgery, Seoul National University Hospital. On the software, bone cutting can be done at any place and any direction. Separated bone segment can be mobilized in all 3 dimensional direction. After the 3D simulation on the software program, mock surgery on the RP model can be performed. This planning method was applied to two hemifacial microsomia patients. With this protocol, we could simulate the movement of bony segment after maxillofacial distraction osteogenesis

• Horticultural Science & Technology
• /
• v.31 no.5
• /
• pp.590-597
• /
• 2013

Principle component and emission localization of volatile compounds were investigated according to scent intensity of rose flower. Scent intensity in cultivars and bred-line of Rosa hybrida was divided into three levels; light ('Feel Lip', 'Venus Berry'), medium ('GR07-135'), strong ('Honey Blue'). The major volatile compounds were different depending on the cultivars and selected line; 3,5-dimethoxytoluene (DMT), benzene, 1,3,5-trimethoxy ('Feel Lip'), megastigma-4,6(Z),8(E)-triene ('Venus Berry'), DMT, benzene,1-ethenyl-4-methoxyand, phenylethylalcohol ('GR07-135') and germacrene-D, DMT ('Honey Blue'). The adaxial epidermal cells were conical papillate shape, whereas the abaxial epidermal cells were flat shape. The adaxial epidermal cells of 3 cultivars and 1 selected line were surrounded by thick cell wall and covered by waxy cuticle of 2 cultivars and 1 selected line (except 'Honey Blue'). The adaxial epidermal cells contained starches in 'Feel Lip', osmiophlic droplets in 'Venus Berry', starchs, plastids, vacuoles in 'GR07-135' and plastoglobules, plastids, vacuoles in 'Honey Blue'. Based on these results, it appears that plastids and vacuoles in adaxial epidermal cells with conical papillate shape are associated production and emission of volatile compounds in scent R. hybrida.