• 전자공학회논문지 IE
• /
• v.45 no.4
• /
• pp.60-67
• /
• 2008

Skin color is a very important information for an automatic face recognition. In this paper, we proposed a skin region extraction method using color information and skin color model. We use the adaptive lighting compensation technique for improved performance of skin region extraction. Also, using an preprocessing filter, normally large areas of easily distinct non skin pixels, are eliminated from further processing. And we use the modified ST color space, where undesired effects are reduced and the skin color distribution fits better than others color space. Experimental results show that the proposed method has better performance than the conventional methods, and reduces processing time by $35{\sim}40%$ on average.

• Journal of Biosystems Engineering
• /
• v.30 no.5 s.112
• /
• pp.299-305
• /
• 2005

This study was developed image processing algorithm for recognition of few teats of a cow in the image using black and white camera attached with infrared filter. Spectroscopic analysis was used for selection of a NIR filter to separate teats from udder skin in the image captured. To verify the performance of image processing algorithm was developed and NIR filter was selected, carried out an experiment with cows. NIR band-pass filter was used to pass the 975nm band of light spectrum. The image processing algorithm was developed could recognize all teats and the process time was 0.9 second to recognize the all teats and to acquire end position of teats.

• Smart Media Journal
• /
• v.2 no.4
• /
• pp.9-19
• /
• 2013

This paper gives a review on three illumination issues of face tracking on smart phones: dark scenes, sudden lighting change and backlit effect. First, we propose a fast and robust face tracking method utilizing continuous adaptive mean shift algorithm (CAMSHIFT) and CbCr skin locus. Initially, the skin locus obtained from training video data. After that, a modified CAMSHIFT version based on the skin locus is accordingly provided. Second, we suggest an enhancement method to increase the chance of detecting faces, an important initialization step for face tracking, under dark illumination. The proposed method works comparably with traditional CAMSHIFT or particle filter, and outperforms these methods when dealing with our public video data with the three illumination issues mentioned above.

• Radiation Oncology Journal
• /
• v.6 no.1
• /
• pp.109-116
• /
• 1988

The Alcyon Co-60 gamma rays was studied for electron contamination. The surface dose, attributable almost entirely to contamination electrons, has a linear dependence on field width for square fields and an inverse square dependence on distance from the bottom of the fixed head assembly Build-up and surface dose measurements were taken with and without an acrylic blocking tray in place. Further measurements were made with a copper filter designed to reduce secondary electrons emitted by photon interactions with the acrylic tray. The results are discussed in relation to skin sparing effect for radiation therapy Patients. And to achieve the maximum skin sparing effect, the selection of the optimum SSD and TSD is needed.

• Journal of the Institute of Convergence Signal Processing
• /
• v.15 no.3
• /
• pp.76-79
• /
• 2014

Face recognition is very challenging because of the unexpected changes of pose, expression, and illumination. Facial detection in the mobile environments has additional difficulty since the computational resources are very limited. This paper discusses face detection based on frequency domain matched filtering in the mobile environments. Face detection is performed by a linear or phase-only matched filter and sequential verification stages. The candidate window regions are selected by a number of peaks of the matched filtering outputs. The sequential stages comprise a skin-color test and an edge mask filtering tests, which aim to remove false alarms among selected candidate windows. The algorithms are built with JAVA language on the mobile device operated by the Android platform. The simulation and experimental results show that real-time face detection can be performed successfully in the mobile environments.

• Magazine of the Korean Society of Agricultural Engineers
• /
• v.28 no.2
• /
• pp.63-73
• /
• 1986

The model was developed by applying the principles of Bacot and Vidal to measure the behavior of deformation of the reinforced earth wall, and various tasts were performed by using the plastic fabric filter and the galvanized steel plate as a strip. The results obtained are as follows; 1. When the reinforced earth wall is deformed by the load, the strip is completely reinforced by the backfill materials and changed to the rigid block state, under the state of failure which permits sliding only, the next theoretical equation is formed. (H/L) . tan$\theta$ [cosO-sinOtanO] =2sinO[tan($\theta$ +0) +tanO] 2.The degree of the mutual reinforcement of the backfill material and the strip depend on the physical characteristics of the each material especially the angle of shearing resistance of the backfill material is desirable over 20$^{\circ}$ and, if it is over 400, its function could be a maximum. 3.The distribution of the maximum tensile strain of the reinforcement is changing with the height of reinforced earth wall, and when the height from bottom of the reinforced earth wall is 1.85 to 3. 35m, the maximum tensile strain appears at 2m from the skin element. The maximum tensile strain is increased by the depth of the reinforced earth wall from surface, and increased with the lapse of time after construction. 4.The failure surface of the reinforced earth wall by the concrete skin was about 60$^{\circ}$and the failure behavior of the reinforced earth wall in which the fabric filter was buried was slow, and so the pore pressure could be decreased. 5.It is possible to construct the fabric retained earth wall by the plastic fabric filter only. And the reinforcing effect between the steel plate and the plastic fabric filter is not largely different. however, in the aspect of the economic durability, the plastic fabric filter is more advantageous. 6.The reinforcing action mainly depends on the width and the length of the reinforcing materials, if possible, the full width is advantageous to enlarge the contact area with backfill. but considering the economic aspect, it is neccessary to develop the method controlling the space of the strip.

• Korean Journal of Optics and Photonics
• /
• v.31 no.3
• /
• pp.148-154
• /
• 2020

In this paper we propose a plasmonic color filter with a novel nanopattern. The suggested pattern, called a "polyperiodic hole array" (PPHA), is introduced to solve the angle dependence of the optical response that originates from the periodic structure. We set the diameter and period of the hole to make a green color filter, and set the unit-cell size and metal and dielectric thicknesses in consideration of the propagation length and skin depth. The periodic hole arrays are locally rotated to make a PPHA pattern, resulting in a globally aperiodic yet partially periodic pattern. As a result, compared to a general pattern, the PPHA nanostructured color filter has a maximum 40% improvement in spectral shift when the angle of incidence is increased from 0° to 30°. Transmittance reduction was also alleviated by 30%. This work will improve the performance of nanostructured color filters and help with nanotechnology being applied industrially to imaging devices, including displays and image sensors.

• The Journal of Korean Society for Radiation Therapy
• /
• v.3 no.1
• /
• pp.85-89
• /
• 1989

Dose distribution was evaluated under vaseline and thin lead used as surface bolus, in case with scattering filter and without, for 9-MeV electron using chambers in water phantom. The results were as follows: 1. The skin dose can be remarkably increased with thin lead bolus than with convensional bolus. 2. The skin dose over $110\%$ in the 0.6mm thin lead bolus compared with the maximum dose in normal irradiation, so skin burn or any other complications may be occured in patients.

• Journal of radiological science and technology
• /
• v.31 no.3
• /
• pp.223-228
• /
• 2008

The most critical point in the medical use of radiation is to minimize the patient's entrance dose while maintaining the diagnostic function. Low-energy photons (long wave X-ray) among diagnostic X-rays are unnecessary because they are mostly absorbed and contribute the increase of patient's entrance dose. The most effective method to eliminate the low-energy photons is to use the filtering plate. The experiments were performed by observing the image quality. The skin entrance dose was 0.3 mmCu (copper) filter. A total of 80 images were prepared as two sets of 40 cuts. In the first set (of 40 cuts), 20 cuts were prepared for the non-filter set and another 20 cuts for the Cu filter of signal + noise image set. In the second set of 40 cuts, 20 cuts were prepared for the non-filter set and another 20 cuts for the Cu filter of non-signal image (noisy image) with random location of diameter 4 mm and 3 mm thickness of acryl disc for ROC signal at the chest phantom. P(S/s) and P(S/n) were calculated and the ROC curve was described in terms of sensitivity and specificity. Accuracy were evaluated after reading by five radiologists. The number of optically observable lesions was counted through ANSI chest phantom and contrast-detail phantom by recommendation of AAPM when non-filter or Cu filter was used, and the skin entrance dose was also measured for both conditions. As the result of the study, when the Cu filter was applied, favorable outcomes were observed on, the ROC Curve was located on the upper left area, sensitivity, accuracy and the number of CD phantom lesions were reasonable. Furthermore, if skin entrance dose was reduced, the use of additional filtration may be required to be considered in many other cases.

• Proceedings of the IEEK Conference
• /
• 2002.06c
• /
• pp.13-16
• /
• 2002

This paper proposes a method to track hand gesture and to recognize the gesture pattern using Kalman filter and continuous dynamic programming (CDP). The positions of hands are predicted by Kalman filter, and corresponding pixels to the hands are extracted by skin color filter. The center of gravity of the hands is the same as the input pattern vector. The input gesture is then recognized by matching with the reference gesture patterns using CDP. From experimental results to recognize circle shape gesture and intention gestures such as “Come on” and “Bye-bye”, we show the proposed method is feasible to the hand gesture-based human -computer interaction.