• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.2
• /
• pp.522-539
• /
• 2021

In order to solve the problems of the existing audio fingerprint method when extracting audio fingerprints from long speech segments, such as too large fingerprint dimension, poor robustness, and low retrieval accuracy and efficiency, a robust audio fingerprint retrieval method based on feature dimension reduction and feature combination is proposed. Firstly, the Mel-frequency cepstral coefficient (MFCC) and linear prediction cepstrum coefficient (LPCC) of the original speech are extracted respectively, and the MFCC feature matrix and LPCC feature matrix are combined. Secondly, the feature dimension reduction method based on information entropy is used for column dimension reduction, and the feature matrix after dimension reduction is used for row dimension reduction based on energy feature dimension reduction method. Finally, the audio fingerprint is constructed by using the feature combination matrix after dimension reduction. When speech's user retrieval, the normalized Hamming distance algorithm is used for matching retrieval. Experiment results show that the proposed method has smaller audio fingerprint dimension and better robustness for long speech segments, and has higher retrieval efficiency while maintaining a higher recall rate and precision rate.

• Korean Journal of Optics and Photonics
• /
• v.23 no.4
• /
• pp.167-171
• /
• 2012

This paper reports on the stray light analysis results of a compact imaging spectrometer (COMIS) for a microsatellite STSAT-3. COMIS images Earth's surface and atmosphere with ground sampling distances of 27 m at the 18~62 spectral bands (0.4 ~ 1.05 ${\mu}m$) for the nadir looking at an altitude of 700 km. COMIS has an imaging telescope and an imaging spectrometer box into which three electronics PCBs are embedded. The telescope images a $27m{\times}28km$ area of Earth surface onto a slit of dimensions $11.8{\mu}m{\times}12.1mm$. This corresponds to a ground sampling distance of 27 m and a swath width of 28 km for nadir looking posture at an altitude of 700 km. Then the optics relays and disperses the slit image onto the detector thereby producing a monochrome image of the entrance slit formed on each row of detector elements. The spectrum of each point in the row is imaged along a detector column. The optical mounts and housing structures are designed in order to prevent stray light from arriving onto the image and so deteriorating the signal to noise ratio (SNR). The stray light analysis, performed by a non-sequential ray tracing software (LightTools) with three dimensional housing and lens modeling, confirms that the ghost and stray light arriving at the detector plane has the relative intensity of ${\sim}10^{-5}$ and furthermore it locates outside the concerned image size i.e. the field of view of the optics.

• The korean journal of orthodontics
• /
• v.29 no.4 s.75
• /
• pp.425-433
• /
• 1999

Dr. Youn-Sic Chun and coworkers invented the new machine for getting information about the relative effectiveness of the orthodontic appliances and we named it Calorific machine. The author and colleagues used this machine to compare with tooth moving mechanism by 3 types of the insertion method of precision-TPA for derotating the posterior teeth. We measured the distance of tooth movement and found out the rotational center on the occlusal X-ray film and the tooth movement on the occlusogram and then processed paired t-test by SAS program, The results were as follows : 1. In the intermolar width, the mesial insertion method showed the loss oi distance, and the other methods(-distal insertion method, mesial expansion method-) showed the increase of distance. 2. In the arch length, the distal insertion method was exhibited as most recommandable way for increasing the arch length .3. The rotation center of the mesial insertion method for derotating the molar, was located between mesial pit and central pit of the lower End molar. And, in the distal insertion method, it was located between distal pit and distolingual cusp, and in the mesial expansion method was located at distal pit.

• Imaging Science in Dentistry
• /
• v.39 no.1
• /
• pp.27-33
• /
• 2009

Purpose: The purpose of this study was to evaluate the clinical usefulness of the recently developed multi-detector computed tomography and cone beam computed tomography in pre-operative implant evaluation, by comparing them with the single detector computed tomography, already confirmed for accuracy in this area. Materials and Methods: Five partially edentulous dry human mandibles, with $1{\times}1mm$ gutta percha cones, placed in 5mm intervals posterior to the mental foramen on each side of the buccal part of the mandible, were used in this study. They were scanned as follows: 1) Single detector computed tomography: slice thickness 1mm, 200mA, 120kV 2) Multi-detector computed tomography: slice thickness 0.75mm, 250mA, 120kV 3) Cone beam computed tomography: 15mAs, 120kV Axial images acquired from three computed tomographies were transferred to personal computer, and then reformatted cross-sectional images were generated using V-Implant $2.0^{(R)}$ (CyberMed Inc., Seoul, Korea) software. Among the cross-sectional images of the gutta perch a cone, placed in the buccal body of the mandible, the most precise cross section was selected as the measuring point and the distance from the most superior border of the mandibular canal to the alveolar crest was measured and analyzed 10 times by a dentist. Results: There were no significant intraobserver differences in the distance from the most superior border of the mandibular canal to the alveolar crest (p>0.05). There were no significant differences among single detector computed tomography, multi-detector computed tomography and cone beam computed tomography in the distance from the most superior border of the mandibular canal to the alveolar crest (p>0.05). Conclusion: Multi-detector computed tomography and cone beam computed tomography are clinically useful in the evaluation of pre-operative site for mandibular dental implants, with consideration for radiation exposure dose and scanning time.

• Journal of Korean Society of Steel Construction
• /
• v.26 no.1
• /
• pp.43-53
• /
• 2014

A single plate connection(SPC) consists of a plate welded to the columns and bolts connected to the beam web. The SPC is widely used for a simple shear connection of steel structure because it is easy-to-fabricated, easy-to-installed and economical. The conventional SPC is used for 2 to 12 bolts in a single vertical row. It is designed to limit the plate thickness by bolt diameter to obtain flexible and ductile connections. The design strength for eccentric shear shall be the lesser of the shear strength of bolts or bearing strength of plate and when the design strength is decided by edge distance failure, the results can be very conservative. Although the research on special solution for 'weak-plate/strong-bolt' model with 2 to 4 bolts has been conducted by L. S. Muir, and W. A. Thonton, 2004, study on generalized design procedures did not conduct. This study proposed design procedure for evaluation of the design strength of eccentric shear bolt groups on a single plate connection based on the actual edge distance and the direction of bolt reaction forces by using elastic vector method(EVM) and instantaneous center of rotation method(ICM).

• The korean journal of orthodontics
• /
• v.26 no.1 s.54
• /
• pp.17-32
• /
• 1996

It is important to understand the operating mechanism and force system of fixed appliance that most effective for individual tooth movement in various orthodontic appliances. The archwire system of fixed appliance is devided into 3 types, which is continuous arch, segmented arch and sectional arch. The last two types have longer interbracket distance and simple force operating points, so it is easy to control force system by operator. But the continuous arch has shorter interbracket distance and various bracket geometry, so it is hard to control and anaylze the force system. The purpose of this study was three dimentional force and moment analysis of continuous arch system by finite element method, which is similar situation to three dimentional elastic beam in structural engineering. Several sample form of various bracket geometry and artificial lower crowding typodont made by author were constructed, analyzed and compared each other. The results were as follows : 1. The force magnitude is linear proportional to the degree of displacement or tilting of the bracket. 2. The force magnitude is inversely non-linear proportional to the interbracket distance. 3. In three dimensional typodont model, while the force can be compared with that of the sample form in the area where adjacent bracket geometry is simple, the force is much more than the expected value in the area where adjacent bracket geometry is complex.

• Journal of Korean Society of Coastal and Ocean Engineers
• /
• v.21 no.1
• /
• pp.45-53
• /
• 2009

A possibility of the drag coefficient estimation in numerical water basins was discussed where the numerical solution were calculated by the 3-dimensional hydro-dynamical model (FLOW-$3D^{(R)}$) with the RNG $k-{\varepsilon}$ turbulence model. On the known cases of the drag coefficients for a rectangle, the numerical drag coefficients got $1.34{\sim}1.52$ and the wind tunnel values were $1.3{\sim}1.5$. For a cylinder, the numerical values were calculated as $0.75{\sim}0.78$ in the range of 0.5

• Korean Journal of Optics and Photonics
• /
• v.10 no.1
• /
• pp.21-26
• /
• 1999

The technique for calibrating standard particles larger than 1 $\mu$m by using an optical array sizing method has been established. The system consists of an optical microscope with transmission light source, a CCD camera, and a computer equipped with a frame grabber. The center distance between two spheres located at both ends of a row which consists of N spheres is measured in terms of pixel numbers, and divided by (N-1) to obtain the average pixel numbers per particle. This value is multiplied by length conversion constant, which has been determined in advance, to obtain the mean diameter of polystylene spheres. The length conversion constant is found from the microscopic image of calibrated standard stage micrometer plate. In order to reduce error in finding center positions of the scale and particles in the image, a software filter which dilates bright (or dark) object has been used. Spheres having nominal size of 1, 2, 3, 5, and 10$\mu$m were measured, and the maximum deviation of the measured mean diameter valus from their certified values was 0.7%. Standard particles which is larger than 1 $\mu$m can be calibrated by this method with measuredment uncertainy (k=2) lees than 1.53%.

• Korean Chemical Engineering Research
• /
• v.48 no.3
• /
• pp.401-404
• /
• 2010

A spray-type cleaning agent in utilizing dust-remover on PCB was chosen to study. In cleaning of electronic and semiconductor equipment, a substrate(IPC-A-36) was used to test the jet forces of the agent. And according to the jet forces time of the cleaning agent, the corresponding moving distances were compared with the spray times, and for the pollutants of iron powder and dust, the cleaning efficiency was tested with the IPC-A-36 by a weight method. The moving distance increased with the spray cleaning time longer. For a spray cleaning time of 3sec, the cleaning efficiency decreased with the amount of dust and the iron powder. It was also observed that the dust was remarkably removed, compared to the iron powder.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.41 no.1
• /
• pp.77-82
• /
• 2004

In this paper we present a barcode reader to decode two-dimensional symbology PDF417 and propose a novel method to extract the bar-space Patterns directly from the gray-level barcode image, which employs the location and the distance between extreme points of each row or column of the barcode image. This algerian proves to be very robust from the high convolutional distortion environments such as defocussing and warping, even under badly illuminating condition. If the scanned barcode image is a result of the convolution of a Gaussian-shaped point spread function with a hi-level image, popular image segmentation methods such as image thresholding can not distinguish between very narrow bar-space patterns. The Proposed algorithm shows improved Performance over current barcode readers.