• Korean Journal of Remote Sensing
• /
• v.24 no.1
• /
• pp.57-64
• /
• 2008

In the previous study, the Haar wavelet was used as the sole basis function for the 3D discrete wavelet transform because the number of bands is too small to decompose a remotely sensed image in band direction with other basis functions. However, it is possible to use other basis functions for wavelet decomposition in horizontal and vertical directions because wavelet decomposition is independently performed in each direction. This study aims to classify a high spatial resolution image with the six types of basis function including the Haar function and to compare those results. The other wavelets are more helpful to classify high resolution imagery than the Haar wavelet. In overall accuracy, the Coif4 wavelet has the best result. The improvement of classification accuracy is different depending on the type of class and the type of wavelet. Using the basis functions with long length could be effective for improving accuracy in classification, especially for the classes of small area. This study is expected to be used as fundamental information for selecting optimal basis function according to the data properties in the 3D DWT based image classification.

• Proceedings of the Korean Society of Surveying, Geodesy, Photogrammetry, and Cartography Conference
• /
• 2005.11a
• /
• pp.81-104
• /
• 2005

Accurate 3D models in urban areas are essential for a variety of applications, such as virtual visualization, CIS, and mobile communications. LiDAR(Light Detection and Ranging) is a relatively new technology for directly obtaining 3D points. Because Manual 3D data reconstruction from LiDAR data is very costly and time consuming, many researchs is focused on the automatic extraction of the useful data. In this paper, we classified ground and non-ground points data from LiDAR data by using filtering, and we reconstructed the DTM(Digital Terrain Model) using ground points data, buildings using nonground points data. After the reconstruction, we assessed the accuracy of the DTM and buildings. As a result of, DTM from LiDAR data were 0.16m and 0.59m in high raised apartments areas and low house areas respectively, and buildings were matched with the accuracy of a l/5,000 digital map.

• Journal of information and communication convergence engineering
• /
• v.19 no.3
• /
• pp.148-154
• /
• 2021

With the advent of context-aware computing, many attempts were made to understand emotions. Among these various attempts, Speech Emotion Recognition (SER) is a method of recognizing the speaker's emotions through speech information. The SER is successful in selecting distinctive 'features' and 'classifying' them in an appropriate way. In this paper, the performances of SER using neural network models (e.g., fully connected network (FCN), convolutional neural network (CNN)) with Mel-Frequency Cepstral Coefficients (MFCC) are examined in terms of the accuracy and distribution of emotion recognition. For Ryerson Audio-Visual Database of Emotional Speech and Song (RAVDESS) dataset, by tuning model parameters, a two-dimensional Convolutional Neural Network (2D-CNN) model with MFCC showed the best performance with an average accuracy of 88.54% for 5 emotions, anger, happiness, calm, fear, and sadness, of men and women. In addition, by examining the distribution of emotion recognition accuracies for neural network models, the 2D-CNN with MFCC can expect an overall accuracy of 75% or more.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.2
• /
• pp.929-944
• /
• 2019

Many critical applications require accurate real-time human action recognition. However, there are many hurdles associated with capturing and pre-processing image data, calculating features, and classification because they consume significant resources for both storage and computation. To circumvent these hurdles, this paper presents a recognition machine learning (ML) based system model which uses reduced data structure features by projecting real 3D skeleton modality on virtual 2D space. The MMU VAAC dataset is used to test the proposed ML model. The results show a high accuracy rate of 97.88% which is only slightly lower than the accuracy when using the original 3D modality-based features but with a 75% reduction ratio from using RGB modality. These results motivate implementing the proposed recognition model on an embedded system platform in the future.

• Journal of Technologic Dentistry
• /
• v.41 no.2
• /
• pp.53-61
• /
• 2019

Purpose: The purpose of this study is to evaluate the discrepancy of scan process in dental intra oral scanner by comparing model scanner and anticipate possibility to introduce intra oral scan technique. Methods: 3D superimposition test was conducted to compare the scan discrepancy. The scanners used in this study are the e-oral scanner, the D750 model scanner, and the high precision CMM(3D Coordinate Measuring Machine). The standard of accuracy verification is ISO 5725-1; trueness and precision. Master model was manufactured by dental stone and scanned 5 times by intra oral, model scanner. Reference data was scanned 5 times by high accuracy CMM to evaluate the trueness. Results: Trueness of D750 scanner were $7.4{\mu}m$ $5.1{\mu}m$ $6.8{\mu}m$ at an abutment, an occluasal, a specific area. and trueness of e-scanner were $20.2{\mu}m$ $27.4{\mu}m$ $37.8{\mu}m$ at an abutment, an occluasal, a specific area. Precision of D750 scanner was $7.04{\mu}m$, e-scanner was $15.95{\mu}m$. Conclusion: When conducting in vitro test, The mean difference of trueness between e-scanner and D750 were $12.8{\mu}m$ at an abutment area, $22.3{\mu}m$ at an occlusal area, $31.0{\mu}m$ at a specific area and $8.91{\mu}m$ in precision. The scan discrepancies are within the range of clinical acceptance.

• Imaging Science in Dentistry
• /
• v.48 no.1
• /
• pp.11-19
• /
• 2018

Purpose: This study compared the diagnostic accuracy of cone-beam computed tomography (CBCT) scans obtained with 2 CBCT systems with high- and low-resolution modes for the detection of root perforations in endodontically treated mandibular molars. Materials and Methods: The root canals of 72 mandibular molars were cleaned and shaped. Perforations measuring 0.2, 0.3, and 0.4 mm in diameter were created at the furcation area of 48 roots, simulating strip perforations, or on the external surfaces of 48 roots, simulating root perforations. Forty-eight roots remained intact(control group). The roots were filled using gutta-percha (Gapadent, Tianjin, China) and AH26 sealer (Dentsply Maillefer, Ballaigues, Switzerland). The CBCT scans were obtained using the NewTom 3G (QR srl, Verona, Italy) and Cranex 3D (Soredex, Helsinki, Finland) CBCT systems in high- and low-resolution modes, and were evaluated by 2 observers. The chi-square test was used to assess the nominal variables. Results: In strip perforations, the accuracies of low- and high-resolution modes were 75% and 83% for NewTom 3G and 67% and 69% for Cranex 3D. In root perforations, the accuracies of low- and high-resolution modes were 79% and 83% for NewTom 3G and was 56% and 73% for Cranex 3D. Conclusion: The accuracy of the 2 CBCT systems was different for the detection of strip and root perforations. The Cranex 3D had non-significantly higher accuracy than the NewTom 3G. In both scanners, the high-resolution mode yielded significantly higher accuracy than the low-resolution mode. The diagnostic accuracy of CBCT scans was not affected by the perforation diameter.

• The korean journal of orthodontics
• /
• v.48 no.5
• /
• pp.292-303
• /
• 2018

Objective: Biplanar imaging systems allow for simultaneous acquisition of lateral and frontal cephalograms. The purpose of this study was to compare measurements recorded on three-dimensional (3D) cephalograms constructed from two-dimensional conventional radiographs and biplanar radiographs generated using a new biplanar imaging system with those recorded on cone-beam computed tomography (CBCT)-generated cephalograms in order to evaluate the accuracy of the 3D cephalograms generated using the biplanar imaging system. Methods: Three sets of lateral and frontal radiographs of 15 human dry skulls with prominent facial asymmetry were obtained using conventional radiography, the biplanar imaging system, and CBCT. To minimize errors in the construction of 3D cephalograms, fiducial markers were attached to anatomical landmarks prior to the acquisition of radiographs. Using the 3D $Ceph^{TM}$ program, 3D cephalograms were constructed from the images obtained using the biplanar imaging system (3D $ceph_{biplanar}$), conventional radiography (3D $ceph_{conv}$), and CBCT (3D $ceph_{cbct}$). A total of 34 measurements were obtained compared among the three image sets using paired t-tests and Bland-Altman plotting. Results: There were no statistically significant differences between the 3D $ceph_{biplanar}$ and 3D $ceph_{cbct}$ measurements. In addition, with the exception of one measurement, there were no significant differences between the 3D $ceph_{cbct}$ and 3D $ceph_{conv}$ measurements. However, the values obtained from 3D $ceph_{conv}$ showed larger deviations than those obtained from 3D $ceph_{biplanar}$. Conclusions: The results of this study suggest that the new biplanar imaging system enables the construction of accurate 3D cephalograms and could be a useful alternative to conventional radiography.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.14 no.2
• /
• pp.151-157
• /
• 1996

This study, experimentally, aims at presenting the methodology to construct an efficient digital terrain by com-paring and analyzing the accuracy among the existing Digital Terrain Models, develope 3-D fractal terrain model-ling program by applying digital algorithm of fractal geometry and using turbo pascal, and lastly perform basic research on constructing GSIS-based 3-D fractal terrain modelling system by integrating a PC-based GSIS Pack-age and the 3-D fractal terrain modelling program developed by this paper. The results are as follows -First, the method to produce TIN(Triangulated Irregular Network) by the combination of point data and line data was showed as an alternative to construct efficient Digital Terrain Model. Second, developing GSIS-based 3-D fractal terrain modelling system, applying fractal geometry is the basic research in developing the new terrain modelling method. also, this study presented the possibility of 3-D terrain modelling with the use of fractal.

• Imaging Science in Dentistry
• /
• v.51 no.4
• /
• pp.407-412
• /
• 2021

Purpose: The purpose of this study was to evaluate the accuracy of virtual 3-dimensional (3D) cephalograms constructed using the principle of biplanar radiography by comparing them with cone-beam computed tomography (CBCT) images. Materials and Methods: Thirty orthodontic patients were enrolled in this study. Frontal and lateral cephalograms were obtained with the use of a head posture aligner and reconstructed into 3D cephalograms using biplanar radiography software. Thirty-four measurements representing the height, width, depth, and oblique distance were computed in 3 dimensions, and compared with the measurements from the 3D images obtained by CBCT, using the paired t-test and Bland-Altman analysis. Results: Comparison of height, width, depth, and oblique measurements showed no statistically significant differences between the measurements obtained from 3D cephalograms and those from CBCT images (P>0.05). Bland-Altman plots also showed high agreement between the 3D cephalograms and CBCT images. Conclusion: Accurate 3D cephalograms can be constructed using the principle of biplanar radiography if frontal and lateral cephalograms can be obtained with a head posture aligner. Three-dimensional cephalograms generated using biplanar radiography can replace CBCT images taken for diagnostic purposes.

• Journal of the Korean Society of Safety
• /
• v.28 no.1
• /
• pp.24-28
• /
• 2013

This article evaluates the accuracy of noise measurements for 37 noise measurement applications for iPhone 4 and iPhone 3Gs. Noise levels were measured using simultaneously a precision sound level meter and iPhones installed noise measurement applications at the levels of 70 dB, 80 dB and 90 dB at 1,000 Hz. Measurement errors were estimated by subtracting two measurements between iPhone and sound level meter. It was found that measurement errors of 34 applications(89.2%) were greater than ${\pm}2$ dB which is the maximum allowable error range for the Type II sound level meter. It was only 4 applications that measurement errors lie within ${\pm}2$ dB error range. There was no significant differences among measurements with four iPhone 4s. However, there were significant differences between the measurements with iPhone 4 and iPhone 3Gs using the same application. It was due to the different hardware specifications such as microphone. Therefore, noise measurement applications, for example, which has to utilize hardware of the smartphone, should be programmed to identify hardware specifications and to adopt appropriate correction factors upon hardware specifications. In conclusion, it is necessary to check accuracy and validity before using the noise measurement applications for iPhones. Also, it was suggested that it should develop an evaluation guideline or protocol on accuracy testing for the measurement applications using a smartphone.