• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.31B no.11
• /
• pp.53-62
• /
• 1994

This paper presents a new approach of analyzing errors resulting from nonideal general-purpose lighting environment when the Photometric Stereo Method (PSM) is applied to estimate the surface-orientation of a three-dimensional object. The approach introduces the explicit modeling of the lighting environment including a circular-disk type irradiance object plane and the direct simulation of the error distribution with the model. The light source is modeled as a point source that has a certain amount of beam angle, and the luminance distribution on the irradiance plane is modeled as a Gaussian function with different deviation values. A simulation algorithm is devised to estimate the light source orientation computing the average luminance intensities obtained from the irradiance object planes positioned in three different orientations. The effect of the nonideal lighting model is directly reflected in such simulation, because of the analogy between the PSM and the proposed algorithm. With an instrumental tool designed to provide arbitrary orientations of the object plane at the origin of the coordinate system, experiment can be performed in a systematic way for the error analysis and compensation. Simulations are performed to find out the error distribution by widely varying the light model and the orientation set of the object plane. The simulation results are compared with those of the experiment performed in the same way as the simulation. It is confirmed from the experiment that a fair amount of errors is due to the erroneous effect of the general-purpose lighting environment.

• Journal of KIISE:Software and Applications
• /
• v.32 no.8
• /
• pp.781-788
• /
• 2005

LDA (Linear Discriminant Analysis) is a data discrimination technique that seeks transformation to maximize the ratio of the between-class scatter and the within-class scatter While it has been successfully applied to several applications, it has two limitations, both concerning the underfitting problem. First, it fails to discriminate data with complex distributions since all data in each class are assumed to be distributed in the Gaussian manner; and second, it can lose class-wise information, since it produces only one transformation over the entire range of classes. We propose three extensions of LDA to overcome the above problems. The first extension overcomes the first problem by modeling the within-class scatter using a PCA mixture model that can represent more complex distribution. The second extension overcomes the second problem by taking different transformation for each class in order to provide class-wise features. The third extension combines these two modifications by representing each class in terms of the PCA mixture model and taking different transformation for each mixture component. It is shown that all our proposed extensions of LDA outperform LDA concerning classification errors for handwritten digit recognition and alphabet recognition.

• Journal of KIISE:Software and Applications
• /
• v.32 no.9
• /
• pp.901-911
• /
• 2005

With the development of e-commerce and the proliferation of easily accessible information, information filtering has become a popular technique to prune large information spaces so that users are directed toward those items that best meet their needs and preferences. While many collaborative filtering systems have succeeded in capturing the similarities among users or items based on ratings to provide good recommendations, there are still some challenges for them to be more efficient, especially the user bias problem, non-transitive association problem and cold start problem. Those three problems impede us to capture more accurate similarities among users or items. In this paper, we provide probabilistic model approaches for UCHM and ICHM which are suggested to solve the addressed problems in hopes of achieving better performance. In this probabilistic model, objects (users or items) are classified into groups and predictions are made for users considering the Gaussian distribution of user ratings. Experiments on a real-word data set illustrate that our proposed approach is comparable with others.

• Korean Journal of Remote Sensing
• /
• v.24 no.5
• /
• pp.483-496
• /
• 2008

Qualified ground control points (GCPs) are required to construct a digital elevation model (DEM) from a pushbroom stereo pair. An inverse geolocation algorithm for extracting GCPs from ERS SAR data and the SRTM DEM was recently developed. However, not all GCPs established by this method are accurate enough for direct application to the geometric correction of pushbroom images such as SPOT, IRS, etc, and thus a method for selecting and removing inaccurate points from the sets of GCPs is needed. In this study, we propose a method for evaluating GCP accuracy and winnowing sets of GCPs through orientation modeling of pushbroom image and validate performance of this method using SPOT stereo pair of Daejon City. It has been found that the statistical distribution of GCP positional errors is approximately Gaussian without bias, and that the residual errors estimated by orientation modeling have a linear relationship with the positional errors. Inaccurate GCPs have large positional errors and can be iteratively eliminated by thresholding the residual errors. Forty-one GCPs were initially extracted for the test, with mean the positional error values of 25.6m, 2.5m and -6.1m in the X-, Y- and Z-directions, respectively, and standard deviations of 62.4m, 37.6m and 15.0m. Twenty-one GCPs were eliminated by the proposed method, resulting in the standard deviations of the positional errors of the 20 final GCPs being reduced to 13.9m, 8.5m and 7.5m in the X-, Y- and Z-directions, respectively. Orientation modeling of the SPOT stereo pair was performed using the 20 GCPs, and the model was checked against 15 map-based points. The root mean square errors (RMSEs) of the model were 10.4m, 7.1m and 12.1m in X-, Y- and Z-directions, respectively. A SPOT DEM with a 20m ground resolution was successfully constructed using a automatic matching procedure.

• Wind and Structures
• /
• v.36 no.6
• /
• pp.393-404
• /
• 2023

Skewness and kurtosis are important higher-order statistics for simulating non-Gaussian wind pressure series on low-rise buildings, but their predictions are less studied in comparison with those of the low order statistics as mean and rms. The distribution gradients of skewness and kurtosis on roofs are evidently higher than those of mean and rms, which increases their prediction difficulty. The conventional artificial neural networks (ANNs) used for predicting mean and rms show unsatisfactory accuracy in predicting skewness and kurtosis owing to the limited capacity of shallow learning of ANNs. In this work, the deep neural networks (DNNs) model with the ability of deep learning is introduced to predict the skewness and kurtosis on a low-rise building. For obtaining the optimal generalization of the DNNs model, the hyper parameters are automatically determined by Bayesian Optimization (BO). Moreover, for providing a benchmark for future studies on predicting higher order statistics, the data sets for training and testing the DNNs model are extracted from the internationally open NIST-UWO database, and the prediction errors of all taps are comprehensively quantified by various error metrices. The results show that the prediction accuracy in this study is apparently better than that in the literature, since the correlation coefficient between the predicted and experimental results is 0.99 and 0.75 in this paper and the literature respectively. In the untrained cornering wind direction, the distributions of skewness and kurtosis are well captured by DNNs on the whole building including the roof corner with strong non-normality, and the correlation coefficients between the predicted and experimental results are 0.99 and 0.95 for skewness and kurtosis respectively.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.52 no.1
• /
• pp.89-101
• /
• 2015

This paper proposes a statistical model of 3-D positions when tracking moving targets using the uncooled infrared (IR) stereo camera system. The proposed model is derived from two errors. One is the position error which is caused by the sampling pixels in the digital image. The other is the timing jitter which results from the irregular capture-timing in the infrared cameras. The capture-timing in the IR camera is measured using the jitter meter designed in this paper, and the observed jitters are statistically modeled as Gaussian distribution. This paper derives an integrated probability distribution by combining jitter error with pixel position error. The combined error is modeled as the convolution of two error distributions. To verify the proposed statistical position error model, this paper has some experiments in tracking moving objects with IR stereo camera. The 3-D positions of object are accurately measured by the trajectory scanner, and 3-D positions are also estimated by stereo matching from IR stereo camera system. According to the experiments, the positions of moving object are estimated within the statistically reliable range which is derived by convolution of two probability models of pixel position error and timing jitter respectively. It is expected that the proposed statistical model can be applied to estimate the uncertain 3-D positions of moving objects in the diverse fields.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.14 no.3
• /
• pp.174-187
• /
• 2011

The flowed soil is able to be estimated from topographic data of before and after the debris flow. However, it is often difficult to obtain airborne LiDAR data before the debris flow area. Thus, this study tries to develop a topographic restoration method that can provide spatial distribution of flowed soil and reconstruct the topography before the debris flow using airborne LiDAR data. The topographic restoration method can express a numerical formula induced from a Gaussian mixture model after extracting the cross sections of linear or non-linear in debris flowed area. The topographic restoration method was verified by two ways using airborne LiDAR data of before and after the debris flow. First, each cross section extracted from the debris flow sites to restore the topography was compared with airborne LiDAR data of before the debris flow. Also, the topographic data produced after the topographic restoration method applied to the debris flow sites was verified by airborne LiDAR DEM. Verifying the results of the topographic restoration method, overall fitting accuracy showed high accuracy close to 0.5m.

• Journal of Broadcast Engineering
• /
• v.14 no.2
• /
• pp.144-153
• /
• 2009

Recently, as the necessity of a light-weighted video encoding technique has been rising for applications such as UCC(User Created Contents) or Multiview Video, Distributed Video Coding(DVC) where a decoder, not an encoder, performs the motion estimation/compensation taking most of computational complexity has been vigorously investigated. Wyner-Ziv coding reconstructs an image by eliminating the noise on side information which is decoder-side prediction of original image using channel code. Generally the side information of Wyner-Ziv coding is generated by using frame interpolation between key frames. The channel code such as Turbo code or LDPC code which shows a performance close to the Shannon's limit is employed. The noise model of Wyner-Ziv coding for channel decoding is called Virtual Channel Noise and is generally modeled by Laplacian or Gaussian distribution. In this paper, we propose a Wyner-Ziv coding method based on the frequency-adaptive channel noise modeling in transform domain. The experimental results with various sequences prove that the proposed method makes the channel noise model more accurate compared to the conventional scheme, resulting in improvement of the rate-distortion performance by up to 0.52dB.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.44 no.3
• /
• pp.80-87
• /
• 2007

In this paper, we investigate an enhancement method for Western epigraphic images, which is based on local statistics. Image data is partitioned into two regions, background and information. Statistical and functional analyses are proceeded for image modeling. The Western epigraphic images, for the most part, have shown the Gaussian distribution. It is clarified that each region can be differentiated statistically. The local normalization process algorithm is designed on this model. The parameter is extracted and it‘s properties are verified with the size of moving window. The spatial gray-level distribution is modified and regions are differentiated by adjusting parameter and the size of moving window. Local statistics are utilized for realization of the enhancement, so that difference between regions can be enhanced and noise or speckles of region can be smoothed. Experimental results are presented to show the superiority of the proposed algorithm over the conventional methods.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.15 no.9
• /
• pp.2000-2006
• /
• 2011

In this paper, the drain induced barrier lowering(DIBL) for doping distribution in the channel has been analyzed for double gate MOSFET(DGMOSFET). The DGMOSFET is extensively been studing because of adventages to be able to reduce the short channel effects(SCEs) to occur in convensional MOSFET. DIBL is SCE known as reduction of threshold voltage due to variation of energy band by high drain voltage. This DIBL has been analyzed for structural parameter and variation of channel doping profile for DGMOSFET. For this object, The analytical model of Poisson equation has been derived from Gaussian doping distribution for DGMOSFET. To verify potential and DIBL models based on this analytical Poisson's equation, the results have been compared with those of the numerical Poisson's equation, and DIBL for DGMOSFET has been investigated using this models.