• Nuclear Medicine and Molecular Imaging
• /
• v.43 no.5
• /
• pp.459-467
• /
• 2009

Purpose: The maximum likelihood-expectation maximization (ML-EM) is the statistical reconstruction algorithm derived from probabilistic model of the emission and detection processes. Although the ML-EM has many advantages in accuracy and utility, the use of the ML-EM is limited due to the computational burden of iterating processing on a CPU (central processing unit). In this study, we developed a parallel computing technique on GPU (graphic processing unit) for ML-EM algorithm. Materials and Methods: Using Geforce 9800 GTX+ graphic card and CUDA (compute unified device architecture) the projection and backprojection in ML-EM algorithm were parallelized by NVIDIA's technology. The time delay on computations for projection, errors between measured and estimated data and backprojection in an iteration were measured. Total time included the latency in data transmission between RAM and GPU memory. Results: The total computation time of the CPU- and GPU-based ML-EM with 32 iterations were 3.83 and 0.26 see, respectively. In this case, the computing speed was improved about 15 times on GPU. When the number of iterations increased into 1024, the CPU- and GPU-based computing took totally 18 min and 8 see, respectively. The improvement was about 135 times and was caused by delay on CPU-based computing after certain iterations. On the other hand, the GPU-based computation provided very small variation on time delay per iteration due to use of shared memory. Conclusion: The GPU-based parallel computation for ML-EM improved significantly the computing speed and stability. The developed GPU-based ML-EM algorithm could be easily modified for some other imaging geometries.

• The Korean Journal of Nuclear Medicine Technology
• /
• v.19 no.1
• /
• pp.12-16
• /
• 2015

Purpose Principal component analysis (PCA) is a method often used in the neuroimagre analysis as a multivariate analysis technique for describing the structure of high dimensional correlation as the structure of lower dimensional space. PCA is a statistical procedure that uses an orthogonal transformation to convert a set of observations of correlated variables into a set of values of linearly independent variables called principal components. In this study, in order to investigate the usefulness of PCA in the brain PET image analysis, we tried to analyze C[11]-PIB PET image as a representative case. Materials and Methods Nineteen subjects were included in this study (normal = 9, AD/MCI = 10). For C[11]-PIB, PET scan were acquired for 20 min starting 40 min after intravenous injection of 9.6 MBq/kg C[11]-PIB. All emission recordings were acquired with the Biograph 6 Hi-Rez (Siemens-CTI, Knoxville, TN) in three-dimensional acquisition mode. Transmission map for attenuation-correction was acquired using the CT emission scans (130 kVp, 240 mA). Standardized uptake values (SUVs) of C[11]-PIB calculated from PET/CT. In normal subjects, 3T MRI T1-weighted images were obtained to create a C[11]-PIB template. Spatial normalization and smoothing were conducted as a pre-processing for PCA using SPM8 and PCA was conducted using Matlab2012b. Results Through the PCA, we obtained linearly uncorrelated independent principal component images. Principal component images obtained through the PCA can simplify the variation of whole C[11]-PIB images into several principal components including the variation of neocortex and white matter and the variation of deep brain structure such as pons. Conclusion PCA is useful to analyze and extract the main pattern of C[11]-PIB image. PCA, as a method of multivariate analysis, might be useful for pattern recognition of neuroimages such as FDG-PET or fMRI as well as C[11]-PIB image.

• Journal of Intelligence and Information Systems
• /
• v.24 no.2
• /
• pp.221-241
• /
• 2018

Deep learning is getting attention recently. The deep learning technique which had been applied in competitions of the International Conference on Image Recognition Technology(ILSVR) and AlphaGo is Convolution Neural Network(CNN). CNN is characterized in that the input image is divided into small sections to recognize the partial features and combine them to recognize as a whole. Deep learning technologies are expected to bring a lot of changes in our lives, but until now, its applications have been limited to image recognition and natural language processing. The use of deep learning techniques for business problems is still an early research stage. If their performance is proved, they can be applied to traditional business problems such as future marketing response prediction, fraud transaction detection, bankruptcy prediction, and so on. So, it is a very meaningful experiment to diagnose the possibility of solving business problems using deep learning technologies based on the case of online shopping companies which have big data, are relatively easy to identify customer behavior and has high utilization values. Especially, in online shopping companies, the competition environment is rapidly changing and becoming more intense. Therefore, analysis of customer behavior for maximizing profit is becoming more and more important for online shopping companies. In this study, we propose 'CNN model of Heterogeneous Information Integration' using CNN as a way to improve the predictive power of customer behavior in online shopping enterprises. In order to propose a model that optimizes the performance, which is a model that learns from the convolution neural network of the multi-layer perceptron structure by combining structured and unstructured information, this model uses 'heterogeneous information integration', 'unstructured information vector conversion', 'multi-layer perceptron design', and evaluate the performance of each architecture, and confirm the proposed model based on the results. In addition, the target variables for predicting customer behavior are defined as six binary classification problems: re-purchaser, churn, frequent shopper, frequent refund shopper, high amount shopper, high discount shopper. In order to verify the usefulness of the proposed model, we conducted experiments using actual data of domestic specific online shopping company. This experiment uses actual transactions, customers, and VOC data of specific online shopping company in Korea. Data extraction criteria are defined for 47,947 customers who registered at least one VOC in January 2011 (1 month). The customer profiles of these customers, as well as a total of 19 months of trading data from September 2010 to March 2012, and VOCs posted for a month are used. The experiment of this study is divided into two stages. In the first step, we evaluate three architectures that affect the performance of the proposed model and select optimal parameters. We evaluate the performance with the proposed model. Experimental results show that the proposed model, which combines both structured and unstructured information, is superior compared to NBC(Naïve Bayes classification), SVM(Support vector machine), and ANN(Artificial neural network). Therefore, it is significant that the use of unstructured information contributes to predict customer behavior, and that CNN can be applied to solve business problems as well as image recognition and natural language processing problems. It can be confirmed through experiments that CNN is more effective in understanding and interpreting the meaning of context in text VOC data. And it is significant that the empirical research based on the actual data of the e-commerce company can extract very meaningful information from the VOC data written in the text format directly by the customer in the prediction of the customer behavior. Finally, through various experiments, it is possible to say that the proposed model provides useful information for the future research related to the parameter selection and its performance.

• Journal of Korean Tunnelling and Underground Space Association
• /
• v.19 no.1
• /
• pp.95-107
• /
• 2017

In this study, a preliminary study was undertaken for development of a tunnel incident automatic detection system based on a machine learning algorithm which is to detect a number of incidents taking place in tunnel in real time and also to be able to identify the type of incident. Two road sites where CCTVs are operating have been selected and a part of CCTV images are treated to produce sets of training data. The data sets are composed of position and time information of moving objects on CCTV screen which are extracted by initially detecting and tracking of incoming objects into CCTV screen by using a conventional image processing technique available in this study. And the data sets are matched with 6 categories of events such as lane change, stoping, etc which are also involved in the training data sets. The training data are learnt by a resilience neural network where two hidden layers are applied and 9 architectural models are set up for parametric studies, from which the architectural model, 300(first hidden layer)-150(second hidden layer) is found to be optimum in highest accuracy with respect to training data as well as testing data not used for training. From this study, it was shown that the highly variable and complex traffic and incident features could be well identified without any definition of feature regulation by using a concept of machine learning. In addition, detection capability and accuracy of the machine learning based system will be automatically enhanced as much as big data of CCTV images in tunnel becomes rich.

• Korean Journal of Remote Sensing
• /
• v.36 no.5_1
• /
• pp.847-860
• /
• 2020

The 2-pass DInSAR (Differential Interferometric SAR) processing steps for DEM generation consist of the co-registration of SAR image pair, interferogram generation, phase unwrapping, calculation of DEM errors, and geocoding, etc. It requires complicated steps, and the accuracy of data processing at each step affects the performance of the finally generated DEM. In this study, we developed an improved method for enhancing the performance of the DEM generation method based on the 2-pass DInSAR technique of TanDEM-X bistatic SAR images was developed. The developed DEM generation method is a method that can significantly reduce both the DEM error in the unwrapped phase image and that may occur during geocoding step. The performance analysis of the developed algorithm was performed by comparing the vertical accuracy (Root Mean Square Error, RMSE) between the existing method and the newly proposed method using the ground control point (GCP) generated from GPS survey. The vertical accuracy of the DInSAR-based DEM generated without correction for the unwrapped phase error and geocoding error is 39.617 m. However, the vertical accuracy of the DEM generated through the proposed method is 2.346 m. It was confirmed that the DEM accuracy was improved through the proposed correction method. Through the proposed 2-pass DInSAR-based DEM generation method, the SRTM DEM error observed by DInSAR was compensated for the SRTM 30 m DEM (vertical accuracy 5.567 m) used as a reference. Through this, it was possible to finally create a DEM with improved spatial resolution of about 5 times and vertical accuracy of about 2.4 times. In addition, the spatial resolution of the DEM generated through the proposed method was matched with the SRTM 30 m DEM and the TanDEM-X 90m DEM, and the vertical accuracy was compared. As a result, it was confirmed that the vertical accuracy was improved by about 1.7 and 1.6 times, respectively, and more accurate DEM generation was possible with the proposed method. If the method derived in this study is used to continuously update the DEM for regions with frequent morphological changes, it will be possible to update the DEM effectively in a short time at low cost.

• Radiation Oncology Journal
• /
• v.25 no.1
• /
• pp.54-61
• /
• 2007

[ $\underline{Purpose}$ ]: The purpose of this study is to develop a practical method for determining accurate marker positions for prostate cancer radiotherapy using CT images and kV x-ray images obtained from the use of the on- board imager (OBI). $\underline{Materials\;and\;Methods}$: Three gold seed markers were implanted into the reference position inside a prostate gland by a urologist. Multiple digital image processing techniques were used to determine seed marker position and the center-of-mass (COM) technique was employed to determine a representative reference seed marker position. A setup discrepancy can be estimated by comparing a computed $COM_{OBI}$ with the reference $COM_{CT}$. A proposed algorithm was applied to a seed phantom and to four prostate cancer patients with seed implants treated in our clinic. $\underline{Results}$: In the phantom study, the calculated $COM_{CT}$ and $COM_{OBI}$ agreed with $COM_{actual}$ within a millimeter. The algorithm also could localize each seed marker correctly and calculated $COM_{CT}$ and $COM_{OBI}$ for all CT and kV x-ray image sets, respectively. Discrepancies of setup errors between 2D-2D matching results using the OBI application and results using the proposed algorithm were less than one millimeter for each axis. The setup error of each patient was in the range of $0.1{\pm}2.7{\sim}1.8{\pm}6.6\;mm$ in the AP direction, $0.8{\pm}1.6{\sim}2.0{\pm}2.7\;mm$ in the SI direction and $-0.9{\pm}1.5{\sim}2.8{\pm}3.0\;mm$ in the lateral direction, even though the setup error was quite patient dependent. $\underline{Conclusion}$: As it took less than 10 seconds to evaluate a setup discrepancy, it can be helpful to reduce the setup correction time while minimizing subjective factors that may be user dependent. However, the on-line correction process should be integrated into the treatment machine control system for a more reliable procedure.

• The KIPS Transactions:PartB
• /
• v.14B no.4
• /
• pp.311-320
• /
• 2007

This paper presents vision-based 3D facial expression animation technique and system which provide the robust 3D head pose estimation and real-time facial expression control. Many researches of 3D face animation have been done for the facial expression control itself rather than focusing on 3D head motion tracking. However, the head motion tracking is one of critical issues to be solved for developing realistic facial animation. In this research, we developed an integrated animation system that includes 3D head motion tracking and facial expression control at the same time. The proposed system consists of three major phases: face detection, 3D head motion tracking, and facial expression control. For face detection, with the non-parametric HT skin color model and template matching, we can detect the facial region efficiently from video frame. For 3D head motion tracking, we exploit the cylindrical head model that is projected to the initial head motion template. Given an initial reference template of the face image and the corresponding head motion, the cylindrical head model is created and the foil head motion is traced based on the optical flow method. For the facial expression cloning we utilize the feature-based method, The major facial feature points are detected by the geometry of information of the face with template matching and traced by optical flow. Since the locations of varying feature points are composed of head motion and facial expression information, the animation parameters which describe the variation of the facial features are acquired from geometrically transformed frontal head pose image. Finally, the facial expression cloning is done by two fitting process. The control points of the 3D model are varied applying the animation parameters to the face model, and the non-feature points around the control points are changed by use of Radial Basis Function(RBF). From the experiment, we can prove that the developed vision-based animation system can create realistic facial animation with robust head pose estimation and facial variation from input video image.

• Journal of Intelligence and Information Systems
• /
• v.18 no.1
• /
• pp.1-21
• /
• 2012

In recent years, mobile phone has experienced an extremely fast evolution. It is equipped with high-quality color displays, high resolution cameras, and real-time accelerated 3D graphics. In addition, some other features are includes GPS sensor and Digital Compass, etc. This evolution advent significantly helps the application developers to use the power of smart-phones, to create a rich environment that offers a wide range of services and exciting possibilities. To date mobile AR in outdoor research there are many popular location-based AR services, such Layar and Wikitude. These systems have big limitation the AR contents hardly overlaid on the real target. Another research is context-based AR services using image recognition and tracking. The AR contents are precisely overlaid on the real target. But the real-time performance is restricted by the retrieval time and hardly implement in large scale area. In our work, we exploit to combine advantages of location-based AR with context-based AR. The system can easily find out surrounding landmarks first and then do the recognition and tracking with them. The proposed system mainly consists of two major parts-landmark browsing module and annotation module. In landmark browsing module, user can view an augmented virtual information (information media), such as text, picture and video on their smart-phone viewfinder, when they pointing out their smart-phone to a certain building or landmark. For this, landmark recognition technique is applied in this work. SURF point-based features are used in the matching process due to their robustness. To ensure the image retrieval and matching processes is fast enough for real time tracking, we exploit the contextual device (GPS and digital compass) information. This is necessary to select the nearest and pointed orientation landmarks from the database. The queried image is only matched with this selected data. Therefore, the speed for matching will be significantly increased. Secondly is the annotation module. Instead of viewing only the augmented information media, user can create virtual annotation based on linked data. Having to know a full knowledge about the landmark, are not necessary required. They can simply look for the appropriate topic by searching it with a keyword in linked data. With this, it helps the system to find out target URI in order to generate correct AR contents. On the other hand, in order to recognize target landmarks, images of selected building or landmark are captured from different angle and distance. This procedure looks like a similar processing of building a connection between the real building and the virtual information existed in the Linked Open Data. In our experiments, search range in the database is reduced by clustering images into groups according to their coordinates. A Grid-base clustering method and user location information are used to restrict the retrieval range. Comparing the existed research using cluster and GPS information the retrieval time is around 70~80ms. Experiment results show our approach the retrieval time reduces to around 18~20ms in average. Therefore the totally processing time is reduced from 490~540ms to 438~480ms. The performance improvement will be more obvious when the database growing. It demonstrates the proposed system is efficient and robust in many cases.

• The Journal of the Acoustical Society of Korea
• /
• v.39 no.1
• /
• pp.16-23
• /
• 2020

In High-Intensity Focused Ultrasound (HIFU) treatment, effective localization of HIFU focus is important for developing a safe treatment plan. While Magnetic Resonance Imaging guided HIFU (MRIgHIFU) can visualize the ultrasound path during the treatment for localizing HIFU focus, it is challenging in ultrasound imaging guided HIFU (USIgHIFU). In the present study, a real-time ultrasound beam visualization technique capable of localizing HIFU focus is presented for USIgHIFU. In the proposed method, a short pulse, with the same center frequency of an imaging ultrasound transducer below the regulated acoustic intensity (i.e., Ispta < 720 mW/㎠), was transmitted through a HIFU transducer whereupon backscattered signals were received by the imaging transducer. To visualize the HIFU beam path, the backscattered signals underwent dynamic receive focusing and subsequent echo processing. From in vitro experiments with bovine serum albumin gel phantoms, the HIFU beam path was clearly depicted with low acoustic intensity (i.e., Ispta of 94.8 mW/㎠) and the HIFU focus was successfully localized before any damages were produced. This result indicates that the proposed ultrasound beam path visualization method can be used for localizing the HIFU focus in real time while minimizing unwanted tissue damage in USIgHIFU treatment.

• Journal of Korean Society of Coastal and Ocean Engineers
• /
• v.29 no.3
• /
• pp.162-168
• /
• 2017

Shallow-water bathymetry survey has been conducted using high definition color images obtained at the altitude of 100 m above sea level using a drone. Shallow-water bathymetry data are one of the most important input data for the research of beach erosion problems. Especially, accurate bathymetry data within closure depth are critically important, because most of the interesting phenomena occur in the surf zone. However, it is extremely difficult to obtain accurate bathymetry data due to wave-induced currents and breaking waves in this region. Therefore, optical remote sensing technique using a small drone is considered to be attractive alternative. This paper presents the potential utilization of image processing algorithms using multi-variable linear regression applied to red, green, blue and grey band images for estimating shallow water depth using a drone with HD camera. Optical remote sensing analysis conducted at Wolpo beach showed promising results. Estimated water depths within 5 m showed correlation coefficient of 0.99 and maximum error of 0.2 m compared with water depth surveyed through manual as well as ship-board echo-sounder measurements.