• The KIPS Transactions:PartC
• /
• v.15C no.3
• /
• pp.149-156
• /
• 2008

JPEG2000 is the upcoming image coding standard that provides better compression rate and image quality compared with JPEG. Lazy-mode steganography guarantees the safe communication under the two information loss stages in JPEG2000. However, it causes the severe changes of the code-block noise variance sequence after embedding and that is detectable under the steganalysis using the Hilbert-Huang transform (HHT) based sequential analysis. In this paper, a JPEG2000 lazy-mode steganography method is presented. The code blocks which produce the sudden variation of the noise variance after embedding are estimated by calculating low precision code-block variance (LPV) and low precision code-block noise variance (LPNV). By avoiding those code-blocks from embedding, our algorithm preserves the sequence and makes stego images secure under the HHT-based steganalytic detection. In addition, it prevents a severe degradation of image quality by using JPEG2000 quality layer information. On various 2048 images, experiments are performed to show the effective reduction of the noise variation after message embedding and the stable performance against HHT-based steganalysis.

• Investigative Magnetic Resonance Imaging
• /
• v.6 no.2
• /
• pp.137-146
• /
• 2002

Purpose : This paper introduces an improved classification approach which adopts a deterministic relaxation method and an agglomerative clustering technique for the classification of MRI using neural network. The proposed approach can solve the problems of convergency to local optima and computational burden caused by a large number of input patterns when a neural network is used for image classification. Materials and methods : Application of Hopfield neural network has been solving various optimization problems. However, major problem of mapping an image classification problem into a neural network is that network is opt to converge to local optima and its convergency toward the global solution with a standard stochastic relaxation spends much time. Therefore, to avoid local solutions and to achieve fast convergency toward a global optimization, we adopt MFA to a Hopfield network during the classification. MFA replaces the stochastic nature of simulated annealing method with a set of deterministic update rules that act on the average value of the variable. By minimizing averages, it is possible to converge to an equilibrium state considerably faster than standard simulated annealing method. Moreover, the proposed agglomerative clustering algorithm which determines the underlying clusters of the image provides initial input values of Hopfield neural network. Results : The proposed approach which uses agglomerative clustering and deterministic relaxation approach resolves the problem of local optimization and achieves fast convergency toward a global optimization when a neural network is used for MRI classification. Conclusion : In this paper, we introduce a new paradigm to classify MRI using clustering analysis and deterministic relaxation for neural network to improve the classification results.

• Proceedings of the Korea Institute of Convergence Signal Processing
• /
• 2001.06a
• /
• pp.173-176
• /
• 2001

In this thesis, a new measurement system is implemented for depth data extraction based on the camera calibration of the known pattern. The relation between 3D world coordinate and 2D image coordinate is analyzed. A new camera calibration algorithm is established from the analysis and then, the internal variables and external variables of the CCD camera are obtained. Suppose that the measurement plane is horizontal plane, from the 2D plane equation and coordinate transformation equation the approximation values corresponding minimum values using Newton-Rabbson method is obtained and they are stored into the look-up table for real time processing . A slit laser light is projected onto the object, and a 2D image obtained on the x-z plane in the measurement system. A 3D shape image can be obtained as the 2D (x-z)images are continuously acquired, during the object is moving to the y direction. The 3D shape images are displayed on computer monitor by use of OpenGL software. In a measuremental result, we found that the resolution of pixels have $\pm$ 1% of error in depth data. It seems that the error components are due to the vibration of mechanic and optical system. We expect that the measurement system need some of mechanic stability and precision optical system in order to improve the system.

• The Journal of the Acoustical Society of Korea
• /
• v.21 no.5
• /
• pp.471-477
• /
• 2002

Recently, lip-reading has been studied actively as an auxiliary method of automatic speech recognition(ASR) in noisy environments. However, almost of research results were obtained based on the database constructed in indoor condition. So, we dont know how developed lip-reading algorithms are robust to dynamic variation of image. Currently we have developed a lip-reading system based on image-transform based algorithm. This system recognize 22 words and this word recognizer achieves word recognition of up to 53.54%. In this paper we present how stable the lip-reading system is in environmental variance and what the main variant factors are about dropping off in word-recognition performance. For studying lip-reading robustness we consider spatial valiance (translation, rotation, scaling) and illumination variance. Two kinds of test data are used. One Is the simulated lip image database and the other is real dynamic database captured in car environment. As a result of our experiment, we show that the spatial variance is one of degradations factors of lip reading performance. But the most important factor of degradation is not the spatial variance. The illumination variances make severe reduction of recognition rates as much as 70%. In conclusion, robust lip reading algorithms against illumination variances should be developed for using lip reading as a complementary method of ASR.

• Journal of the Korea Convergence Society
• /
• v.8 no.6
• /
• pp.29-36
• /
• 2017

In this paper, we propose a method for generating improved saliency map by learning multiple features to improve the accuracy and reliability of saliency map which has similar result to human visual perception type. In order to overcome the inaccurate result of reverse selection or partial loss in color based salient area estimation in existing salience map generation, the proposed method generates multi feature data based on learning. The features to be considered in the image are analyzed through the process of distinguishing the color pattern and the region having the specificity in the original image, and the learning data is composed by the combination of the similar protrusion area definition and the specificity area using the LAB color space based color analysis. After combining the training data with the extrinsic information obtained from low level features such as frequency, color, and focus information, we reconstructed the final saliency map to minimize the inaccurate saliency area. For the experiment, we compared the ground truth image with the experimental results and obtained the precision-recall value.

• Journal of Korean Neurosurgical Society
• /
• v.62 no.4
• /
• pp.376-381
• /
• 2019

Objective : To define optimal method that calculate the safe direction of cervical pedicle screw placement using computed tomography (CT) image based three dimensional (3D) cortical shell model of human cervical spine. Methods : Cortical shell model of cervical spine from C3 to C6 was made after segmentation of in vivo CT image data of 44 volunteers. Three dimensional Cartesian coordinate of all points constituting surface of whole vertebra, bilateral pedicle and posterior wall were acquired. The ideal trajectory of pedicle screw insertion was defined as viewing direction at which the inner area of pedicle become largest when we see through the biconcave tubular pedicle. The ideal trajectory of 352 pedicles (eight pedicles for each of 44 subjects) were calculated using custom made program and were changed from global coordinate to local coordinate according to the three dimensional position of posterior wall of each vertebral body. The transverse and sagittal angle of trajectory were defined as the angle between ideal trajectory line and perpendicular line of posterior wall in the horizontal and sagittal plane. The averages and standard deviations of all measurements were calculated. Results : The average transverse angles were $50.60^{\circ}{\pm}6.22^{\circ}$ at C3, $51.42^{\circ}{\pm}7.44^{\circ}$ at C4, $47.79^{\circ}{\pm}7.61^{\circ}$ at C5, and $41.24^{\circ}{\pm}7.76^{\circ}$ at C6. The transverse angle becomes more steep from C3 to C6. The mean sagittal angles were $9.72^{\circ}{\pm}6.73^{\circ}$ downward at C3, $5.09^{\circ}{\pm}6.39^{\circ}$ downward at C4, $0.08^{\circ}{\pm}6.06^{\circ}$ downward at C5, and $1.67^{\circ}{\pm}6.06^{\circ}$ upward at C6. The sagittal angle changes from caudad to cephalad from C3 to C6. Conclusion : The absolute values of transverse and sagittal angle in our study were not same but the trend of changes were similar to previous studies. Because we know 3D address of all points constituting cortical shell of cervical vertebrae. we can easily reconstruct 3D model and manage it freely using computer program. More creative measurement of morphological characteristics could be carried out than direct inspection of raw bone. Furthermore this concept of measurement could be used for the computing program of automated robotic screw insertion.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.21 no.3
• /
• pp.59-66
• /
• 2021

Humans mainly recognize surrounding objects using visual and auditory information among the five senses (sight, hearing, smell, touch, taste). Major research related to the latest object recognition mainly focuses on analysis using image sensor information. In this paper, after emitting various chirp audio signals into the observation space, collecting echoes through a 2-channel receiving sensor, converting them into spectral images, an object recognition experiment in 3D space was conducted using an image learning algorithm based on deep learning. Through this experiment, the experiment was conducted in a situation where there is noise and echo generated in a general indoor environment, not in the ideal condition of an anechoic room, and the object recognition through echo was able to estimate the position of the object with 83% accuracy. In addition, it was possible to obtain visual information through sound through learning of 3D sound by mapping the inference result to the observation space and the 3D sound spatial signal and outputting it as sound. This means that the use of various echo information along with image information is required for object recognition research, and it is thought that this technology can be used for augmented reality through 3D sound.

• Journal of Internet Computing and Services
• /
• v.25 no.1
• /
• pp.29-37
• /
• 2024

In this paper, we propose an efficient image processing system to detect and track the movement of specific objects such as patients. The proposed system extracts the outline area of an object from a binarized difference image by applying a thinning algorithm that enables more precise detection compared to previous algorithms and is advantageous for mixed-mode design. The binarization and thinning steps, which require a lot of computation, are designed based on RTL (Register Transfer Level) and replaced with optimized hardware blocks through logic circuit synthesis. The designed binarization and thinning block was synthesized into a logic circuit using the standard 180n CMOS library and its operation was verified through simulation. To compare software-based performance, performance analysis of binary and thinning operations was also performed by applying sample images with 640 × 360 resolution in a 32-bit FPGA embedded system environment. As a result of verification, it was confirmed that the mixed-mode design can improve the processing speed by 93.8% in the binary and thinning stages compared to the previous software-only processing speed. The proposed mixed-mode system for object recognition is expected to be able to efficiently monitor patient movements even in an edge computing environment where artificial intelligence networks are not applied.

• Progress in Medical Physics
• /
• v.23 no.4
• /
• pp.309-316
• /
• 2012

Multimodal-imaging technique has been rapidly developed for improvement of diagnosis and evaluation of therapeutic effects. In despite of integrated hardware, registration accuracy was decreased due to a discrepancy between multimodal image and insufficiency of count in accordance with different acquisition method of each modality. The purpose of this study was to improve the PET image by event data resampling through analysis of data format, noise and statistical properties of small animal PET list data. Inveon PET listmode data was acquired as static data for 10 min after 60 min of 37 MBq/0.1 ml $^{18}F$-FDG injection via tail vein. Listmode data format was consist of packet containing 48 bit in which divided 8 bit header and 40 bit payload space. Realigned sinogram was generated from resampled event data of original listmode by using adjustment of LOR location, simple event magnification and nonparametric bootstrap. Sinogram was reconstructed for imaging using OSEM 2D algorithm with 16 subset and 4 iterations. Prompt coincidence was 13,940,707 count measured from PET data header and 13,936,687 count measured from analysis of list-event data. In simple event magnification of PET data, maximum was improved from 1.336 to 1.743, but noise was also increased. Resampling efficiency of PET data was assessed from de-noised and improved image by shift operation of payload value of sequential packet. Bootstrap resampling technique provides the PET image which noise and statistical properties was improved. List-event data resampling method would be aid to improve registration accuracy and early diagnosis efficiency.

• KIPS Transactions on Software and Data Engineering
• /
• v.5 no.6
• /
• pp.295-306
• /
• 2016

The purpose of this paper is both to analyze research trends and to predict industrial flows using the meta-data from the previous studies on data quality. There have been many attempts to analyze the research trends in various fields till lately. However, analysis of previous studies on data quality has produced poor results because of its vast scope and data. Therefore, in this paper, we used a text mining, social network analysis for time series network analysis to analyze the vast scope and data of data quality collected from a Web of Science index database of papers published in the international data quality-field journals for 10 years. The analysis results are as follows: Decreases in Mathematical & Computational Biology, Chemistry, Health Care Sciences & Services, Biochemistry & Molecular Biology, Biochemistry & Molecular Biology, and Medical Information Science. Increases, on the contrary, in Environmental Sciences, Water Resources, Geology, and Instruments & Instrumentation. In addition, the social network analysis results show that the subjects which have the high centrality are analysis, algorithm, and network, and also, image, model, sensor, and optimization are increasing subjects in the data quality field. Furthermore, the industrial connection analysis result on data quality shows that there is high correlation between technique, industry, health, infrastructure, and customer service. And it predicted that the Environmental Sciences, Biotechnology, and Health Industry will be continuously developed. This paper will be useful for people, not only who are in the data quality industry field, but also the researchers who analyze research patterns and find out the industry connection on data quality.