• Journal of the Korea Society of Computer and Information
• /
• v.14 no.3
• /
• pp.65-74
• /
• 2009

Scene change detection is pretreatment to index and search video information in video search system, and it is very important technology for overall performance. Existing scene change detection used single characteristic of pixel value difference, histogram difference, etc or mixed single characteristics that have complementary relationship. However, accuracy of those researches is very poor for special video such as infrared camera, night shooting. Therefore, this paper is proposed the method that is mixed color histogram and at algorithm for scene change detection at the specific movie. To verify the usefulness of a proposed method, we did an experiment which used color histogram only and KLT algorithm with color histogram. In result, evaluation index of proposed method is improved about 11.4% at the specific movie.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.33 no.1
• /
• pp.63-74
• /
• 2023

In order to ensure safety to invasion of privacy, Federated Learning(FL) that learns using parameters is emerging. However a paper that leaks training data using gradients was recently published. Our paper implements an experiment to leak training data using gradients in a federated learning environment, and proposes a method to improve reconstruction performance by improving existing attacks that leak training data. Experiments using Yale face database B, MNIST dataset on the proposed method show that federated learning is not safe from invasion of privacy by reconstructing up to 100 data out of 100 training data when performance of federated learning is high at accuracy=99~100%. In addition, by comparing the performance (MSE, PSNR, SSIM) of pixels and the performance of identification by Human Test, we want to emphasize the importance of the performance of identification rather than the performance of pixels.

• Progress in Medical Physics
• /
• v.34 no.3
• /
• pp.23-32
• /
• 2023

Purpose: In radiomics analysis, to evaluate features, and predict genetic characteristics and survival time, the pixel values of lesions depicted in computed tomography (CT) and magnetic resonance imaging (MRI) images are used. CT and MRI offer three-dimensional images, thus producing three-dimensional features (Features_3d) as output. However, in reports, the superiority between Features_3d and two-dimensional features (Features_2d) is distinct. In this study, we aimed to investigate whether a difference exists in the prediction accuracy of radiomics analysis of lung cancer using Features_2d and Features_3d. Methods: A total of 38 cases of large cell carcinoma (LCC) and 40 cases of squamous cell carcinoma (SCC) were selected for this study. Two- and three-dimensional lesion segmentations were performed. A total of 774 features were obtained. Using least absolute shrinkage and selection operator regression, seven Features_2d and six Features_3d were obtained. Results: Linear discriminant analysis revealed that the sensitivities of Features_2d and Features_3d to LCC were 86.8% and 89.5%, respectively. The coefficients of determination through multiple regression analysis and the areas under the receiver operating characteristic curve (AUC) were 0.68 and 0.70 and 0.93 and 0.94, respectively. The P-value of the estimated AUC was 0.87. Conclusions: No difference was found in the prediction accuracy for LCC and SCC between Features_2d and Features_3d.

• The Korean Journal of Nuclear Medicine
• /
• v.32 no.5
• /
• pp.414-424
• /
• 1998

Purpose: Cross-modality coregistration of positron emission tomography (PET) and magnetic resonance imaging (MR) could enhance the clinical information. In this study we propose a refined technique to improve the robustness of registration, and to implement more realistic visualization of the coregistered images. Materials and Methods: Using the sinogram of PET emission scan, we extracted the robust head boundary and used boundary-enhanced PET to coregister PET with MR. The pixels having 10% of maximum pixel value were considered as the boundary of sinogram. Boundary pixel values were exchanged with maximum value of sinogram. One hundred eighty boundary points were extracted at intervals of about 2 degree using simple threshold method from each slice of MR images. Best affined transformation between the two point sets was performed using least square fitting which should minimize the sum of Euclidean distance between the point sets. We reduced calculation time using pre-defined distance map. Finally we developed an automatic coregistration program using this boundary detection and surface matching technique. We designed a new weighted normalization technique to display the coregistered PET and MR images simultaneously. Results: Using our newly developed method, robust extraction of head boundary was possible and spatial registration was successfully performed. Mean displacement error was less than 2.0 mm. In visualization of coregistered images using weighted normalization method, structures shown in MR image could be realistically represented. Conclusion: Our refined technique could practically enhance the performance of automated three dimensional coregistration.

• The Journal of Korean Academy of Prosthodontics
• /
• v.36 no.2
• /
• pp.302-322
• /
• 1998

This study was accompolished to analyze and compare the number and area of the occlusal contact points in healthy volunteers and athletes with normal occlusion. For this study, 15 athletes(13 amles and 2 females with average age 20) and 21 healthy volunteers(14 mles and 7 females with average age 20.09) were selected. The visual display acquired by scanning of occlusal record in maximal intercuspation was converted into 16 gray value image. Then, using computer program(J & Lee Occlusal Analyser), the pixel which was in definite range of the gray value was recognized, and the numbers of recognized pixel were calculated to area. The results were as follows ; 1. The average numbers of total occlusal contact feints were 31.05 in control group, and 34.67 in athlete group. The average area of total occlusal contacts was $100.25mm^2$ in control group, and $127.78mm^2$ in athlete group. 2. In control group, the average numbers of occlusal contact points were revealed in order as follows ; the first molar(8.48), the second molar(8.24), the second premolar(4.71), the lateral incisor(2.90), the first premor(2.43), the central incisor(2.19), and the canine(2.1). The least average in canine(2.1) was similar to the average(2.19) in central incisor and (2.09) in lateral incisor. In athlete group, the average numbers of occlusal contact points were revealed in order as follows ; the first molar(8.97), the second molar(8.47), the second premolar(5.60), the canine(3.80), the lateral incisor(3.33), the first premolar(2.67), and the central incisor(1.93). 3. In control group, the average areas of occlusal contact surface were revealed in order as follows ; the first molar($39.47mm^3$), the second molar($37.54mm^3$), the second premolar($9.54mm^3$) the first premolar($6.18mm^3$), canine($3.49mm^3$), the central incisor($2.76mm^3$), and the lateral incisor($1.28mm^3$). In athlete group, the average areas of occlusal contact surface were revealed in order as follows ; the first molar($44.11mm^3$), the second molar($40.69mm^3$), the second premolar($16.50mm^3$), the first premolar($9.39mm^3$), the canine($5.08mm^3$), the lateral incisor($3.7mm^3$), and the central incisor($2.25mm^3$). 4. With aging in both control and athlete group, there was a decreasing tendancy in average number of occlusal contact point, and was an increasing tendancy in average area of occlusal contact surface. In comparison at each age, both the numbers and area of occlusal contact were greater in athlete group than in control group. It was not significant in the numbers of occlusal contact points beween athlete and control group(p>0.1), but significant in the area of occlusal contact surface(p<0.03). 5. In comparision as to the kind of sports(Gymnastics : 2, Rugby : 3, Soccor : 5, Ice hocky : 5), the numbers of occlusal contact points were the most in ice hocky, and the area of occlusal contact surface was the greatest in gymnastics. With increasing a career in athlete group, there was a decreasing tendancy in average numbers of occlusal contact points and was an increasing tendancy in average area of occlusal contact surface. 6. By T-scan analyzing, the contact numbers on the anterior teeth were greater in control group than in athlete group, and on the posterior teeth were greater in athlete group than in control group. And the results acquired by T-scan were lesser than that caquired by the silicone bite records. It was not significant in the posterior teeth, but significant in the anterior teeth. In T-scan records, the numbers of occlusal contact points on second molar were the greatest, but in the silicone records, the numbers on first molar were the greatest.

• Journal of Dental Rehabilitation and Applied Science
• /
• v.18 no.4
• /
• pp.235-249
• /
• 2002

This study was accompolished to analyze and compare the number and area of the occlusal contact points in healthy volunteers and athletes with normal occlusion. Before the study, 89 athletes(sports career:8.6 years, average age 20) at Kyung Hee University were selected, and survey was accomplished for athlete's recognition about sports dentistry. For this study, 15 athletes(13 amles and 2 females with average age 20) and 21 healthy volunteers(14 mles and 7 females with average age 20.09) at Kyung Hee University were selected. The visual display acquired by scanning of occlusal record in maximal intercuspation was converted into 16 gray value image. Then, using computer program(J & Lee Occlusal Analyser), the pixel which was in definite range of the gray value was recognized, and the numbers of recognized pixel were calculated to area. The results were as follows : (1) On the survey about sports dentistry, 28% of 89 athletes didn't agree that human occlusion may be important, and 30% didn't have any idea of the influence of human occlusion during their sports activities. (2) The average numbers of total occlusal contact points were 31.05 in control group, and 34.67 in athlete group. The average area of total occlusal contacts was $100.25mm^2$ in control group, and $127.78mm^2$ in athlete group. (3) In control group, the average numbers of occlusal contact points were revealed in order as follows; the first molar(8.48), the second molar(8.24), the second premolar(4.71), the lateral incisor(2.90), the first premor(2.43), the central incisor(2.19), and the canine(2.1). The least average in canine(2.1) was similar to the average(2.19) in central incisor and (2.09) in lateral incisor. In athlete group, the average numbers of occlusal contact points were revealed in order as follows; the first molar(8.87), the second molar(8.47), the second premolar(5.60), the canine(3.80), the lateral incisor(3.33), the first premolar(2.67), and the central incisor(1.93). (4) In control group, the average areas of occlusal contact surface were revealed in order as follows; the first molar($39.47mm^3$), the second molar($37.54mm^3$), the second premolar($9.54mm^3$) the first premolar($6.18mm^3$), canine($3.49mm^3$), the central incisor($2.76mm^3$), and the lateral incisor($1.28mm^3$). In athlete group, the average areas of occlusal contact surface were revealed in order as follows; the first molar($44.11mm^3$), the second molar($40.69mm^3$), the second premolar($16.50mm^3$), the first premolar($9.39mm^3$), the canine($5.08mm^3$), the lateral incisor($3.7mm^3$), and the central incisor($2.25mm^3$). (5). With aging in both control and athlete group, there was a decreasing tendancy in average number of occlusal contact point, and was an increasing tendancy in average area of occlusal contact surface. In comparison at each age, both the numbers and area of occlusal contact were greater in athlete group than in control group. It was not significant in the numbers of occlusal contact points beween athlete and control group(p>0.1), but significant in the area of occlusal contact surface(p<0.05). (6) In comparision as to the kind of sports(Gymnastics:2, Rugby:3, Soccor:5, Ice hocky:5), the numbers of occlusal contact points were the most in ice hocky, and the area of occlusal contact surface was the greatest in gymnastics. With increasing a career in athlete group, there was a decreasing tendancy in average numbers of occlusal contact points, and was an increasing tendancy in average area of occlusal contact surface.

• 전자공학회논문지 IE
• /
• v.44 no.3
• /
• pp.23-34
• /
• 2007

For an effective phase multiplexing in a volume holographic system, four types of phase code, pseudo random code(PSC), Hadamard matrix(HAM), pure random code(PRC), equivalent random code(ERC), used as reference beams are generated. In case of $32{\times}32$ address beam, a phase error with 0%, 5%, 10%, 15%, 20%, and 25% error rate, is purposely added to the real phase values in order to consider the practical SLM's nonlinear characteristics of phase modulation in computer simulation. Crosstalks and SNRs(signal-to-ratio) are comparatively analyzed for these phase codes by the auto-correlation and cross-correlation. PSC has the lowest cross-correlation mean value of 0.067 among four types of phase code, which means the SNR of the pseudo random phase code is higher than other phase codes. Also, the standard deviation of the pseudo random phase code indicating the degree of recalled data degradation is the lowest value of 0.0113. In order to analyze the affect by variation of pixel size, simulation is carried out by same method for the case of $32{\times}32$, $64{\times}64$, $128{\times}128$, $256{\times}256$ address beams.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.43 no.2 s.308
• /
• pp.40-47
• /
• 2006

This paper proposes a multitoning method using threshold modulation of MJBNM(Modified Jointly Blue Noise Mask) for banding artifacts reduction. As banding artifacts in multitoning appear as uniform dot distributions around the intermediate output levels, such multitone output results in discontinuity and visually unpleasing patterns in smooth transition regions. Therefore, to reduce these banding artifacts, the proposed method rearranges the dot distribution by introducing pixels in the neighborhood of output levels that occurs banding artifacts. First of all principal cause of banding artifacts are analyzed using mathematical description. Based on this analytical result, a threshold modulation technique of MJBNM which takes account of chrominance error and correlation between channels is applied. The original threshold range of MJBNM is first scaled linearly sot that the minimum and maximum of the scaled range include two pixel more than adjacent two output levels that cover an input value. In an input value is inside the vicinity of any intermediate output levels produce banding artifacts, the output is set to one of neighboring output levels based on the pointwise comparison result according to threshold modulation parameter that determines the dot density and distribution. In this case, adjacent pixels are introduced at the position where the scaled threshold values are located between two output levels and the minimum and maximum threshold values. Otherwise, a conventional multitoning method is applied. As a result, the proposed method effectively decreased the appearance of banding artifacts around the intermediate output levels. To evaluate the quality of the multitone result, HVS-WRMSE according to gray level for gray ramp image and S-CIELAB color difference for color ramp image are compared with other methods.

• Journal of Broadcast Engineering
• /
• v.25 no.2
• /
• pp.166-175
• /
• 2020

Recently, application fields for processing and using digital image contents in various forms and types are rapidly increasing. Since image content is high value-added content, the intellectual property rights of this content must be protected in order to activate the production and use of the digital image content. In this paper, we propose a deep learning based watermark embedding and extraction network. The proposed method is to maximize the robustness of the watermark against malicious/non-malicious attacks while preserving the invisibility of the host image. This network consists of a preprocessing network that changes the watermark to have the same resolution as the host image, a watermark embedding network that embeds watermark data while maintaining the resolution of the host image by three-dimensionally concatenating the changed host image and the watermark information, and a watermark extraction network that reduces the resolution and extracts watermarks. This network verifies the invisibility and robustness of the proposed method by experimenting with various pixel value change attacks and geometric attacks against various watermark data and host images with various resolutions, and shows that this method is universal and practical.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.31 no.1C
• /
• pp.55-64
• /
• 2006

In this paper, a new web-based remote 3D imaging communication system employing an adaptive matching algorithm is suggested. In the proposed method, feature values are extracted from the stereo image pair through estimation of the disparity and similarities between each pixel of the stereo image. And then, the matching window size for disparity estimation is adaptively selected depending on the magnitude of this feature value. Finally, the detected disparity map and the left image is transmitted into the client region through the network channel. And then, in the client region, right image is reconstructed and intermediate views be synthesized by a linear combination of the left and right images using interpolation in real-time. From some experiments on web based-transmission in real-time and synthesis of the intermediate views by using two kinds of stereo images of 'Joo' & 'Hoon' captured by real camera, it is analyzed that PSNRs of the intermediate views reconstructed by using the proposed transmission scheme are highly measured by 30dB for 'Joo', 27dB for 'Hoon' and the delay time required to obtain the intermediate image of 4 view is also kept to be very fast value of 67.2ms on average, respectively.