• 한국통신학회논문지
• /
• 제38A권8호
• /
• pp.650-655
• /
• 2013

광저장장치의 DC-억압을 위한 멀티모드 코딩 기법 중 Guided Scrambling (GS) 코딩기법이 널리 사용된다. 홀로그래픽 저장장치를 위한 DC-억압 GS코딩에서는 후보코드 선택기준으로 심볼의 균등한 분포 및 심볼간 천이의 최대화 기준이 고려되었다. 본 연구에서는 후보코드행렬의 digital sum value (DSV)의 $l_{\infty}$-norm을 최소화하는 minimum DSV (MDSV) 기준 GS코딩을 정수계획법 모형으로 수식화하고, 제안된 모형을 사용하여 MDSV 기준과 최대천이강도 기준이 적용된 GS코딩의 성능을 제어비트수, 행렬크기 및 스크램블링 다항식들의 조합에 대하여 평가한다.

• 한국멀티미디어학회논문지
• /
• 제24권7호
• /
• pp.849-859
• /
• 2021

Prostate cancer (PCa) is a fatal disease that occurs in men. In general, PCa cells are found in the prostate gland. Early diagnosis is the key to prevent the spreading of cancers to other parts of the body. In this case, deep learning-based systems can detect and distinguish histological patterns in microscopy images. The histological grades used for the analysis were benign, grade 3, grade 4, and grade 5. In this study, we attempt to use transfer learning and fine-tuning methods as well as different model architectures to develop and compare the models. We implemented MobileNet, ResNet50, and DenseNet121 models and used three different strategies of freezing layers techniques of fine-tuning, to get various pre-trained weights to improve accuracy. Finally, transfer learning using MobileNet with the half-layer frozen showed the best results among the nine models, and 90% accuracy was obtained on the test data set.

• 디지털산업정보학회논문지
• /
• 제19권3호
• /
• pp.245-251
• /
• 2023

Facial expression recognition plays a significant role in understanding human emotional states. With the advancement of AI and computer vision technologies, extensive research has been conducted in various fields, including improving customer service, medical diagnosis, and assessing learners' understanding in education. In this study, we develop a model that can infer emotions in real-time from a webcam using transfer learning with TensorFlow.js and MobileNet. While existing studies focus on achieving high accuracy using deep learning models, these models often require substantial resources due to their complex structure and computational demands. Consequently, there is a growing interest in developing lightweight deep learning models and transfer learning methods for restricted environments such as web browsers and edge devices. By employing MobileNet as the base model and performing transfer learning, our study develops a deep learning transfer model utilizing JavaScript-based TensorFlow.js, which can predict emotions in real-time using facial input from a webcam. This transfer model provides a foundation for implementing facial expression recognition in resource-constrained environments such as web and mobile applications, enabling its application in various industries.

• 대한치과기공학회지
• /
• 제41권4호
• /
• pp.287-293
• /
• 2019

Purpose: To evaluate the accuracy of the 3D printed die models and to investigate its clinical applicability. Methods: Stone die models were fabricated from conventional impressions(stone die model; SDM, n=7). 3D virtual models obtained from the digital impressions were manufactured as a 3D printed die models using a 3D printer(3D printed die models;3DM, n=7). Reference model, stone die models and 3D printed die models were scanned with a reference scanner. All dies model dataset were superimposed with the reference model file by the "Best fit alignment" method using 3D analysis software. Statistical analysis was performed using the independent t-test and 2-way ANOVA (α=.05). Results: The RMS value of the 3D printed die model was significantly larger than the RMS value of the stone die model (P<.001). As a result of 2-way ANOVA, significant differences were found between the model group (P<.001) and the part (P<.001), and their interaction effects (P<.001). Conclusion: The 3D printed die model showed lower accuracy than the stone die model. Therefore, it is necessary to further improve the performance of 3D printer in order to apply the 3D printed model in prosthodontics.

• 대한치과기공학회지
• /
• 제36권2호
• /
• pp.91-96
• /
• 2014

Purpose: The purpose of this study was evaluate the validity of polyurethane model fabricated by intra oral scanning method. Methods: Ten sam cases of stone models were manufactured from master model, and polyurethane models were made with the intra oral scanning and CNC milling method. One examiner individually measured 6 distances(intercanine distance, intermolar distance, two dental arch lengths(right, left) and two diagonal of dental arch lengths(right, left) on the stone models and the polyurethane models. The Mann-Whitney U test(${\alpha}$=0.05) were used for statistical analysis. Results: The mean difference between measurements made directly on the stone models and those made on the polyurethane models was 0.31-0.38mm. No statistically differences between the two groups were founded 4 distances(p>0.05), but 2 distances were statistically significant(p<0.05). Conclusion: Stone models showed larger than polyurethane models fabricated by intra oral scanning method.

• 대한전자공학회논문지
• /
• 제21권3호
• /
• pp.55-59
• /
• 1984

본 연구는 2차원 디지탈 필터링을 이용한 한글 인식 방법에 관한 것이다. 한글의 실용문자 1,659자의 표준 인쇄체를 초성, 중성, 종성의 위치별로 분석하여 총 170가지의 자모로 분류하고 이들 각 자모에 대한 2차원 디지탈 필터링된 모형을 얻어냈다. 이것을 바탕으로 하여 한글 조합문자에 중첩의 원리를 적용하여 자소를 분해. 인식하는 알고리즘을 제시하였다. 모의 시험의 결과. 인쇄체의 경우 100%의 인식률을 얻었다.

• International Journal of CAD/CAM
• /
• 제3권1_2호
• /
• pp.61-66
• /
• 2003

This paper describes a novel experiment that measures skin movement with respect to the flexional motion of a hand. The study was based on MR images in conjunction with CAD techniques. The MR images of the hand were captured in 3 different postures with surface markers. The surface markers attached to the skin where employed to trace skin movement during the flexional motion of the hand. After reconstructing 3D isosurfaces from the segmented MR images, the global registration was applied to the 3D models based on the particular bone shape of different postures. Skin movement was interpreted by measuring the centers of the surface markers in the registered models.

• 대한치과교정학회지
• /
• 제45권6호
• /
• pp.275-281
• /
• 2015

Objective: To assess the accuracy of anterior tooth movement using clear aligners in integrated three-dimensional digital models. Methods: Cone-beam computed tomography was performed before and after treatment with clear aligners in 32 patients. Plaster casts were laser-scanned for virtual setup and aligner fabrication. Differences in predicted and achieved root and crown positions of anterior teeth were compared on superimposed maxillofacial digital images and virtual models and analyzed by Student's t-test. Results: The mean discrepancies in maxillary and mandibular crown positions were $0.376{\pm}0.041mm$ and $0.398{\pm}0.037mm$, respectively. Maxillary and mandibular root positions differed by $2.062{\pm}0.128mm$ and $1.941{\pm}0.154mm$, respectively. Conclusions: Crowns but not roots of anterior teeth can be moved to designated positions using clear aligners, because these appliances cause tooth movement by tilting motion.

• 대한건축학회연합논문집
• /
• 제21권6호
• /
• pp.31-40
• /
• 2019

This study aims to establish Nammun-ro 2ga in Cheongju city in the 1960s as three-dimensional digital information data for the restoration of urban archetypes. For this purpose, referring to the existing restoration map and model of Cheongju urban area in the 1960s, and the results of this study are as follows. Firstly, the buildings that can be generally classified are prepared through the modeling of parametric families. Secondly, the untypical models(combined and broken roofs, atypical and large scale buildings) of them are simply performed through solid modeling. And then, these simplified models are simulated through a sky view, a walking sight, and information analysis. Through this study, it will be possible to visualize and regenerate the low and dense area of Cheongju city in the 1960s.

• 대한의용생체공학회:의공학회지
• /
• 제44권5호
• /
• pp.354-361
• /
• 2023

Most of spinal diseases are diagnosed based on the subjective judgment of a specialist, so numerous studies have been conducted to find objectivity by automating the diagnosis process using deep learning. In this paper, we propose a method that combines segmentation and feature extraction, which are frequently used techniques for diagnosing spinal diseases. Four models, U-Net, U-Net++, DeepLabv3+, and M-Net were trained and compared using 1000 X-ray images, and key-points were derived using Douglas-Peucker algorithms. For evaluation, Dice Similarity Coefficient(DSC), Intersection over Union(IoU), precision, recall, and area under precision-recall curve evaluation metrics were used and U-Net++ showed the best performance in all metrics with an average DSC of 0.9724. For the average Euclidean distance between estimated key-points and ground truth, U-Net was the best, followed by U-Net++. However the difference in average distance was about 0.1 pixels, which is not significant. The results suggest that it is possible to extract key-points based on segmentation and that it can be used to accurately diagnose various spinal diseases, including spondylolisthesis, with consistent criteria.