• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.05a
• /
• pp.346-347
• /
• 2021

For pattern recognition for machine learning, the larger the amount of learning data, the better its performance. However, it is not always possible to secure a large amount of learning data with the types and information of patterns that must be detected in daily life. Therefore, it is necessary to significantly inflate a small data set for general machine learning. In this study, we study techniques to augment data so that machine learning can be performed. A representative method of performing machine learning using a small data set is the transfer learning technique. Transfer learning is a method of obtaining a result by performing basic learning with a general-purpose data set and then substituting the target data set into the final stage. In this study, a learning model trained with a general-purpose data set such as ImageNet is used as a feature extraction set using augmented data to detect a desired pattern.

• Journal of the Microelectronics and Packaging Society
• /
• v.30 no.4
• /
• pp.69-78
• /
• 2023

With the rapid growth of artificial intelligence, the demand for semiconductors is enormously increasing everywhere. To ensure the manufacturing quality and quantity simultaneously, the importance of automatic defect detection during the packaging process has been re-visited by adapting various deep learning-based methodologies into automatic packaging defect inspection. Deep learning (DL) models require a large amount of data for training, but due to the nature of the semiconductor industry where security is important, sharing and labeling of relevant data is challenging, making it difficult for model training. In this study, we propose a new framework for securing sufficient data for DL models with fewer computing resources through a divide-and-conquer approach. The proposed method divides high-resolution images into pre-defined sub-regions and assigns conditional labels to each region, then trains individual sub-regions and boundaries with boundary loss inducing the globally coherent and seamless images. Afterwards, full-size image is reconstructed by combining divided sub-regions. The experimental results show that the images obtained through this research have high efficiency, consistency, quality, and generality.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.46 no.3
• /
• pp.1-6
• /
• 2023

In the realm of dental prosthesis fabrication, obtaining accurate impressions has historically been a challenging and inefficient process, often hindered by hygiene concerns and patient discomfort. Addressing these limitations, Company D recently introduced a cutting-edge solution by harnessing the potential of intraoral scan images to create 3D dental models. However, the complexity of these scan images, encompassing not only teeth and gums but also the palate, tongue, and other structures, posed a new set of challenges. In response, we propose a sophisticated real-time image segmentation algorithm that selectively extracts pertinent data, specifically focusing on teeth and gums, from oral scan images obtained through Company D's oral scanner for 3D model generation. A key challenge we tackled was the detection of the intricate molar regions, common in dental imaging, which we effectively addressed through intelligent data augmentation for enhanced training. By placing significant emphasis on both accuracy and speed, critical factors for real-time intraoral scanning, our proposed algorithm demonstrated exceptional performance, boasting an impressive accuracy rate of 0.91 and an unrivaled FPS of 92.4. Compared to existing algorithms, our solution exhibited superior outcomes when integrated into Company D's oral scanner. This algorithm is scheduled for deployment and commercialization within Company D's intraoral scanner.

• Journal of radiological science and technology
• /
• v.47 no.1
• /
• pp.29-37
• /
• 2024

This study aims to evaluate the performance of the U-Net based learning model that may vary depending on the histogram equalization algorithm. The subject of the experiment were 17 radiology students of this college, and 1,727 data sets in which the region of interest was set in the thyroid after acquiring ultrasound image data were used. The training set consisted of 1,383 images, the validation set consisted of 172 and the test data set consisted of 172. The equalization algorithm was divided into Histogram Equalization(HE) and Contrast Limited Adaptive Histogram Equalization(CLAHE), and according to the clip limit, it was divided into CLAHE8-1, CLAHE8-2. CLAHE8-3. Deep Learning was learned through size control, histogram equalization, Z-score normalization, and data augmentation. As a result of the experiment, the Attention U-Net showed the highest performance from CLAHE8-2 to 0.8355, and the U-Net and BSU-Net showed the highest performance from CLAHE8-3 to 0.8303 and 0.8277. In the case of mIoU, the Attention U-Net was 0.7175 in CLAHE8-2, the U-Net was 0.7098 and the BSU-Net was 0.7060 in CLAHE8-3. This study attempted to confirm the effects of U-Net, Attention U-Net, and BSU-Net models when histogram equalization is performed on ultrasound images. The increase in Clip Limit can be expected to increase the ROI match with the prediction mask by clarifying the boundaries, which affects the improvement of the contrast of the thyroid area in deep learning model learning, and consequently affects the performance improvement.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.46 no.2
• /
• pp.84-91
• /
• 2009

This paper presents an online partial 3D modeling methodology that uses a mobile augmented reality system and aerial photographs, and a tracking methodology that compares the 3D model with a video image. Instead of relying on models which are created in advance, the system generates a 3D model for a real building on the fly by combining frontal and aerial views. A user's initial pose is estimated using an aerial photograph, which is retrieved from a database according to the user's GPS coordinates, and an inertial sensor which measures pitch. We detect edges of the rooftop based on Graph cut, and find edges and a corner of the bottom by minimizing the proposed cost function. To track the user's position and orientation in real-time, feature-based tracking is carried out based on salient points on the edges and the sides of a building the user is keeping in view. We implemented camera pose estimators using both a least squares estimator and an unscented Kalman filter (UKF). We evaluated the speed and accuracy of both approaches, and we demonstrated the usefulness of our computations as important building blocks for an Anywhere Augmentation scenario.

• Maxillofacial Plastic and Reconstructive Surgery
• /
• v.29 no.6
• /
• pp.485-493
• /
• 2007

Introduction: Although several types of calcium-phosphate coumpound have been frequently applied to osseous defects at maxillofacial area for many years, there is a controversy about its efficiency on bone conductivity comprared to xenograft bone substitute. Alloplastic carbonapatite has been introduced to improve disadvantages of hydroxyapatite and to mimic natural bone containing carbon elements. However, a preclinical study about its efficiency of osteoconductivity has not been reported. This study was performed to evaluate the early osteoconductive potential of synthetic carbonapatite with multiple pores relative to anorganic bovine xenograft. Materials and methods: Total 5 beagle dogs were used for maxillary augmentation model. The control (anorganic bovine xenograft) and experimental groups (synthetic carbonapatite) were randomly distributed in the mouth split design. After bone graft, all animals were sacrificed 4 weeks after surgery. Histological specimens with Masson Trichrome staining were made and histomorphometrically analysed with image analyser. The statistical analysis was performed using paired t-test. Results: In both groups, all animals had no complications. The experimental group showed relatively much new bone formation around and along the bone substitutes, whereas it was clearly reduced in the control group. The ratios of new bone area to total area, to material area and to the residual area excluding materials were higher in the experimental group ($0.13{\pm}0.03,\;0.40{\pm}0.13,\;0.20{\pm}0.06$ respectively) than in the control group ($0.01{\pm}0.01,\;0.03{\pm}0.02,\;0.03{\pm}0.03$, respectively). And the differences between both groups were statistically significant (p<0.001, <0.01, <0.01, respectively), while the ratio of material area to total area in two groups was not significant. Conclusion: Carbonapatite showed a high osteoconductivity in the early stage of bone healing compared to bovine derived anorganic bone substitute. This study suggests that this bone materials can be applied as a reliable bone substitute in the clinical treatment.

• Journal of Broadcast Engineering
• /
• v.27 no.3
• /
• pp.391-401
• /
• 2022

As deep learning-based object detection and recognition research have been developed recently, the scope of application to industry and real life is expanding. But deep learning-based systems in the construction system are still much less studied. Calculating materials in the construction system is still manual, so it is a reality that transactions of wrong volumn calculation are generated due to a lot of time required and difficulty in accurate accumulation. A fast and accurate automatic drawing recognition system is required to solve this problem. Therefore, we propose an AI-based automatic drawing recognition accumulation system that detects and recognizes steel materials in construction drawings. To accurately detect steel materials in construction drawings, we propose data augmentation techniques and spatial attention modules for improving small object detection performance based on YOLOv4. The detected steel material area is recognized by text, and the number of steel materials is integrated based on the predicted characters. Experimental results show that the proposed method increases the accuracy and precision by 1.8% and 16%, respectively, compared with the conventional YOLOv4. As for the proposed method, Precision performance was 0.938. The recall was 1. Average Precision AP_0.5 was 99.4% and AP_0.5:0.95 was 67%. Accuracy for character recognition obtained 99.9.% by configuring and learning a suitable dataset that contains fonts used in construction drawings compared to the 75.6% using the existing dataset. The average time required per image was 0.013 seconds in the detection, 0.65 seconds in character recognition, and 0.16 seconds in the accumulation, resulting in 0.84 seconds.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.16 no.5
• /
• pp.331-337
• /
• 2023

In this paper, we propose a deep learning model to detect lumbar 3 (L3) CT images to determine the occurrence and degree of sarcopenia. In addition, we would like to propose an optimization technique that uses oversampling ratio and class weight as design parameters to address the problem of performance degradation due to data imbalance between L3 level and non-L3 level portions of CT data. In order to train and test the model, a total of 150 whole-body CT images of 104 prostate cancer patients and 46 bladder cancer patients who visited Gangneung Asan Medical Center were used. The deep learning model used ResNet50, and the design parameters of the optimization technique were selected as six types of model hyperparameters, data augmentation ratio, and class weight. It was confirmed that the proposed optimization-based L3 level extraction model reduced the median L3 error by about 1.0 slices compared to the control model (a model that optimized only 5 types of hyperparameters). Through the results of this study, accurate L3 slice detection was possible, and additionally, we were able to present the possibility of effectively solving the data imbalance problem through oversampling through data augmentation and class weight adjustment.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.12 no.9
• /
• pp.1703-1710
• /
• 2008

As the performance of digital camera keeps going on, the digital camera becomes into wide use recently so that the number of digital images taken by digital camera is increased rapidly. Given the augmentation of digital images, the digital photo frame has appeared for storing and enjoying the digital image and receives attention as the price of digital frame becomes down. However, users with the existing digital frame only could display the digital images and stream media stored in local disk. In the paper, we design and implement a new digital photo frame baled on ubiquitous which could provide a more variety of service to the user. With the proposed digital photo frame, the user could obtain the environment information and the digital image from multimedia sensor network. The users also could enjoy the various stream media stored in the remote streaming server seamlessly. In addition, the user could update the digital image through the web site at anywhere.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.28 no.4
• /
• pp.55-61
• /
• 2024

Recently, the number of aging concrete structures is steadily increasing. This is because many of these structures are reaching their expected lifespan. Such structures require accurate inspections and persistent maintenance. Otherwise, their original functions and performance may degrade, potentially leading to safety accidents. Therefore, research on objective inspection technologies using deep learning and computer vision is actively being conducted. High-resolution images can accurately observe not only micro cracks but also spalling and exposed rebar, and deep learning enables automated detection. High detection performance in deep learning is only guaranteed with diverse and numerous training datasets. However, surface damage to concrete is not commonly captured in images, resulting in a lack of training data. To overcome this limitation, this study proposed a method for generating concrete surface damage images, including cracks, spalling, and exposed rebar, using stable diffusion. This method synthesizes new damage images by paired text and image data. For this purpose, a training dataset of 678 images was secured, and fine-tuning was performed through low-rank adaptation. The quality of the generated images was compared according to three base models of stable diffusion. As a result, a method to synthesize the most diverse and high-quality concrete damage images was developed. This research is expected to address the issue of data scarcity and contribute to improving the accuracy of deep learning-based damage detection algorithms in the future.