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.

This paper proposes a model and train method that can real-time detect objects and distances estimation based on a monocular camera by applying deep learning. It used YOLOv2 model which is applied to autonomous or robot due to the fast image processing speed. We have changed and learned the loss function so that the YOLOv2 model can detect objects and distances at the same time. The YOLOv2 loss function added a term for learning bounding box values x, y, w, h, and distance values z as 클래스ification losses. In addition, the learning was carried out by multiplying the distance term with parameters for the balance of learning. we trained the model location, recognition by camera and distance data measured by lidar so that we enable the model to estimate distance and objects from a monocular camera, even when the vehicle is going up or down hill. To evaluate the performance of object detection and distance estimation, MAP (Mean Average Precision) and Adjust R square were used and performance was compared with previous research papers. In addition, we compared the original YOLOv2 model FPS (Frame Per Second) for speed measurement with FPS of our model.

A new algorithm has been propose to detect the reflected light region as disturbances in a real-time vision system. There have been several attempts to detect existing reflected light region. The conventional mathematical approach requires a lot of complex processes so that it is not suitable for a real-time vision system. On the other hand, when a simple detection process has been applied, the reflected light region can not be detected accurately. Therefore, in order to detect reflected light region for a real-time vision system, the detection process requires a new algorithm that is as simple and accurate as possible. In order to extract the reflected light, the proposed algorithm has been adopted several filter equations and clustering processes in the HSI (Hue Saturation Intensity) color space. Also the proposed algorithm used the pre-defined reflected light data generated through the clustering processes to make the algorithm simple. To demonstrate the effectiveness of the proposed algorithm, several images with the reflected region have been used and the reflected regions are detected successfully.

One of the most frequently performed tasks in human-robot interaction (HRI), intelligent vehicles, and security systems is face related applications such as face recognition, facial expression recognition, driver state monitoring, and gaze estimation. In these applications, accurate head pose estimation is an important issue. However, conventional methods have been lacking in accuracy, robustness or processing speed in practical use. In this paper, we propose a novel method for estimating head pose with a monocular camera. The proposed algorithm is based on a deep neural network for multi-task learning using a small grayscale image. This network jointly detects multi-view faces and estimates head pose in hard environmental conditions such as illumination change and large pose change. The proposed framework quantitatively and qualitatively outperforms the state-of-the-art method with an average head pose mean error of less than 4.5° in real-time.

증강현실 기술을 게임에 접목하려는 시도가 많이 이루어지고 있다. 본 논문에서는 실시간으 로 스마트 기기의 카메라로부터 손동작 2차원 영상을 획득하고 HSV 컬러 영역을 활용하여 해 당 색상의 물체 또는 신체를 실시간으로 검출해, 게임의 요소로 활용하여 즐길 수 있는 2D 퍼 즐게임을 개발하였다. 이는 스마트 기기에 컴퓨터 비전을 접목한 형태의 게임으로, 교육 및 엔 터테인먼트 환경에서 스마트 기기의 휴대성과 가상현실 체험의 현장감을 접목시킨 새로운 인터 랙티브 기술이 될 것이다.

This paper presents a robust lane detection algorithm based on RGB color and shape information during autonomous car control in realtime. For realtime control, our algorithm increases its processing speed by employing minimal elements. Our algorithm extracts yellow and white pixels by computing the average and standard deviation values calculated from specific regions, and constructs elements based on the extracted pixels. By clustering elements, our algorithm finds the yellow center and white stop lanes on the road. Our algorithm is insensitive to the environment change and its processing speed is realtime-executable. Experimental results demonstrate the feasibility of our algorithm.

Recently, Total Organic Carbon (TOC) which can be measured instantly can be used as an organic pollutant index instead of BOD or COD due to the diversity of pollutants and non-degradable problem. The primary purpose of the present study is to reveal the properties of time series data for TOC which have been measured by real-time monitoring in Juam Lake and, in particularly, to understand the long- and short-term characteristics with the extraction of the respective components based on the different return periods. For the purpose, we proposed Discrete Wavelet Transform (DWT) as the methodology. The results from the DWT showed that the different components according to the respective periodicities could be extracted from the time series data for TOC and the variation of each component with respect to time could emerge from the return periods and the respective energy ratios of the decomposed components against the raw data.