• Journal of the Korea Computer Graphics Society
• /
• v.29 no.1
• /
• pp.1-11
• /
• 2023

Deep-learning-based image segmentation is one of the most widely employed lane detection approaches, and it requires a post-process for extracting the key points on the lanes. A general approach for key-point extraction is using a fixed threshold defined by a user. However, finding the best threshold is a manual process requiring much effort, and the best one can differ depending on the target data set (or an image). We propose a novel key-point extraction algorithm that automatically adapts to the target image without any manual threshold setting. In our adaptive key-point extraction algorithm, we propose a line-level normalization method to distinguish the lane region from the background clearly. Then, we extract a representative key point for each lane at a line (row of an image) using a kernel density estimation. To check the benefits of our approach, we applied our method to two lane-detection data sets, including TuSimple and CULane. As a result, our method achieved up to 1.80%p and 17.27% better results than using a fixed threshold in the perspectives of accuracy and distance error between the ground truth key-point and the predicted point.

• Smart Structures and Systems
• /
• v.33 no.6
• /
• pp.399-414
• /
• 2024

This study discusses the long-term deformation monitoring and shape sensing of bridge girder surfaces with an automated extraction scheme for point clouds in the Region Of Interest (ROI), invariant to the position of a Light Detection And Ranging system (LiDAR). Advanced smart construction necessitates continuous monitoring of the deformation and shape of bridge girders during the construction phase. An automated scheme is proposed for reconstructing geometric model of ROI in the presence of noisy non-stationary background. The proposed scheme involves (i) denoising irrelevant background point clouds using dimensions from the design model, (ii) extracting the outer boundaries of the bridge girder by transforming and processing the point cloud data in a two-dimensional image space, (iii) extracting topology of pre-defined targets using the modified Otsu method, (iv) registering the point clouds to a common reference frame or design coordinate using extracted predefined targets placed outside ROI, and (v) defining the bounding box in the point clouds using corresponding dimensional information of the bridge girder and abutments from the design model. The surface-fitted reconstructed geometric model in the ROI is superposed consistently over a long period to monitor bridge shape and derive deflection during the construction phase, which is highly correlated. The proposed scheme of combining 2D-3D with the design model overcomes the sensitivity of 3D point cloud registration to initial match, which often leads to a local extremum.

• Korean Journal of Optics and Photonics
• /
• v.14 no.4
• /
• pp.413-422
• /
• 2003

We have calculated the power extraction efficiency and the lasing wavelength distribution of index-coupled DFB lasers at threshold for various kL and facet reflectivity combinations, and compared with those above-threshold. The power extraction efficiency increases as the asymmetry of the facet reflectivities increases. The power extraction efficiency above-threshold is slightly larger than that at threshold. Since the relative photon density around the center region increases as kL increases, the power extraction efficiency decreases. The uniformity of the distribution of lasing wavelength over the stop band increases due to the relief of mode degeneracy as the asymmetry of the facet reflectivities increases. In the case of AR-HR combination, the lasing wavelength distributions at threshold are similar to those above-threshold. However, in the case of AR-AR combination, the lasing wavelength at threshold is concentrated on both edges of the stop band, while it is concentrated only on the longer wavelength edge above-threshold. As kL increases, the range of the lasing wavelength distribution increases due to the increase of the stop band. The effect of AR reflectivity on the power extraction and the lasing wavelength distribution is very weak.

• Maxillofacial Plastic and Reconstructive Surgery
• /
• v.41
• /
• pp.4.1-4.10
• /
• 2019

Background: The mandibular third molar (M3) is typically the last permanent tooth to erupt because of insufficient space and thick soft tissues covering its surface. Problems such as alveolar bone loss, development of a periodontal pocket, exposure of cementum, gingival recession, and dental caries can be found in the adjacent second molars (M2) following M3 extraction. The specific aims of the study were to assess the amount and rate of bone regeneration on the distal surface of M2 and to evaluate the aspects of bone regeneration in terms of varying degree of impaction. Methods: Four series of panoramic radiographic images were obtained from the selected cases, including images from the first visit, immediately after extraction, 6 weeks, and 6 months after extraction. ImageJ software^® (NIH, USA) was used to measure linear distance from the region of interest to the distal root of the adjacent M2. Radiographic infrabony defect (RID) values were calculated from the measured radiographic bone height and cementoenamel junction with distortion compensation. Repeated measures of analysis of variance and one-way analysis of variance were conducted to analyze the statistical significant difference between RID and time, and a Spearman correlation test was conducted to assess the relationship between Pederson's difficulty index (DI) and RID. Results: A large RID (> 6 mm) can be reduced gradually and consistently over time. More than half of the samples recovered nearly to their normal healthy condition (RID ≤ 3 mm) by the 6-month follow-up. DI affected the first 6 weeks of post-extraction period and only showed a significant positive correlation with respect to the difference between baseline and final RID. Conclusions: Additional treatments on M2 for a minimum of 6 months after an M3 extraction could be recommended. Although DI may affect bone regeneration during the early healing period, further study is required to elucidate any possible factors associated with the healing process. The DI does not cause any long-term adverse effects on bone regeneration after surgical extraction.

• Journal of KIISE:Software and Applications
• /
• v.27 no.7
• /
• pp.731-741
• /
• 2000

This paper proposes a method to detect a facial region and extract facial features which is crucial for visual recognition of human faces. In this paper, we extract the MER(Minimum Enclosing Rectangle) of a face and facial components using projection analysis on both edge image and binary image. We use an active contour model(snakes) for extraction of the contours of eye, mouth, eyebrow, and face in order to reflect the individual differences of facial shapes and converge quickly. The determination of initial contour is very important for the performance of snakes. Particularly, we detect Minimum Enclosing Rectangle(MER) of facial components and then determine initial contours using general shape of facial components within the boundary of the obtained MER. We obtained experimental results to show that MER extraction of the eye, mouth, and face was performed successfully. But in the case of images with bright eyebrow, MER extraction of eyebrow was performed poorly. We obtained good contour extraction with the individual differences of facial shapes. Particularly, in the eye contour extraction, we combined edges by first order derivative operator and zero crossings by second order derivative operator in designing energy function of snakes, and we achieved good eye contours. For the face contour extraction, we used both edges and grey level intensity of pixels in designing of energy function. Good face contours were extracted as well.

• Journal of Oral Medicine and Pain
• /
• v.32 no.3
• /
• pp.263-272
• /
• 2007

The purpose of this study was to analyze the nerve damage after tooth extraction and implant surgery, and to establish a predictive model for assessment and management of dysesthesia. In this questionnaire study, the subjects chosen for this study were 276 dentists who answered the questionnaire about dysesthesia after tooth extraction and implant surgery. The analysis of the results consist of the sex and age distribution, affected site, associated symptoms, rate and duration of the recovery. The results are summarized as follows. : 1. There were no significant difference between the sex and the dysesthesia. 2. The most common affected site was the mandibular region. In the group of the implant surgery, 100% affected the mandibular site. The tooth extraction group was 93.2% affected. 3. Pain was one of the most associated symptom with dysesthesia-46.5% of the tooth extraction and 44.8% of the implant surgery. 4. The recovery ratio was 72.3% in the tooth extraction, 71.8% in the implant surgery. Most of them, they recovered in $1{\sim}6$ months. In conclusion, most of dysesthesia may be recovered within 1 year. However, the possibility of persistent dysesthesia should not be neglected. Therefore, practitioners must discuss the possibility of nerve injury with their patients, and include this possibility in the consent forms. Various methods of monitoring recovery of sensation should be considered for objective assessment of prognosis. In addition, immediate referral to orofacial pain specialists can offer the patients an opportunity for more effective and noninvasive treatments.

• Microbiology and Biotechnology Letters
• /
• v.51 no.1
• /
• pp.99-108
• /
• 2023

Several challenges arise in DNA extraction and gene amplification for airborne fungal metagenome analysis from a particulate matter (PM) samples. In this study, various conditions were tested to optimize the DNA extraction method from PM samples and polymerase chain reaction (PCR) conditions with primer set and annealing temperature. As a result of comparative evaluation of DNA extraction under various conditions, chemical cell lysis using buffer and proteinase K for 20 minutes and bead beating treatment were followed by using a commercial DNA extraction kit to efficiently extract DNA from the PM filter samples. To optimize the PCR conditions, PCR was performed using 10 primer sets for amplifying the ITS2 gene region. The concentration of the PCR amplicon was relatively high when the annealing temperature was 58℃ with the ITS3tagmix3/ITS4 primer set. Even under these conditions, when the concentration of the PCR product was low, nested PCR was performed using the primary PCR amplicon as the template DNA to amplify the ITS2 gene at a satisfactory concentration. Using the methods optimized in this study, DNA extraction and PCR were performed on 15 filter samples that collected PM2.5 in Seoul, and the ITS2 gene was successfully amplified in all samples. The optimized methods can be used for research on analyzing and interpreting the fungal metagenome of atmospheric PM samples.

• The KIPS Transactions:PartB
• /
• v.16B no.4
• /
• pp.257-262
• /
• 2009

A method of the region extraction and recognition of a speed sign in the real road environment is proposed. The region of speed sign is extracted by using color information and then numbers are segmented in the region. We improve the recognition rate by performing an incline compensation of the speed sign for directions clockwise and counterclockwise. In image sequences of the real road environment, a robust recognition results are achieved with speed signs at normal condition as well as inclined.

• Spatial Information Research
• /
• v.10 no.3
• /
• pp.439-453
• /
• 2002

Researches using DEM are actively progressed in field of water resource, and especially the extraction of watershed and stream based on topographic and hydrologic analysis, is widely used in GIS analysis recently. However, because it is extracted that watershed which is type of sliver polygon and stream which is type of straight line in low topographic region, they're not efficient for application. This study determined buffer zone from stream and tried smooth elevation change to buffer zone in order to efficiently extract abnormal watershed and stream being occurred in low topographic region. Especially, we applied equal-distance assignment model using Voronoi Diagram to determine smooth elevation change. Also, we extracted watershed and stream using stream recognition DEM and origional DEM, and evaluated the efficiency of research through comparing the shape of watershed and stream in low topographic region.

• Journal of Intelligence and Information Systems
• /
• v.8 no.1
• /
• pp.15-25
• /
• 2002

This paper proposes a vehicle license plate recognition system, which has relatively a simple structure and is highly tolerant of noise, by using the DCT(Discrete Cosine Transform) coefficients extracted from the character region of a license plate and the LVQ(Learning Vector Quantization) neural network. The image of a license plate is taken from a captured vehicle image based on RGB color information, and the character region is derived by the histogram of the license plate and the relative position of individual characters in the plate. The feature vector obtained by the DCT of extracted character region is utilized as an input to the LVQ neural classifier fur the recognition process. In the experiment, 109 vehicle images captured under various types of circumstances were tested with the proposed method, and the relatively high extraction rate of license plates and recognition rate were achieved.