• International Journal of Contents
• /
• 제3권2호
• /
• pp.25-29
• /
• 2007

In this paper, we discuss the mesh segmentation problem which divides a given 3D mesh into several disjoint sets. To solve the problem, we propose a greedy method based on the merging priority metric defined for representing the geometric properties of meaningful parts. The proposed priority metric is a weighted function using five geometric parameters, those are, a distribution of Gaussian map, boundary path concavity, boundary path length, cardinality, and segmentation resolution. In special, we can control by setting up the weight values of the proposed geometric parameters to obtain visually better mesh segmentation. Finally, we carry out an experiment on several 3D mesh models using the proposed methods and visualize the results.

• 정보처리학회논문지:소프트웨어 및 데이터공학
• /
• 제12권12호
• /
• pp.505-518
• /
• 2023

3차원 포인트 클라우드 의미적 분할은 각 포인트별로 해당 포인트가 속한 물체나 영역의 분류 레이블을 예측함으로써, 포인트 클라우드를 서로 다른 물체들이나 영역들로 나누는 컴퓨터 비전 작업이다. 기존의 3차원 의미적 분할 모델들은 RGB 영상들에서 추출하는 2차원 시각적 특징과 포인트 클라우드에서 추출하는 3차원 기하학적 특징의 특성을 충분히 고려한 특징 융합을 수행하지 못한다는 한계가 있다. 따라서, 본 논문에서는 2차원-3차원 멀티-모달 특징을 이용하는 새로운 3차원 의미적 분할 모델 MMCA-Net을 제안한다. 제안 모델은 중기 융합 전략과 멀티-모달 교차 주의집중 기반의 융합 연산을 적용함으로써, 이질적인 2차원 시각적 특징과 3차원 기하학적 특징을 효과적으로 융합한다. 또한 3차원 기하학적 인코더로 PTv2를 채용함으로써, 포인트들이 비-정규적으로 분포한 입력 포인트 클라우드로부터 맥락정보가 풍부한 3차원 기하학적 특징을 추출해낸다. 본 논문에서는 제안 모델의 성능을 분석하기 위해 벤치마크 데이터 집합인 ScanNetv2을 이용한 다양한 정량 및 정성 실험들을 진행하였다. 성능 척도 mIoU 측면에서 제안 모델은 3차원 기하학적 특징만을 이용하는 PTv2 모델에 비해 9.2%의 성능 향상을, 2차원-3차원 멀티-모달 특징을 사용하는 MVPNet 모델에 비해 12.12%의 성능 향상을 보였다. 이를 통해 본 논문에서 제안한 모델의 효과와 유용성을 입증하였다.

• 대한원격탐사학회:학술대회논문집
• /
• 대한원격탐사학회 2002년도 Proceedings of International Symposium on Remote Sensing
• /
• pp.59-64
• /
• 2002

This paper presents a method for line segment extraction for 3-d building reconstruction. Building roofs are described as a set of planar polygonal patches, each of which is extracted by watershed-based image segmentation, line segment matching and coplanar grouping. Coplanar grouping and polygonal patch formation are performed per region by selecting 3-d line segments that are matched using epipolar geometry and flight information. The algorithm has been applied to high resolution aerial images and the results show accurate 3-d building reconstruction.

• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2019년도 춘계학술대회
• /
• pp.432-435
• /
• 2019

Convolutional Neural Networks (CNNs) have recently been gaining popularity in the medical image analysis field because of their image segmentation capabilities. In this paper, we present a CNN that performs automated brain tumor segmentations of sparsely annotated 3D Magnetic Resonance Imaging (MRI) scans. Our CNN is based on 3D U-net architecture, and it includes separate Dilated and Depth-wise Convolutions. It is fully-trained on the BraTS 2018 data set, and it produces more accurate results even when compared to the winners of the BraTS 2017 competition despite having a significantly smaller amount of parameters.

• 한국멀티미디어학회논문지
• /
• 제3권4호
• /
• pp.339-346
• /
• 2000

의료 영상은 환자에 대한 해부학적인 진단 정보를 얻기 위한 영상으로 정확한 병변 인식과 판단을 위해서는 조직별 분할이 선행되어야 한다. 본 논문에서는 T1 강조 영상 그리고 T2 강조 영상, PD 영상의 특징을 상호보완적으로 이용한 자동적인 영상 분할 방법을 제안한다. 제안한 분할 알고리듬은 PD 영상으로부터 대뇌마스크를 획득하고, 대뇌마스크를 T1 과 T2, PD의 입력 영상에 씌워 각각의 대뇌 영상을 획득하여 T1과 T2, PD를 축으로 하는 3차원 공간상에서 스케일 스페이스 필터링과, 3차원 클러스터링을 이용하여 대뇌 내부조직에 해당하는 클러스터를 찾아서 분할에 이용한다. 대뇌 영상분할은 이들 클러스터의 중심 값을 FCM 알고리듬의 초기 중심 값으로 두고 FCM 알고리듬을 이용하여 분할한다. 제안한 분할 알고리듬은 정확한 클러스터의 중심 값을 계산함으로 초기 값의 영향을 많이 받는 FCM 알고리듬의 단점을 보완하였고 다중 스펙트럼 영상의 특성을 조합하여 분할에 이용함으로 단일 스펙트럼 영상만을 이용하는 방법보다 향상된 분할 결과를 얻을 수 있었다.

• 대한의용생체공학회:의공학회지
• /
• 제43권5호
• /
• pp.341-352
• /
• 2022

Medical image segmentation is the most important task in radiation therapy. Especially, when segmenting medical images, the liver is one of the most difficult organs to segment because it has various shapes and is close to other organs. Therefore, automatic segmentation of the liver in computed tomography (CT) images is a difficult task. Since tumors also have low contrast in surrounding tissues, and the shape, location, size, and number of tumors vary from patient to patient, accurate tumor segmentation takes a long time. In this study, we propose a method algorithm for automatically segmenting the liver and tumor for this purpose. As an advantage of setting the boundaries of the tumor, the liver and tumor were automatically segmented from the CT image using the 2D CoordConv DeepLab V3+ model using the CoordConv layer. For tumors, only cropped liver images were used to improve accuracy. Additionally, to increase the segmentation accuracy, augmentation, preprocess, loss function, and hyperparameter were used to find optimal values. We compared the CoordConv DeepLab v3+ model using the CoordConv layer and the DeepLab V3+ model without the CoordConv layer to determine whether they affected the segmentation accuracy. The data sets used included 131 hepatic tumor segmentation (LiTS) challenge data sets (100 train sets, 16 validation sets, and 15 test sets). Additional learned data were tested using 15 clinical data from Seoul St. Mary's Hospital. The evaluation was compared with the study results learned with a two-dimensional deep learning-based model. Dice values without the CoordConv layer achieved 0.965 ± 0.01 for liver segmentation and 0.925 ± 0.04 for tumor segmentation using the LiTS data set. Results from the clinical data set achieved 0.927 ± 0.02 for liver division and 0.903 ± 0.05 for tumor division. The dice values using the CoordConv layer achieved 0.989 ± 0.02 for liver segmentation and 0.937 ± 0.07 for tumor segmentation using the LiTS data set. Results from the clinical data set achieved 0.944 ± 0.02 for liver division and 0.916 ± 0.18 for tumor division. The use of CoordConv layers improves the segmentation accuracy. The highest of the most recently published values were 0.960 and 0.749 for liver and tumor division, respectively. However, better performance was achieved with 0.989 and 0.937 results for liver and tumor, which would have been used with the algorithm proposed in this study. The algorithm proposed in this study can play a useful role in treatment planning by improving contouring accuracy and reducing time when segmentation evaluation of liver and tumor is performed. And accurate identification of liver anatomy in medical imaging applications, such as surgical planning, as well as radiotherapy, which can leverage the findings of this study, can help clinical evaluation of the risks and benefits of liver intervention.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 2001년도 춘계학술대회 논문집
• /
• pp.435-438
• /
• 2001

Reverse engineering technology refers to the process that creates a CAD model of an existing part using measuring devices. Recently, non-contact scanning devices have become more accurate and the speed of data acquisition has increased drastically. However, they generate thousands of points per second and various types of point data. Therefore, it becomes a major issue to handle the huge amount and various types of point data. To generate a CAD model from scanned point data efficiently, these point data should be well arranged through point data handling processes such as data reduction and segmentation. This paper proposes a new point data handling method using 3D grids. The geometric information of a part is extracted from point cloud data by estimating normal values of the points. The non-uniform 3D grids for data reduction and segmentation are generated based on the geometric information. Through these data reduction and segmentation processes, it is possible to create CAD models autmatically and efficiently. The proposed method is applied to two quardric medels and the results are discussed.

• Journal of International Society for Simulation Surgery
• /
• 제3권1호
• /
• pp.28-32
• /
• 2016

Background Congenital muscular torticollis (CMT) is a benign condition. With early diagnosis and appropriate management, it can be cured completely, leaving no residual deformity. However, long-standing, untreated CMT can lead to permanent craniofacial deformities and asymmetry.Methods Nineteen patients presented to the author with congenital muscular torticollis. Three dimensional computed tomography (3-D CT) scans was obtained upon patient’s admission. Adjustment of skull’s position to Frankfort horizontal plan was done. Cervical vertebral segmentation was done which allowed a 3D module to be separately created for each vertebra to detect any anatomical or positional changes.Results The segmented vertebrae showed an apparent anatomical changes, which were most noticeable at the level of the atlas and axis vertebrae. These changes decreased gradually till reaching the seventh cervical vertebra, which appeared to be normal in all patients. The changes in the atlas vertebra were mostly due to its intimate relation with the skull base, while the changes of the axis were the most significantConclusion Cervical vertebral segmentation is a reliable tool for isolation and studying cervical vertebral pathological changes of each vertebra separately. The accuracy of the procedures in addition to the availability of many software that can be used for segmentation will allow many surgeons to use segmentation of the vertebrae for diagnosis and even for preoperative simulation planning.

• Smart Structures and Systems
• /
• 제30권5호
• /
• pp.501-511
• /
• 2022

Three-dimensional (3D) models have become crucial for improving civil infrastructure analysis, and they can be used for various purposes such as damage detection, risk estimation, resolving potential safety issues, alarm detection, and structural health monitoring. 3D point cloud data is used not only to make visual models but also to analyze the states of structures and to monitor them using semantic data. This study proposes automating the generation of high-quality 3D point cloud data and removing noise using deep learning algorithms. In this study, large-format aerial images of civilian infrastructure, such as cut slopes and dams, which were captured by drones, were used to develop a workflow for automatically generating a 3D point cloud model. Through image cropping, downscaling/upscaling, semantic segmentation, generation of segmentation masks, and implementation of region extraction algorithms, the generation of the point cloud was automated. Compared with the method wherein the point cloud model is generated from raw images, our method could effectively improve the quality of the model, remove noise, and reduce the processing time. The results showed that the size of the 3D point cloud model created using the proposed method was significantly reduced; the number of points was reduced by 20-50%, and distant points were recognized as noise. This method can be applied to the automatic generation of high-quality 3D point cloud models of civil infrastructures using aerial imagery.

• 대한의용생체공학회:의공학회지
• /
• 제24권4호
• /
• pp.247-257
• /
• 2003

This paper proposes a three-dimensional (3D) segmentation algorithm for extracting a diagnostic object from ultrasound images by using a LoG operator In the proposed algorithm, 2D cutting planes are first obtained by the equiangular revolution of a cross sectional Plane on a reference axis for a 3D volume data. In each 2D ultrasound image. a region of interest (ROI) box that is included tightly in a diagnostic object of interest is set. Inside the ROI box, a LoG operator, where the value of $\sigma$ is adaptively selected by the distance between reference points and the variance of the 2D image, extracts edges in the 2D image. In Post processing. regions of the edge image are found out by region filling, small regions in the region filled image are removed. and the contour image of the object is obtained by morphological opening finally. a 3D volume of the diagnostic object is rendered from the set of contour images obtained by post-processing. Experimental results for a tumor and gall bladder volume data show that the proposed method yields on average two times reduction in error rate over Krivanek's method when the results obtained manually are used as a reference data.