• The KIPS Transactions:PartB
• /
• v.11B no.2
• /
• pp.133-140
• /
• 2004

Recently, the dental information systems were rapidly developed in order to store and process the data of patients. But, these systems should serve a doctor a good quality information against disease for diagnostic and surgery purpose so as to success in this field. This function of the system it important to persuade patients to undergo proper surgical operation they needed. Hence, 3D teeth model capable of simulating the dental surgery and treatment is necessary Teeth manipulation of dentistry is performed on individual tooth in dental clinic. io, 3D teeth reconstruction system should have the techniques of segmentation and 3D reconstruction adequate for individual tooth. In this paper, we propose the techniques of adaptive optimal segmentation to segment the individual area of tooth, and reconstruction method of tooth based on contour-based method. Each tooth can be segmented from neighboring teeth and alveolar bone in CT images using adaptive optimal threshold computed differently on tooth. Reconstruction of individual tooth using results of segmentation can be manipulated according to user's input and make the simulation of dental surgery and treatment possible.

• Journal of KIISE:Software and Applications
• /
• v.27 no.8
• /
• pp.815-826
• /
• 2000

This study is for the segmentation and volume calculation of the white matter and gray matter from brain MRI. In general, the volume of white and gray matter is reduced by contraction of each components in the case of mental retardation which are Alzheimer's disease and Down's syndrome. As results, it is useful for diagnostic and early detection for various mental retardation through the tracing of variation for its volume from the brain MRI. But, until now, it was very difficult to calculate the partial volume of each components existing in some thickness, because MR image was represented by single gray value after scanning by MR scanner. Accordingly, new segmentation algorithm proposed in this paper is to calculate the partial volume of the white and gray matter existing in some thickness through the analysis of the blurred gray value, and is to determine the threshold for segmentation of white and gray matter, and is to calculate the volume of each segmented component. And finally, proposed algorithm was applied the models which was created manually, and then acquired results was compared with that of original model.

• Korean Journal of Remote Sensing
• /
• v.34 no.6_2
• /
• pp.1239-1249
• /
• 2018

Optical remote sensing sensors provide visually more familiar images than radar images. However, it is difficult to discriminate sea ice types in optical images using spectral information based machine learning algorithms. This study addresses two topics. First, we propose a semantic segmentation which is a part of the state-of-the-art deep learning algorithms to identify ice types by learning hierarchical and spatial features of sea ice. Second, we propose a new approach by combining of semi-supervised and active learning to obtain accurate and meaningful labels from unlabeled or unseen images to improve the performance of supervised classification for multiple images. Therefore, we successfully added new labels from unlabeled data to automatically update the semantic segmentation model. This should be noted that an operational system to generate ice type products from optical remote sensing data may be possible in the near future.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2013.10a
• /
• pp.150-151
• /
• 2013

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.26 no.2
• /
• pp.28-36
• /
• 2022

As port structures are exposed to various extreme external loads such as wind (typhoons), sea waves, or collision with ships; it is important to evaluate the structural safety periodically. To monitor the port structure, especially the rubber fender, a fender segmentation system using a vision sensor and deep learning method has been proposed in this study. For fender segmentation, a new deep learning network that improves the encoder-decoder framework with the receptive field block convolution module inspired by the eccentric function of the human visual system into the DenseNet format has been proposed. In order to train the network, various fender images such as BP, V, cell, cylindrical, and tire-types have been collected, and the images are augmented by applying four augmentation methods such as elastic distortion, horizontal flip, color jitter, and affine transforms. The proposed algorithm has been trained and verified with the collected various types of fender images, and the performance results showed that the system precisely segmented in real time with high IoU rate (84%) and F1 score (90%) in comparison with the conventional segmentation model, VGG16 with U-net. The trained network has been applied to the real images taken at one port in Republic of Korea, and found that the fenders are segmented with high accuracy even with a small dataset.

• KIPS Transactions on Software and Data Engineering
• /
• v.12 no.5
• /
• pp.217-228
• /
• 2023

Video instance segmentation is an intelligent visual task with high complexity because it not only requires object instance segmentation for each image frame constituting a video, but also requires accurate tracking of instances throughout the frame sequence of the video. In special, human instance segmentation in drama videos has an unique characteristic that requires accurate tracking of several main characters interacting in various places and times. Also, it is also characterized by a kind of the class imbalance problem because there is a significant difference between the frequency of main characters and that of supporting or auxiliary characters in drama videos. In this paper, we introduce a new human instance datatset called MHIS, which is built upon drama videos, Miseang, and then propose a novel video data augmentation method, CDVA, in order to overcome the data imbalance problem between character classes. Different from the previous video data augmentation methods, the proposed CDVA generates more realistic augmented videos by deciding the optimal location within the background clip for a target human instance to be inserted with taking rich spatio-temporal context embedded in videos into account. Therefore, the proposed augmentation method, CDVA, can improve the performance of a deep neural network model for video instance segmentation. Conducting both quantitative and qualitative experiments using the MHIS dataset, we prove the usefulness and effectiveness of the proposed video data augmentation method.

• Journal of Korea Multimedia Society
• /
• v.13 no.12
• /
• pp.1786-1797
• /
• 2010

Automatic understanding of contents in document image is a very hard problem due to involvement with mathematically challenging problems originated mainly from the over-determined system induced by document segmentation process. In both academic and industrial areas, there have been incessant and various efforts to improve core parts of content retrieval technologies by the means of separating out segmentation related issues using semi-structured document, e.g., invoice,. In this paper we proposed classification models for text lines on invoice document in which text lines were clustered into the five categories in accordance with their contents: purchase order header, invoice header, summary header, surcharge header, purchase items. Our investigation was concentrated on the performance of machine learning based models in aspect of linear-discriminant-analysis (LDA) and non-LDA (logic based). In the group of LDA, na$\"{\i}$ve baysian, k-nearest neighbor, and SVM were used, in the group of non LDA, decision tree, random forest, and boost were used. We described the details of feature vector construction and the selection processes of the model and the parameter including training and validation. We also presented the experimental results of comparison on training/classification error levels for the models employed.

• Journal of Korea Multimedia Society
• /
• v.8 no.4
• /
• pp.492-501
• /
• 2005

In this paper, we proposed the new segmentation method of optic nerve head and optic cub to consider the depth of optic nerve head on stereo fundus image. We analyzed the error factor of stereo matching on stereo fundus image, and compensated them. For robust extraction of optic nerve head and optic cub, we proposed the modified active contour model to consider the 3D depth of optic nerve head. As experiment result to various stereo fundus images, we confirmed that proposed method can segment optic nerve head and optic cup effectively.

• 제어로봇시스템학회:학술대회논문집
• /
• 2005.06a
• /
• pp.877-881
• /
• 2005

In this paper, a novel method for analyzing a textile fabric structure is proposed to segment each yarn of the textile fabric from voxel data made out of its X-ray computed tomography (CT) images. In order to segment the each yarn, directions of fibers, of which yarn consists, are firstly estimated by correlating the voxel with a fiber model. Second, each fiber is reconstructed by clustering the voxel of the fiber using the estimated fiber direction as a similarity. Then, each yarn is reconstructed by clustering the reconstructed fibers using a distance which is newly defined as a dissimilarity. Consequently, each yarn of the textile fabric is segmented from the voxel data. The effectiveness of the proposed method is confirmed by experimentally applying the method to voxel data of a sample plain woven fabric, which is made of polyester two folded yarn. The each two folded yarn is correctly segmented by the proposed method.

• Journal of Institute of Control, Robotics and Systems
• /
• v.16 no.11
• /
• pp.1089-1095
• /
• 2010

We propose an image segmentation method for auto-teaching system of PCB (Printed Circuit Board) assembly inspection machines. The inspection machine acquires images of all components in PCB, and then compares each image with its standard image to find the assembly errors such as misalignment, inverse polarity, and tombstone. The component window that is the area of component to be acquired by camera, is one of the teaching data for operating the inspection machines. To reduce the teaching time of the machine, we newly develop the image processing method to extract the component window automatically from the image of PCB. The proposed method segments the component window by excluding the soldering parts as well as board background. We binarize the input image by use of HSI color model because it is difficult to discriminate the RGB colors between components and backgrounds. The linear combination of the binarized images then enhances the component window from the background. By use of the horizontal and vertical projection of histogram, we finally obtain the component widow. The experimental results are presented to verify the usefulness of the proposed method.