• Annual Conference of KIPS
• /
• 2023.05a
• /
• pp.485-487
• /
• 2023

컴퓨터 비전의 하위 태스크(Task)인 의미론적 분할(Semantic Segmentation)은 자율주행, 해상에서 선박찾기 등 다양한 분야에서 연구되고 있다. 기존 FCN(Fully Conovlutional Networks) 기반 의미론적 분할 모델은 다운샘플링(Dowsnsampling)과정에서 공간정보의 손실이 발생하여 정확도가 하락했다. 본 논문에서는 공간정보 손실을 완화하고자 PSA(Polarized Self-attention)의 공간정보 강조 모듈을 HRNet(High-resolution Networks)의 합성곱 블록 사이에 추가한다. 실험결과 파라미터는 3.1M, GFLOPs는 3.2G 증가했으나 mIoU는 0.26% 증가했다. 공간정보가 의미론적 분할 정확도에 영향이 미치는 것을 확인했다.

• ETRI Journal
• /
• v.32 no.6
• /
• pp.881-890
• /
• 2010

Conventional fire detection systems use physical sensors to detect fire. Chemical properties of particles in the air are acquired by sensors and are used by conventional fire detection systems to raise an alarm. However, this can also cause false alarms; for example, a person smoking in a room may trigger a typical fire alarm system. In order to manage false alarms of conventional fire detection systems, a computer vision-based fire detection algorithm is proposed in this paper. The proposed fire detection algorithm consists of two main parts: fire color modeling and motion detection. The algorithm can be used in parallel with conventional fire detection systems to reduce false alarms. It can also be deployed as a stand-alone system to detect fire by using video frames acquired through a video acquisition device. A novel fire color model is developed in CIE $L^*a^*b^*$ color space to identify fire pixels. The proposed fire color model is tested with ten diverse video sequences including different types of fire. The experimental results are quite encouraging in terms of correctly classifying fire pixels according to color information only. The overall fire detection system's performance is tested over a benchmark fire video database, and its performance is compared with the state-of-the-art fire detection method.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.26 no.2
• /
• pp.165-173
• /
• 2008

This study aims at generating automatically three dimensional geometric models of roads using LIDAR data and a digital map. The main processes in the proposed method are (1) generating a polygon encompassing a road region using a road layer from the digital map, (2) extracting LIDAR points within the road region using the polygon, (3) organizing the points into surface patches and grouping the patches into surface clusters, (4) searching the road surface clusters and generating the surface model from the points linked to the clusters, (5) refining the boundary using a digital map. By applying the proposed method to real data, we successfully generated the linear and surface information of the roads.

• The Korean Journal of Applied Statistics
• /
• v.22 no.1
• /
• pp.115-127
• /
• 2009

The purpose of statistical analyses of array-CGH experiment data is to divide the whole genome into regions of equal copy number, to quantify the copy number in each region and finally to evaluate its significance of being different from two. Several statistical procedures have been proposed which include the circular binary segmentation, and a Gaussian based local regression for detecting break points (GLAD) by estimating a piecewise constant function. We propose in this note a penalized spline regression and its simultaneous confidence band(SCB) approach to evaluate the statistical significance of regions of genetic gain/loss. The region of which the simultaneous confidence band stays above 0 or below 0 can be considered as a region of genetic gain or loss. We compare the performance of the SCB procedure with GLAD and hidden Markov model approaches through a simulation study in which the data were generated from AR(1) and AR(2) models to reflect spatial dependence of the array-CGH data in addition to the independence model. We found that the SCB method is more sensitive in detecting the low level copy number alterations.

• Proceedings of the KSRS Conference
• /
• 2005.10a
• /
• pp.241-244
• /
• 2005

Many building detection methods mainly rely on line segments extracted from aerial or satellite imagery. Building detection methods based on line segments, however, are difficult to succeed in high resolution satellite imagery such as IKONOS imagery, for most buildings in IKONOS imagery have small size of roofs with low contrast between roof and background. In this paper, we propose an efficient method to extract line segments and group them at the same time. First, edge preserving filtering is applied to the imagery to remove the noise. Second, we segment the imagery by watershed method, which collects the pixels with similar intensities to obtain homogeneous region. The boundaries of homogeneous region are not completely coincident with roof boundaries due to low contrast in the vicinity of the roof boundaries. Finally, to resolve this problem, we set up snake model with segmented region boundaries as initial snake's positions. We used a greedy algorithm to fit a snake to roof boundary. Experimental results show our method can obtain more .correct roof boundary with small size and low contrast from IKONOS imagery. Snake algorithm, building roof detection, watershed segmentation, edge-preserving filtering

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.24 no.10B
• /
• pp.1902-1911
• /
• 1999

In this paper we present a content-based image retrieval algorithm using the visual feature vectors which describe the spatial characteristics of objects. The proposed technique uses the Gaussian mixture model(GMM) to represent multi-colored objects and the expectation maximization(EM) algorithm is employed to estimate the maximum likelihood(ML) parameters of the model. After image segmentation is performed based on GMM, the shape and color features are extracted from each object using Fourier descriptors and color histograms, respectively. Image retrieval consists of two steps: first, the shape-based query is carried out to find the candidate images whose objects have the similar shapes with the query image and second, the color-based query is followed. The experimental results show that the proposed algorithm is effective in image retrieving by using the spatial and visual features of segmented objects.

• Journal of Korea Multimedia Society
• /
• v.22 no.12
• /
• pp.1385-1395
• /
• 2019

In this paper, we propose a method to fabricate a patient-specific breast implant using MRI images and 3D scan data. Existing breast implants for breast reconstruction surgery are primarily fabricated products for shaping, and among the limited types of implants, products similar to the patient's breast have been used. In fact, the larger the difference between the shape of the breast and the implant, the more frequent the postoperative side effects and the lower the satisfaction. Previous researches on the fabrication of patient-specific breast implants have used limited information based on only MRI images or on only 3D scan data. In this paper, we propose an algorithm for the fabrication of patient-specific breast implants that combines MRI images with 3D scan data, considering anatomical suitability for external shape, volume, and pectoral muscle. Experimental results show that we can produce precise breast implants using the proposed algorithm.

• Journal of Korea Multimedia Society
• /
• v.10 no.6
• /
• pp.716-725
• /
• 2007

In this paper we describe a motion tracking algorithm for 3D human animation using stereo vision system. This allows us to extract the motion data of the end effectors of human body by following the movement through segmentation process in HIS or RGB color model, and then blob analysis is used to detect robust shape. When two hands or two foots are crossed at any position and become disjointed, an adaptive algorithm is presented to recognize whether it is left or right one. And the real motion is the 3-D coordinate motion. A mono image data is a data of 2D coordinate. This data doesn't acquire distance from a camera. By stereo vision like human vision, we can acquire a data of 3D motion such as left, right motion from bottom and distance of objects from camera. This requests a depth value including x axis and y axis coordinate in mono image for transforming 3D coordinate. This depth value(z axis) is calculated by disparity of stereo vision by using only end-effectors of images. The position of the inner joints is calculated and 3D character can be visualized using inverse kinematics.

• The Transactions of the Korea Information Processing Society
• /
• v.5 no.10
• /
• pp.2713-2724
• /
• 1998

This paper proposes a new method for the parameter estimation in Markov Random Field(MRF) model of textured color images. The MRF models allow an image region to bel described using a finite number of parameters that characterize spatial interactionsl within and between bands of al color image. An important problem is estimation of the parameters since the randorn field model-based textured color image is the mostly parametric images of natural scenes to verify the validit of the proposed method proves that the method is not affected by the size of the image and shows well-segmented images.

• Culinary science and hospitality research
• /
• v.24 no.2
• /
• pp.112-118
• /
• 2018

The purpose of this study is to develop and test a model that explains the effect of experiential marketing factors on: 1) brand image, 2) satisfaction, and 3) loyalty in context of restaurants located in complex shopping mall. In addition, the study clarified how these variables relate to each other. Survey were distributed to customers who have visiting experience(s) in a restaurants in complex shopping mall. Total 360 participants were distributed and 344 questionnaires were used for analysing. The confirmatory factor analysis and structural equation modeling(SEM) have been employed research methods for frequency analysis, reliability analysis and measurement model validation. The findings of this study identified that relation factor of experiential marketing elements was only significant factor on brand image Furthermore, sense and recognition were critical components of customer satisfaction. Last, present study also identified the significant relationship between satisfaction and customer loyalty. These findings may contribute to provide valuable marketing strategic for this business segmentation, and it can be utilized as a fundamental study to establish an efficient business plan to increase revenue especially for restaurant business located in complex shopping mall.