• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.48 no.2
• /
• pp.11-23
• /
• 2011

In this paper, an accurate and fully automatic video object segmentation method is proposed for video conferencing systems in which the real-time performance is required. The proposed method consists of two steps: 1) accurate object extraction on the initial frame, 2) real-time object extraction from the next frame using the result of the first step. Object extraction on the initial frame starts with generating a cumulative edge map obtained from frame differences in the beginning. This is because we can estimate the initial shape of the foreground object from the cumulative motion. This estimated shape is used to assign the seeds for both object and background, which are needed for Graph-Cut segmentation. Once the foreground object is extracted by Graph-Cut segmentation, real-time object extraction is conducted using the extracted object and the double edge map obtained from the difference between two successive frames. Experimental results show that the proposed method is suitable for real-time processing even in VGA resolution videos contrary to previous methods, being a useful tool for immersive video conferencing systems.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.18 no.3
• /
• pp.687-696
• /
• 2014

This paper presents a semantic object extraction method using user's stroke input, color, and depth information. It is supposed that a semantically meaningful object is surrounded with a few strokes from a user, and has similar depths all over the object. In the proposed method, deciding the region of interest (ROI) is based on the stroke input, and the semantically meaningful object is extracted by using color and depth information. Specifically, the proposed method consists of two steps. The first step is over-segmentation inside the ROI using color and depth information. The second step is semantically meaningful object extraction where over-segmented regions are classified into the object region and the background region according to the depth of each region. In the over-segmentation step, we propose a new marker extraction method where there are two propositions, i.e. an adaptive thresholding scheme to maximize the number of the segmented regions and an adaptive weighting scheme for color and depth components in computation of the morphological gradients that is required in the marker extraction. In the semantically meaningful object extraction, we classify over-segmented regions into the object region and the background region in order of the boundary regions to the inner regions, the average depth of each region being compared to the average depth of all regions classified into the object region. In experimental results, we demonstrate that the proposed method yields reasonable object extraction results.

• Proceedings of the KSRS Conference
• /
• v.2
• /
• pp.610-614
• /
• 2006

In this paper the advanced method of geodesic active contours was developed for the task of building detection from aerial and satellite images. Automatic extraction of man-made structures including buildings, building blocks or roads from remote sensing data is useful for land use mapping, scene understanding, robotic navigation, image retrieval, surveillance, emergency management procedures, cadastral etc. A level set method based on a region-driven segmentation model was implemented with which building boundaries were detected, through this curve propagation technique. The essence of this approach is to optimize the position and the geometric form of the curve by measuring information along that curve, and within the regions that compose the image partition. To this end, one can consider uniform intensities inside objects and the background. Thus, given an initial position of the curve, one can determine global, region-driven functions and provide a statistical description of the inside and outside object area. The calculus of variations and a gradient descent method was used to optimize the variational functional by an iterative steady state process. Experimental results demonstrate the potential of the proposed processing scheme.

• Journal of Information Processing Systems
• /
• v.9 no.1
• /
• pp.117-140
• /
• 2013

Hindi is the most widely spoken language in India, with more than 300 million speakers. As there is no separation between the characters of texts written in Hindi as there is in English, the Optical Character Recognition (OCR) systems developed for the Hindi language carry a very poor recognition rate. In this paper we propose an OCR for printed Hindi text in Devanagari script, using Artificial Neural Network (ANN), which improves its efficiency. One of the major reasons for the poor recognition rate is error in character segmentation. The presence of touching characters in the scanned documents further complicates the segmentation process, creating a major problem when designing an effective character segmentation technique. Preprocessing, character segmentation, feature extraction, and finally, classification and recognition are the major steps which are followed by a general OCR. The preprocessing tasks considered in the paper are conversion of gray scaled images to binary images, image rectification, and segmentation of the document's textual contents into paragraphs, lines, words, and then at the level of basic symbols. The basic symbols, obtained as the fundamental unit from the segmentation process, are recognized by the neural classifier. In this work, three feature extraction techniques-: histogram of projection based on mean distance, histogram of projection based on pixel value, and vertical zero crossing, have been used to improve the rate of recognition. These feature extraction techniques are powerful enough to extract features of even distorted characters/symbols. For development of the neural classifier, a back-propagation neural network with two hidden layers is used. The classifier is trained and tested for printed Hindi texts. A performance of approximately 90% correct recognition rate is achieved.

• Progress in Medical Physics
• /
• v.15 no.2
• /
• pp.112-119
• /
• 2004

We tried to build human models based on medical images of live Korean, instead of using standard data of human body structures. Characteristics of mechanical structures of human bodies were obtained from medical images such as CT and MR images. For each constitutional part of mechanical structures CT images were analyzed in terms of gray levels and MR images were analyzed in terms of pulse sequence. Characteristic features of various mechanical structures were extracted from the analyses. Based on the characteristics of each structuring element we peformed image segmentation on CT and MR images. We delineated bones, muscles, ligaments and tendons from CT and MR images using image segmentation or manual drawing. For the image segmentation we compared the edge detection method, region growing method and intensity threshold method and applied an optimal compound of these methods for the best segmentation results. Segmented mechanical structures of the head/neck part were three dimensionally reconstructed.

• Journal of the Korean Society for Industrial and Applied Mathematics
• /
• v.11 no.2
• /
• pp.41-47
• /
• 2007

A fine segmentation algorithm is proposed for extracting objects in an image, which have both weak boundaries and highly non-convex shapes. The image has simple background colors or simple object colors. Two concepts, geometric attraction-driven flow (GADF) and edge-regions are combined to detect boundaries of objects in a sub-pixel resolution. The main strategy to segment the boundaries is to construct initial curves close to objects by using edge-regions and then to make a curve evolution in GADF. Since the initial curves are close to objects regardless of shapes, highly non-convex shapes are easily detected and dependence on initial curves in boundary-based segmentation algorithms is naturally removed. Weak boundaries are also detected because the orientation of GADF is obtained regardless of the strength of boundaries. For a fine segmentation, we additionally propose a local region competition algorithm to detect perceptible boundaries which are used for the extraction of objects without visual loss of detailed shapes. We have successfully accomplished the fine segmentation of objects from images taken in the studio and aphids from images of soybean leaves.

• Asia-pacific Journal of Multimedia Services Convergent with Art, Humanities, and Sociology
• /
• v.7 no.3
• /
• pp.383-391
• /
• 2017

In image processing, image segmentation has been studied by various methods in a long time. Image segmentation is the process of partitioning a digital image into multiple objects and face detection is a typical image segmentation field being used in a variety of applications that identifies human faces in digital images. In this paper, we propose a method for extracting the contours of faces included in images. Using the Viola-Jones algorithm, to do this, we detect the approximate locations of faces from images. But, the Viola-Jones algorithm could detected the approximate location of face not the correct position. In order to extract a more accurate face region from image, we use skin color in this paper. In details, face region would be extracted using the analysis of horizontal and vertical histograms on the skin area. Finally, the face contour is extracted using snake algorithm for the extracted face area. In this paperr, a modified snake energy function is proposed for face contour extraction based snake algorithm proposed by Williams et al.[7]

• Journal of Advanced Navigation Technology
• /
• v.24 no.2
• /
• pp.148-154
• /
• 2020

Wrinkles are one of the main features of skin aging. Conventional image processing-based wrinkle detection is difficult to effectively cope with various skin images. In particular, Wrinkle extraction performance is significantly decreased when the wrinkles are not strong and similar to the surrounding skin. In this paper, deep learning is applied to extract wrinkles from microscopic skin images. In general, the microscope image is equipped with a wide-angle lens, so the brightness at the boundary area of the image is dark. In this paper, to solve this problem, the brightness of the skin image is estimated and corrected. In addition, We apply the structure of semantic segmentation network suitable for wrinkle extraction. The proposed method obtained an accuracy of 99.6% in test experiments on skin images collected in our laboratory.

• 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.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.27 no.1
• /
• pp.723-733
• /
• 2009

This study proposes a sequence of procedures to extract building boundaries and planar patches through segmentation of rasterized lidar data. Although previous approaches to building extraction have been shown satisfactory, there still exist needs to increase the degree of automation. The methodologies proposed in this study are as follows: Firstly, lidar data are rasterized into grid form in order to exploit its rapid access to neighboring elevations and image operations. Secondly, propagation of errors in raw data is taken into account for in assessing the quality of gradients-driven properties and further in choosing suitable parameters. Thirdly, extraction of planar patches is conducted through a sequence of processes: histogram analysis, least squares fitting, and region merging. Experimental results show that the geometric components of building models could be extracted by the proposed approach in a streamlined way.