• Journal of the Korean Institute of Telematics and Electronics
• /
• v.20 no.1
• /
• pp.7-14
• /
• 1983

Linear feature such as lines and curves are one of important features in image processing. In this paper, new method of linear feature detection is suggested. Also, we have studied approximation technique which transforms detected linear feature into data structure for the practical. This method is based on graph theory and principle of this method is based on minimal spanning tree concept which is widely used in edge linking process. By postprocessing, Hairs and inconsistent line segments are removed. To approximate and describe traced linear feature, piecewise linear approximation is adapted. The algorithm is demonstrated through computer simulations.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2011.11a
• /
• pp.150-153
• /
• 2011

중간시점 영상은 스테레오 정합 방식을 이용하여 구한 깊이 지도를 이용하여 생성하는 것이 일반적인 방법이다. 그러나 대부분의 스테레오 정합 방식들은 좌, 우 영상의 조명환경이나 기하학적으로 수평 또는 수직방향이 일치하지 않으면 잘못된 깊이 지도를 획득하는 단점이 있다. 이러한 단점들은 정합 과정을 통해 획득한 깊이 지도를 이용하여 다시점 영상 생성 시 더 많은 홀과 경계 잡음을 생성하게 된다. 이러한 문제점을 보완하기 위하여 본 논문에서는 RGB 컬러 카메라 1 대와 깊이 카메라를 이용하여 중간 영상을 생성하는 방법을 제안한다. 제안된 기법을 이용하여 화상회의 시 사실감 및 현실감을 증대할 수 있는 eye-contact 시점 영상을 생성하고 이때 발생하는 홀과 경계잡음 제거를 실시간으로 처리하기 위한 효율적인 기법을 제안한다.

• Smart Media Journal
• /
• v.1 no.3
• /
• pp.36-42
• /
• 2012

Recently, the smart applications, such as smart phone and smart TV, become a hot issue in IT consumer markets. In particular, the smart TV provides 3D video services, hence efficient coding methods for 3D video data are required. Three-dimensional (3D) video involves stereoscopic or multi-view images to provide depth experience through 3D display systems. Binocular cues are perceived by rendering proper viewpoint images obtained at slightly different view angles. Since the number of viewpoints of the multi-view video is limited, 3D display devices should generate arbitrary viewpoint images using available adjacent view images. In this paper, after we explain a view synthesis method briefly, we propose a new algorithm to compensate view synthesis errors around object boundaries. We describe a 3D warping technique exploiting the depth map for viewpoint shifting and a hole filling method using multi-view images. Then, we propose an algorithm to remove boundary noises that are generated due to mismatches of object edges in the color and depth images. The proposed method reduces annoying boundary noises near object edges by replacing erroneous textures with alternative textures from the other reference image. Using the proposed method, we can generate perceptually inproved images for 3D video systems.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.30 no.11C
• /
• pp.1076-1082
• /
• 2005

Recently, there have been much efforts in the image segmentation using morphological approach. Among them, the watershed algorithm is one of powerful tools which can take advantages of both of the conventional edge-based segmentation and region-based segmentation. The concept of watershed is based on topographic analogy. But, its high sensitivity to noise yields a very large number of resulting segmented regions which leads to oversegmentation. So we suggest the restricted waterfall algorithm which reduce the oversegmentation by eliminate not only local minima but also local maxima. As a result, the restricted waterfall algorithm has a good segmented image than the other methods, and has a better binary image than the histogram thresholding method.

• Journal of the Korea Computer Graphics Society
• /
• v.13 no.3
• /
• pp.7-10
• /
• 2007

We propose an image filtering technique that removes various image styles. To destylize a given image, we define image styles as repeated patterns existing in the image. For dll pixels of the image, we compute image styles as style vectors. We remove image styles by using bilateral filtering based on these style vectors. Destylization results show well smoothed images while preserving feature boundaries. Our method effectively removes image styles and reveals image structures clearly, and results can be applied to several applications such as texture transfer.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2016.11a
• /
• pp.110-113
• /
• 2016

본 논문에서는 카메라로 촬영한 동영상에서 키 프레임을 추출하고 특징점을 기반으로 영상을 정합하는 파노라마 영상 생성 기법을 제안한다. 제안한 기법에서는 다양한 동영상의 히스토그램, 에지 등의 정보를 이용해 강인한 키 프레임을 추출하고 추출된 다수의 키 프레임 영상에 실린더 투영 방법과 FAST(Feature from Accelerated Segment Test) 기법을 적용하여 자연스러운 정합 영상을 획득할 수 있다. 정합된 특징점의 오차율을 최소화하기 위해 RANSAC(Random Sample Consensus)을 사용하고 여러 장의 다른 시점 영상을 정합할 때 생길 수 있는 경계선을 제거하고 보정하기 위해 선형가중치 함수도 사용한다. 실험을 통해 제안하는 기법으로 자연스러운 파노라마 영상을 생성할 수 있었다.

• KIPS Transactions on Software and Data Engineering
• /
• v.8 no.11
• /
• pp.457-464
• /
• 2019

The weight of pig is one of the main factors in determining the health and growth state of pigs, their shipment, the breeding environment, and the ration of feed, and thus measuring the pig's weight is an important issue in productivity perspective. In order to estimate the pig's weight by using the number of pig's pixels from images, acquired from a Top-view camera, the posture determining and the head removal from images are necessary to measure the accurate number of pixels. In this research, we propose the fast and accurate method to determine the pig's posture by using a fast image processing technique, find the head location by using a fast deep learning technique, and remove pig's head by using light weighted image processing technique. First, we determine the pig's posture by comparing the length from the center of the pig's body to the outline of the pig in the binary image. Then, we train the location of pig's head, body, and hip in images using YOLO(one of the fast deep learning based object detector), and then we obtain the location of pig's head and remove an outside area of head by using head location. Finally, we find the boundary of head and body by using Convex-hull, and we remove pig's head. In the Experiment result, we confirmed that the pig's posture was determined with an accuracy of 0.98 and a processing speed of 250.00fps, and the pig's head was removed with an accuracy of 0.96 and a processing speed of 48.97fps.

• Journal of Broadcast Engineering
• /
• v.26 no.1
• /
• pp.26-38
• /
• 2021

In this paper, DLKF (Deep Learning Keypoint Filtering), the deep learning-based keypoint filtering method for the rapidization of the image registration method for remote sensing images is proposed. The complexity of the conventional feature-based image registration method arises during the feature matching step. To reduce this complexity, this paper proposes to filter only the keypoints detected in the artificial structure among the keypoints detected in the keypoint detector by ensuring that the feature matching is matched with the keypoints detected in the artificial structure of the image. For reducing the number of keypoints points as preserving essential keypoints, we preserve keypoints adjacent to the boundaries of the artificial structure, and use reduced images, and crop image patches overlapping to eliminate noise from the patch boundary as a result of the image segmentation method. the proposed method improves the speed and accuracy of registration. To verify the performance of DLKF, the speed and accuracy of the conventional keypoints extraction method were compared using the remote sensing image of KOMPSAT-3 satellite. Based on the SIFT-based registration method, which is commonly used in households, the SURF-based registration method, which improved the speed of the SIFT method, improved the speed by 2.6 times while reducing the number of keypoints by about 18%, but the accuracy decreased from 3.42 to 5.43. Became. However, when the proposed method, DLKF, was used, the number of keypoints was reduced by about 82%, improving the speed by about 20.5 times, while reducing the accuracy to 4.51.

• The KIPS Transactions:PartA
• /
• v.14A no.2
• /
• pp.77-84
• /
• 2007

The intima media thickness(IMT) of the carotid artery from B mode ultrasound images has recently been proposed as the most useful index of individual atherosclerosis and can be used to predict major cardiovascular events. Ultrasonic measurements of the IMT are conventionally obtained by manually tracing interfaces between tissue layers. The drawbacks of this method are the inter and intra observer variability and its inefficiency. In this paper, we present a multiresolution snake method combined with the dynamic programming, which overcomes the various noises and sensitivity to initialization of conventional snake. First, an image pyramid is constructed using the Gaussian pyramid that maintains global edge information with smoothing in the images, and then the boundaries are automatically detected in the lowest resolution level by minimizing a cost function based on dynamic programming. The cost function includes cost terms which are representing image features and geometrical continuity of the vessel interfaces. Since the detected boundaries are selected as initial contour of the snake for the next level, this automated approach solves the problem of the initialization. Moreover, the proposed snake improves the problem of converging th the local minima by defining the external energy based on multiple image features. In this paper, our method has been validated by computing the correlation between manual and automatic measurements. This automated detection method has obtained more accurate and reproducible results than conventional edge detection by considering multiple image features.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2003.05a
• /
• pp.161-164
• /
• 2003

A study on the gray images morphing using morphology and spline interpolation is realized. The adapted method is that join the 2 broken line together with the dilation of morphology to the morphing image, remove the holes to be a soft edge at the image boundaries and extract the featuring points with the cubic spline interpolation. As a result of experiment the method elicits more smoothing image morphing in detail and speed up the interframe processing.