• Journal of the Society of Naval Architects of Korea
• /
• v.57 no.3
• /
• pp.140-151
• /
• 2020

It is important to identify the location of ship hull blocks with exact block identification number when scheduling the shipbuilding process. The wrong information on the location and identification number of some hull block can cause low productivity by spending time to find where the exact hull block is. In order to solve this problem, it is necessary to equip the system to track the location of the blocks and to identify the identification numbers of the blocks automatically. There were a lot of researches of location tracking system for the hull blocks on the stockyard. However there has been no research to identify the hull blocks on the stockyard. This study compares the performance of 5 Convolutional Neural Network (CNN) models with multi-view image set on the classification of the hull blocks to identify the blocks on the stockyard. The CNN models are open algorithms of ImageNet Large-Scale Visual Recognition Competition (ILSVRC). Four scaled hull block models are used to acquire the images of ship hull blocks. Learning and transfer learning of the CNN models with original training data and augmented data of the original training data were done. 20 tests and predictions in consideration of five CNN models and four cases of training conditions are performed. In order to compare the classification performance of the CNN models, accuracy and average F1-Score from confusion matrix are adopted as the performance measures. As a result of the comparison, Resnet-152v2 model shows the highest accuracy and average F1-Score with full block prediction image set and with cropped block prediction image set.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.5
• /
• pp.2366-2395
• /
• 2018

Multi-view super-resolution (MVSR) refers to the process of reconstructing a high-resolution (HR) image from a set of low-resolution (LR) images captured from different viewpoints typically by different cameras. These multi-view images are usually obtained by a camera array. In our previous work [1], we super-resolved multi-view LR images via image fusion (IF) and blind deblurring (BD). In this paper, we present a new MVSR method that jointly realizes IF and BD based on an integrated energy function optimization. First, we reformulate the MVSR problem into a multi-channel blind deblurring (MCBD) problem which is easier to be solved than the former. Then the depth map of the desired HR image is calculated. Finally, we solve the MCBD problem, in which the optimization problems with respect to the desired HR image and with respect to the unknown blur are efficiently addressed by the alternating direction method of multipliers (ADMM). Experiments on the Multi-view Image Database of the University of Tsukuba and images captured by our own camera array system demonstrate the effectiveness of the proposed method.

• Journal of Multimedia Information System
• /
• v.1 no.1
• /
• pp.1-10
• /
• 2014

In this paper, we propose a new view synthesis technique for coding of multi-view color and depth data in arbitrary camera arrangements. We treat each camera position as a 3-D point in world coordinates and build clusters of those vertices. Color and depth data within a cluster are gathered into one camera position using a hierarchical representation based on the concept of layered depth image (LDI). Since one camera can cover only a limited viewing range, we set multiple reference cameras so that multiple LDIs are generated to cover the whole viewing range. Therefore, we can enhance the visual quality of the reconstructed views from multiple LDIs comparing with that from a single LDI. From experimental results, the proposed scheme shows better coding performance under arbitrary camera configurations in terms of PSNR and subjective visual quality.

• Journal of Broadcast Engineering
• /
• v.13 no.5
• /
• pp.685-695
• /
• 2008

This paper presents a method to segment the object of interest from a set of multi-view images with minimal user interaction. Specifically, after the user segments an initial image, we first estimate the transformations between foreground and background of the segmented image and the neighboring image, respectively. From these transformations, we obtain regions in the neighboring image that respectively correspond to the foreground and the background of the segmented image. We are then able to segment the neighboring image based on these regions, and iterate this process to segment the whole image set. Transformation of foregrounds are estimated by feature-based registration with free-form deformation, while transformation of backgrounds are estimated by homography constrained to affine transformation. Here, both are based on correspondence point pairs. Segmentation is done by estimating pixel color distributions and defining a shape prior based on the obtained foreground and background regions and applying them to a Markov random field (MRF) energy minimization framework for image segmentation. Experimental results demonstrate the effectiveness of the proposed method.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 1997.11a
• /
• pp.135-138
• /
• 1997

Multi-view 영상 DB의 surround view 입체 화상을 재현하기 위해서는 영상 정보를 효과적으로 합성, 보간하는 방법이 필연적으로 요구된다. 다시점 입체 영상의 표시 기능을 구현하기 위해서 이론적으로는 충분한 개수의 카메라를 상하 좌우로 배치하여 영상을 획득하고 보는 이의 위치에 가장 적합한 영상을 선택하여 표현하면 된다. 이 경우 많은 수의 카메라로부터 입력되는 방대한 영상 데이터의 저장, 처리, 전송, 설치 문제 등 현실적으로 많은 제약점이 존재한다. 이러한 문제점을 해결하기 위해 기존의 블록 매칭과 같은 3D 구조 추정에 기초하여 블록 단위의 깊이 정보와 텍스쳐 정보의 추출, 다시점 영상 set의 가장 왼쪽 영상으로부터 시작하여 차례로 한 줄씩 쌓아서 얻어진 EPI(Epipolar Plane Image)에 기초한 유사 영상 값의 자취선 탐색 방법에 의한 시차와 텍스쳐 정보 추출로 전체 다안화상을 다해상도로 표현하여 전송하는 방식을 제안한다. 다해상도의 영상 표현과 점진적 전송 방법을 통해 다시점 영상 검색 시스템의 성능향상을 위한 네트워크 모델링을 제안하고자 한다.

• Journal of the Korean Institute of Telematics and Electronics S
• /
• v.35S no.7
• /
• pp.53-61
• /
• 1998

In this paper, we propose an efficient multi-view images coding algorithm to find the optimal depth and texture from the set of multi-view images. The proposed algorithm consists of two consecutive steps, i) the depth estraction step, and ii) the texture extraction step, comparedwith the traditional algorithem which finds the depth and texture concurrently. The X-Y plane of the normalized object space is divided into traingular paatches and the Z value of the node is determined in the first step and then the texture of the each patch is extracted in the second step. In the depth extraction step, the depth of the node is determined by applying the block based disparity compensation method to the windowed area centered at the node. In the second step, the texture of the traingular patches is extracted from the multi-view images by applying the affine transformation based disparity compensation method to the traingular pateches with the depth extracted from the first step. Experimental results show that the SNR(Singnal-to- Noise Ratio) of images enconded by our algorithm is better than that of images encoded by the traditional algorithm by the amount about 4dB for for the test sets of multi-view images called dragon, kid, city and santa.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.10 no.10
• /
• pp.5129-5152
• /
• 2016

Multi-view super-resolution (MVSR) aims to estimate a high-resolution (HR) image from a set of low-resolution (LR) images that are captured from different viewpoints (typically by different cameras). MVSR is usually applied in camera array imaging. Given that MVSR is an ill-posed problem and is typically computationally costly, we super-resolve multi-view LR images of the original scene via image fusion (IF) and blind deblurring (BD). First, we reformulate the MVSR problem into two easier problems: an IF problem and a BD problem. We further solve the IF problem on the premise of calculating the depth map of the desired image ahead, and then solve the BD problem, in which the optimization problems with respect to the desired image and with respect to the unknown blur are efficiently addressed by the alternating direction method of multipliers (ADMM). Our approach bridges the gap between MVSR and BD, taking advantages of existing BD methods to address MVSR. Thus, this approach is appropriate for camera array imaging because the blur kernel is typically unknown in practice. Corresponding experimental results using real and synthetic images demonstrate the effectiveness of the proposed method.

• Journal of Broadcast Engineering
• /
• v.12 no.2
• /
• pp.128-136
• /
• 2007

Since the multi-view video is a set of video sequences captured by multiple array cameras for the same three-dimensional scene, it can provide multiple viewpoint images using geometrical manipulation and intermediate view generation. Although multi-view video allows us to experience more realistic feeling with a wide range of images, the amount of data to be processed increases in proportion to the number of cameras. Therefore, we need to develop efficient coding methods. One of the possible approaches to multi-view video coding is to generate an intermediate image using view interpolation method and to use the interpolated image as an additional reference frame. The previous view interpolation method for multi-view video coding employs fixed size block matching over the pre-determined disparity search range. However, if the disparity search range is not proper, disparity error may occur. In this paper, we propose an efficient view interpolation method using initial disparity estimation, variable block-based estimation, and pixel-level estimation using adjusted search ranges. In addition, we propose a multi-view video coding method based on H.264/AVC to exploit the intermediate image. Intermediate images have been improved about $1{\sim}4dB$ using the proposed method compared to the previous view interpolation method, and the coding efficiency have been improved about 0.5 dB compared to the reference model.

• Proceedings of the Korean Society of Surveying, Geodesy, Photogrammetry, and Cartography Conference
• /
• 2010.04a
• /
• pp.53-56
• /
• 2010

An aerial multi-looking camera system equips itself with five separate cameras which enables acquiring one vertical image and four oblique images at the same time. This provides diverse information about the site compared to aerial photographs vertically. However, multi-looking Aerial Camera for building a 3D spatial information don't use a large-size CCD camera, do uses a medium-size CCD camera, if acquiring forward, backward, left and right imagery of Certain objects, Aerial photographing set overlap and sidelap must be considered. Especially, Sideward-looking camera set up by the sidelap to determine whether a particular object can be acquisition Through our research we analyzed of sideward footprint and aerial photographing efficiency of Multi-view imagery by sidelap changing.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2009.01a
• /
• pp.727-731
• /
• 2009

In this paper, we propose a method to generate a free viewpoint image using multi-viewpoint images which are taken by cameras arranged circularly. In past times, we have proposed the method to generate a free viewpoint image based on Ray-Space method. However, with that method, we can not generate a walk-through view seen from a virtual viewpoint among objects. The method we propose in this paper realizes the generation of such view. Our method gets information of the positions of objects using shape from silhouette method at first, and selects appropriate cameras which acquired rays needed for generating a virtual image. A free viewpoint image can be generated by collecting rays which pass over the focal point of a virtual camera. However, when the requested ray is not available, it is necessary to interpolate it from neighboring rays. Therefore, we estimate the depth of the objects from a virtual camera and interpolate ray information to generate the image. In the experiments with the virtual sequences which were captured at every 6 degrees, we set the virtual camera at user's choice and generated the image from that viewpoint successfully.