• Journal of KIISE:Software and Applications
• /
• v.29 no.1_2
• /
• pp.114-125
• /
• 2002

In this paper, we are proposing a hierarchical segmentation method that first segments the video data into units of shots by detecting cut and dissolve, and then decides types of camera operations or object movements in each shot. In our previous work[1], each picture group is divided into one of the three detailed categories, Shot(in case of scene change), Move(in case of camera operation or object movement) and Static(in case of almost no change between images), by analysing DC(Direct Current) component of I(Intra) frame. In this process, we have designed two-stage hierarchical neural network with inputs of various multiple features combined. Then, the system detects the accurate shot position, types of camera operations or object movements by searching P(Predicted), B(Bi-directional) frames of the current picture group selectively and hierarchically. Also, the statistical distributions of macro block types in P or B frames are used for the accurate detection of cut position, and another neural network with inputs of macro block types and motion vectors method can reduce the processing time by using only DC coefficients of I frames without decoding and by searching P, B frames selectively and hierarchically. The proposed method classified the picture groups in the accuracy of 93.9-100.0% and the cuts in the accuracy of 96.1-100.0% with three different together is used to detect dissolve, types of camera operations and object movements. The proposed types of video data. Also, it classified the types of camera movements or object movements in the accuracy of 90.13% and 89.28% with two different types of video data.

• Journal of Broadcast Engineering
• /
• v.26 no.3
• /
• pp.247-257
• /
• 2021

This paper proposes a new technique for calibrating a multi-view RGB-D camera using a 3D (dimensional) skeleton. In order to calibrate a multi-view camera, consistent feature points are required. In addition, it is necessary to acquire accurate feature points in order to obtain a high-accuracy calibration result. We use the human skeleton as a feature point to calibrate a multi-view camera. The human skeleton can be easily obtained using state-of-the-art pose estimation algorithms. We propose an RGB-D-based calibration algorithm that uses the joint coordinates of the 3D skeleton obtained through the posture estimation algorithm as a feature point. Since the human body information captured by the multi-view camera may be incomplete, the skeleton predicted based on the image information acquired through it may be incomplete. After efficiently integrating a large number of incomplete skeletons into one skeleton, multi-view cameras can be calibrated by using the integrated skeleton to obtain a camera transformation matrix. In order to increase the accuracy of the calibration, multiple skeletons are used for optimization through temporal iterations. We demonstrate through experiments that a multi-view camera can be calibrated using a large number of incomplete skeletons.

• Journal of IKEEE
• /
• v.28 no.3
• /
• pp.474-478
• /
• 2024

In this paper, we proposes a deep learning network for quality inspection in a multi-camera inline inspection system for pharmaceutical containers. The proposed deep learning network is specifically designed for pharmaceutical containers by using data produced in real manufacturing environments, leading to more accurate quality inspection. Additionally, the use of an inline-capable deep learning network allows for an increase in inspection speed. The development of the deep learning network for quality inspection in the multi-camera inline inspection system consists of three steps. First, a dataset of approximately 10,000 images is constructed from the production site using one line camera for foreign substance inspection and three area cameras for dimensional inspection. Second, the pharmaceutical container data is preprocessed by designating regions of interest (ROI) in areas where defects are likely to occur, tailored for foreign substance and dimensional inspections. Third, the preprocessed data is used to train the deep learning network. The network improves inference speed by reducing the number of channels and eliminating the use of linear layers, while accuracy is enhanced by applying PReLU and residual learning. This results in the creation of four deep learning modules tailored to the dataset built from the four cameras. The performance of the proposed deep learning network for quality inspection in the multi-camera inline inspection system for pharmaceutical containers was evaluated through experiments conducted by a certified testing agency. The results show that the deep learning modules achieved a classification accuracy of 99.4%, exceeding the world-class level of 95%, and an average classification speed of 0.947 seconds, which is superior to the world-class level of 1 second. Therefore, the effectiveness of the proposed deep learning network for quality inspection in a multi-camera inline inspection system for pharmaceutical containers has been demonstrated.

• The Journal of the Korea Contents Association
• /
• v.5 no.4
• /
• pp.111-121
• /
• 2005

The appearance of digital culture swiftly has changed the culture in domestic and international arenas before and after the year 2004 and the image and animation have become two of the most important expression media in contemporary age. Among the Japanese animations that have demonstrated the rapid development of cartoon and animation in the world, the director Dakahata Isao's 'Graves of the Fireflies' that has influenced many works has been evaluated as one of the noticeable works that has a unique method and scenario dramatization in terms of producing the original novel into an animation. This study investigates the metaphor and symbolism shown in this work according to each sequence, divides the production ability in the work into three elements and applies them to the important elements such as camera, colors and mise-en-scene when the original work is depicted into image. It can be summarized in more detail as in the following: firstly, I study the rhythm of camera corresponding to the symbolism of the angle that the camera has and production; secondly, I analyze the artistic elements appeared in the process of expressing the original work into the image, especially the production for the colors and symbolism contained in them and the composition of screen. Thirdly, I analyze how effectively the atmosphere for the situations for the original work is expressed in animation with the aid of one of the image elements, mis-en-scene. It is expected that the analyzed findings will be effective as a way of overcoming the limitation of expressions that the original work in text and the study on these processes will become good examples to the relevant workers and will be the good references to the producers who are interested in the creation of animation in Korea.

• Journal of Digital Convergence
• /
• v.15 no.10
• /
• pp.525-535
• /
• 2017

The purpose of this study aims to examine consumer perception and current status of water park. The Naver and Daum were used for data collection channels and the keyword 'water park' was used for data retrieval. The data analysis period was limited to the study period from January 1, 2015 to December 31, 2016 for a total of two years. First, as a result of the frequency analysis, hidden cameras, Lotte water park, arrests, suspects, gimhae were in top 5 in 2015, Lotte water park, swimming, summer, opening, admission ticket were in top 5 in 2016. Second, as a result of the connection degree central analysis, hidden camera, arrest, suspect, female, shower room were in top 5 in 2015, swimming, Lotte water park, summer and One Mount, admission ticket were in top 5 in 2016. Third, as a result of the N-GRAM network graph, the water park/hidden camera, the hidden camera/hidden camera, the suspect/arrest, the Gimhae/Lotte water park, water park/suspect were in top 5 in 2015, and One Mount/water park, Gimhae/Lotte water park, water park/admission ticket, water park/water park, water park/opening were in top 5 in 2016. Fourth, as a result of the CONCOR analysis, three groups in 2015 and two groups in 2016 were formed.

• Annals of Coloproctology
• /
• v.34 no.6
• /
• pp.292-298
• /
• 2018

Purpose: This study compared the perioperative clinical outcomes of reduced-port laparoscopic surgery (RPLS) with those of conventional multiport laparoscopic surgery (MPLS) for patients with sigmoid colon cancer and investigated the safety and feasibility of RPLS performed by 1 surgeon and 1 camera operator. Methods: From the beginning of 2010 until the end of 2014, 605 patients underwent a colectomy for sigmoid colon cancer. We compared the characteristics, postoperative outcomes, and pathologic results for the patients who underwent RPLS and for the patients who underwent MPLS. We also compared the clinical outcomes of single-incision laparoscopic surgery (SILS) and 3-port laparoscopic surgery. Results: Of the 115 patients in the RPLS group, 59 underwent SILS and 56 underwent 3-port laparoscopic surgery. The MPLS group included 490 patients. The RPLS group had shorter operating time ($137.4{\pm}43.2minutes$ vs. $155.5{\pm}47.9minutes$, P < 0.001) and shorter incision length ($5.3{\pm}2.2cm$ vs. $7.8{\pm}1.2cm$, P < 0.001) than the MPLS group. In analyses of SILS and 3-port laparoscopic surgery, the SILS group showed younger age, longer operating time, and shorter incision length than the 3-port surgery group and exhibited a more advanced T stage, more lymphatic invasion, and larger tumor size. Conclusion: RPLS performed by 1 surgeon and 1 camera operator appears to be a feasible and safe surgical option for the treatment of patients with sigmoid colon cancer, showing comparable clinical outcomes with shorter operation time and shorter incision length than MPLS. SILS can be applied to patients with favorable tumor characteristics.

• Journal of the Korean Society of Radiology
• /
• v.13 no.4
• /
• pp.661-666
• /
• 2019

Diverging collimators is used to obtain reduced images of an object, or to detect a wide filed-of-view (FOV) using a small gamma camera. In the gamma camera using the diverging collimators, the block scintillator, and the pixel scintillator array, gamma rays are obliquely incident on the scintillator surface when the source is located the periphery of the FOV. Therefore, the spatial resolution is reduced because it is obliquely detected in depth direction. In this study, we designed a novel system to improve the spatial resolution in the periphery of the FOV. Using a tapered crystal array to configure the scintillation pixels to coincide with the angle of the collimator's hole allows imaging to one scintillation pixel location, even if events occur to different depths. That is, even if is detected at various points in the diagonal direction, the gamma rays interact with one crystal pixel, so resolution does not degrade. The resolution of the block scintillator and the tapered crystal array was compared and evaluated through Geant4 Application for Tomographic Emission (GATE) simulation. The spatial resolution of the obtained image was 4.05 mm in the block scintillator and 2.97 mm in the tapered crystal array. There was a 26.67% spatial resolution improvement in the tapered crystal array compared to the block scintillation.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.24 no.1
• /
• pp.189-196
• /
• 2024

Currently, it is almost impossible to move the focus of a previously taken photo to a different location. This paper challenges a technology that can move the focus of a captured photo to another location after shooting. To achieve this goal, this paper proposed and implemented a method for taking photos with various focuses at the moment the camera took pictures and storing them in a single JPEG file to extract photos focused on the user's preferred location. In this paper, two methods are implemented: taking various photos by quickly moving the focal length of the lens from close to far away, and taking various photos focused on each object by recognizing objects in the camera viewfinder. Various photos taken are stored in a single JPEG to maintain compatibility with traditional photo viewers. At this time, this JPEG file used the All-in-JPEG format proposed in previous research to store a variety of images. This paper verified its practicality by implementing these technologies in an Android app named OnePIC.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.52 no.12
• /
• pp.150-157
• /
• 2015

Pedestrian detection has largely been researched as one of the important technologies for autonomous driving vehicle and preventing accidents. There are two categories for pedestrian detection, camera-based and LIDAR-based. LIDAR-based methods have the advantage of the wide angle of view and insensitivity of illuminance change while camera-based methods have not. However, there are several problems with 3D LIDAR, such as insufficient resolution to detect distant pedestrians and decrease in detection rate in a complex situation due to segmentation error and occlusion. In this paper, two methods using GM-PHD filter are proposed to improve the poor rates of pedestrian detection algorithms based on 3D LIDAR. First one improves detection performance and resolution of object by automatic accumulation of points in previous frames onto current objects. Second one additionally enhances the detection results by applying the GM-PHD filter which is modified in order to handle the poor situation to classified multi target. A quantitative evaluation with autonomously acquired road environment data shows the proposed methods highly increase the performance of existing pedestrian detection algorithms.

• Journal of KIISE:Computer Systems and Theory
• /
• v.35 no.2
• /
• pp.84-93
• /
• 2008

Modeling and rendering of cloth texture have been regarded as one of the most important factors to enhance reality of the contents in the digital contents industry. So far, however, two major approaches to realistically describe the cloth texture were developed: building analytical reflectance model for target cloth and sometimes thread itself and obtaining overall reflectance model using optical equipments. However, yielding a plausible analytic reflection model satisfying many subtle characteristics of a cloth is not an easy task; moreover, fine-detailed modeling of the cloth pattern across the target clothes should also be accompanied by huge amount of computation. The method to obtain overall reflectance model needs expensive measurement equipments and data size becomes huge. Since it applies in the end the reflectance model obtained at one point of a cloth to across whole the visible area of the target clothes, it cannot properly reproduce the pattern of the clothes nor the texture. To address the aforementioned problems, this paper proposes a simple low cost camera rig and a novel method for realistic modeling and rendering of the cloth texture by analyzing photos taken by the proposed camera rig, which can reproduce even the texture pattern applied to the whole clothes, overcoming the one-point reflectance model.