• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.11 no.7
• /
• pp.1341-1347
• /
• 2007

The displacement field (Optical flow) has been calculated by bottom-up approaches based on local processing. In contrast with them, in this paper, a top-down approach based on expanding in turn from the lowest order mode the whole motion in an image pair of sequential images is proposed. The intensity of medical images usually represents a quantity which is conserved during the motion. Hence sequential images are ideally related by a coordinate transformation. The displacement field can be determined from the generalized moments of the two images. The equations which transform arbitrary generalized moments from a source image to a target image are expressed as a function of the displacement field. The appareent displacement field is then computed iteratively by a projection method which utilizes the functional derivatives of the linearized moment equations. This method is demonstrated using a pair of sequential heart images. For comparative evaluation, we applied Horn and Schunck's method, a standard multigrid method, and our proposed algorithm to sequential image.

• Journal of KIISE:Computer Systems and Theory
• /
• v.31 no.3_4
• /
• pp.224-238
• /
• 2004

Multi-modality image registration is a widely used image processing technique to obtain composite information from two different kinds of image sources. This study proposes an image registration method based on moment information and surface distance, which improves the previous surface-based registration method. The proposed method ensures stable registration results with low registration error without being subject to the initial position and direction of the object. In the preprocessing step, the surface points of the object are extracted, and then moment information is computed based on the surface points. Moment information is matched prior to fine registration based on the surface distance, in order to ensure stable registration results even when the initial positions and directions of the objects are very different. Moreover, surface comer sampling algorithm has been used in extracting representative surface points of the image to overcome the limits of the existed random sampling or systematic sampling methods. The proposed method has been applied to brain MRI(Magnetic Resonance Imaging) and PET(Positron Emission Tomography), and its accuracy and stability were verified through registration error ratio and visual inspection of the 2D/3D registration result images.

• Journal of Intelligence and Information Systems
• /
• v.28 no.1
• /
• pp.89-106
• /
• 2022

Recently, the amount of video data collected from smartphones, CCTVs, black boxes, and high-definition cameras has increased rapidly. According to the increasing video data, the requirements for analysis and utilization are increasing. Due to the lack of skilled manpower to analyze videos in many industries, machine learning and artificial intelligence are actively used to assist manpower. In this situation, the demand for various computer vision technologies such as object detection and tracking, action detection, emotion detection, and Re-ID also increased rapidly. However, the object detection and tracking technology has many difficulties that degrade performance, such as re-appearance after the object's departure from the video recording location, and occlusion. Accordingly, action and emotion detection models based on object detection and tracking models also have difficulties in extracting data for each object. In addition, deep learning architectures consist of various models suffer from performance degradation due to bottlenects and lack of optimization. In this study, we propose an video analysis system consists of YOLOv5 based DeepSORT object tracking model, SlowFast based action recognition model, Torchreid based Re-ID model, and AWS Rekognition which is emotion recognition service. Proposed model uses single-linkage hierarchical clustering based Re-ID and some processing method which maximize hardware throughput. It has higher accuracy than the performance of the re-identification model using simple metrics, near real-time processing performance, and prevents tracking failure due to object departure and re-emergence, occlusion, etc. By continuously linking the action and facial emotion detection results of each object to the same object, it is possible to efficiently analyze videos. The re-identification model extracts a feature vector from the bounding box of object image detected by the object tracking model for each frame, and applies the single-linkage hierarchical clustering from the past frame using the extracted feature vectors to identify the same object that failed to track. Through the above process, it is possible to re-track the same object that has failed to tracking in the case of re-appearance or occlusion after leaving the video location. As a result, action and facial emotion detection results of the newly recognized object due to the tracking fails can be linked to those of the object that appeared in the past. On the other hand, as a way to improve processing performance, we introduce Bounding Box Queue by Object and Feature Queue method that can reduce RAM memory requirements while maximizing GPU memory throughput. Also we introduce the IoF(Intersection over Face) algorithm that allows facial emotion recognized through AWS Rekognition to be linked with object tracking information. The academic significance of this study is that the two-stage re-identification model can have real-time performance even in a high-cost environment that performs action and facial emotion detection according to processing techniques without reducing the accuracy by using simple metrics to achieve real-time performance. The practical implication of this study is that in various industrial fields that require action and facial emotion detection but have many difficulties due to the fails in object tracking can analyze videos effectively through proposed model. Proposed model which has high accuracy of retrace and processing performance can be used in various fields such as intelligent monitoring, observation services and behavioral or psychological analysis services where the integration of tracking information and extracted metadata creates greate industrial and business value. In the future, in order to measure the object tracking performance more precisely, there is a need to conduct an experiment using the MOT Challenge dataset, which is data used by many international conferences. We will investigate the problem that the IoF algorithm cannot solve to develop an additional complementary algorithm. In addition, we plan to conduct additional research to apply this model to various fields' dataset related to intelligent video analysis.

• Journal of Korea Multimedia Society
• /
• v.6 no.1
• /
• pp.28-39
• /
• 2003

This paper is an overview of research contributions by the authors to the use of compression techniques to handle high resolution, multi-spectral images. Originally developed in the remote sensing context, the same techniques are here applied to food and medical images. The objective is to point out the potential of this kind of processing in different contexts such as remote sensing, food monitoring, and medical imaging and to stimulate new research exploitations. Compression is based on the simple assumption that it is possible to find out a relationship between pixels close one each other in multi-spectral images it translates to the possibility to say that there is a certain degree of correlation within pixels belonging to the same band in a close neighbourhood. Once found a correlation based on certain coefficient on one band, the coefficients of this relationship are, in turn, quite probably, similar to the ones calculated in one of the other bands. Based upon this second observation, an algorithm was developed, able to reduce the number of bit/pixel from 16 to 4 in satellite remote sensed multi-spectral images. A comparison is carried out between different methods about their speed and compression ratio. As reference it was taken the behaviour of three common algorithms, LZW (Lempel-Ziv-Welch), Huffman and RLE (Run Length Encoding), as they are used in common graphic format such as GIF, JPEG and PCX. The Presented methods have similar results in both speed and compression ratio to the commonly used programs and are to be preferred when the decompression must be carried out on line, inside a main program or when there is the need of a custom made compression algorithm.

• Progress in Medical Physics
• /
• v.4 no.2
• /
• pp.59-67
• /
• 1993

In this paper, we present a result of our study on how to construct a HIMCS(hospital image management and communication system) based on Windows system. The proposed HIMCS is composed of image acquisition units, image archiving and processing units, display units, and all units are interfaced with LAN. We construct high speed image transmission system for distributed database and retrieval of various medical pictures in ward through image transmission system and realize integrated image diagnosis. Hospital image management and communication systems(HIMCS) have been proposed as a way to meet the challenge presented by the growing volume of medical imaging data. These systems are based on the concept of handing images in digital form, there by marking use of advances in computer-based technology. A HIMCS offers a wide spectrum of advantages over filmbased imaging. Digital acquisition of images eliminates the need for film processing and associated costs in both time and materials.

• Journal of Broadcast Engineering
• /
• v.16 no.4
• /
• pp.584-595
• /
• 2011

Recently, RGB images and depth maps have been supplied to academic fields. The depth maps are utilized to the generation of stereoscopic images in the diverse formats according to the users' preference. A variety of methods that use depth maps have been introduced so far. One of applications is a medical field. In this area, the improvement of the perceptual quality of 2D medical images has gained much interest. In this paper, we propose a novel scheme that expands the conventional method to 3D stereoscopic image, thereby achieving the perceptual depth quality improvement as well as 3D stereoscopic perception enhancement at the same time. For this, contrast transformation as well as depth darkening are proposed and their performance is validated through the subjective test. Subjective experiments peformed for stereoscopic enhancement as well as visual fatigue validate that the proposed method achieves better 3D perception than the usage of the original stereoscopic image and suggests the limitation in terms of the visual fatigue.

• Journal of KIISE:Software and Applications
• /
• v.27 no.8
• /
• pp.804-814
• /
• 2000

In this paper, we propose an effective method to detect the regions of interests in the Videostrobokymography System. Videostrobokymography system is a medical image processing system for extracting automatically the diagnosis parameters from the irregular vibratory movements of the vocal fold. We detect the regions of interests through three steps. In the first step, we remove the noise in the input image and we find the minimum energy value in each frame. In the second step, we computed the edge by everage value for the one line. In the third step, the regions of interests can be extracted by using the Merge Algorithm which uses the variance of luminance as the feature points. We experimented this method for the vocal fold images of nineteen patients. In consequence, the regions of interests are detected in most vocal fold images. The method proposed in this study is efficient enough to extract the region of interests in the vocal fold images with the frame rate of 40 frames/second and the resolution of 200${\times}$280 pixels.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.05a
• /
• pp.210-212
• /
• 2021

In modern society, various digital equipment are being distributed due to the influence of the 4th industrial revolution, and they are used in a wide range of fields such as automated processes, intelligent CCTV, medical industry, robots, and drones. Accordingly, the importance of the preprocessing process in a system operating based on an image is increasing, and an algorithm for effectively reconstructing an image is drawing attention. In this paper, we propose a filter algorithm based on a combined weight value to reconstruct an image in a complex noise environment. The proposed algorithm calculates the weight according to the spatial distance and the weight according to the difference between the pixel values for the input image and the pixel values inside the filtering mask, respectively. The final output was filtered by applying the join weights calculated based on the two weights to the mask. In order to verify the performance of the proposed algorithm, we simulated it by comparing it with the existing filter algorithm.

• Journal of the Institute of Convergence Signal Processing
• /
• v.8 no.4
• /
• pp.249-254
• /
• 2007

This paper describes a technique for detecting the boundary of the optic disk in digital image of the retina using inward and outward curve evolution. Optic disk boundary offers medical information about glaucoma progresses. For accurate boundary detection, image inpainting based on PDE removes blood vessels crossing the optic disk. For removing noises and preserving boundary of optic disk in image inpainting process, the anisotropic diffusion filtering is developed. After pre-processing, the optic disk boundary is determined using inward and outward curve evolution. Experimental results show that blurring effect of original region and optic disk boundary is reduced considerably. By the proposed method, we can detect correct disk boundary compare to conventional method.

• The Korean Society of Law and Medicine
• /
• v.20 no.1
• /
• pp.109-132
• /
• 2019

Recently, medical accidents related to surgical procedures have increased. In addition, the media reported that some of these accidents were involved in health crimes. Patient-advocate groups have called for mandatory establishment and management of CCTV in operating rooms. There is a lot of discussion among the interested parties, so it is necessary to review the relevant laws and regulations. The purpose of this study is to identify the characteristics of CCTV in operating rooms and to review legislations related to establishment and management of the CCTV in operating rooms. Medical institutions use CCTV for management of facilities and patient safety and install it in operating rooms optionally. The Constitution guarantees the privacy and the privacy of correspondence of every citizen, but it can be limited by the law for public welfare. Currently, however, there is no existing law about establishment and management of the CCTV in operating rooms and it can be defect of legal system. Under the current legislations, it is likely that the Self-determination can be violated due to the characteristic of healthcare provider when CCTV is mandatorily installed in operating room. In addition, the regulations on access and leakage of confidential information known by operator are insufficient. So that, the safety of the visual data might be threatened. Furthermore, unless the period and the place of storage of the visual data are clearly defined, it is highly unlikely to meet the original purpose of patient safety and prevention of medical accidents. This study is meaningful as there is few previous study on this topic although the need for legal review about this is growing and several bills are being proposed. It is expected that the results of this study can be utilized as basic data for enactment or amendment of the laws and regulations about establishment and management of CCTV in operating rooms.