• 디지털산업정보학회논문지
• /
• 제12권4호
• /
• pp.107-114
• /
• 2016

Recently, 3D related technology has become a hot topic for IT. 3D technologies such as 3DTV, Kinect and 3D printers are becoming more and more popular. According to the flow of the times, the goal of this study is that the general public is exposed to 3D technology easily. we have developed a web-based application program that enables 3D modeling of facial front and side photographs using a mobile phone. In order to realize 3D modeling, two photographs (front and side) are photographed with a mobile camera, and ASM (Active Shape Model) and skin binarization technique are used to extract facial height such as nose from facial and side photographs. Three-dimensional coordinates are generated using the face extracted from the front photograph and the face height obtained from the side photograph. Using the 3-D coordinates generated for the standard face model modeled with the standard face as a control point, the face becomes the face of the subject when the RBF (Radial Basis Function) interpolation method is used. Also, in order to cover the face with the modified face model, the control point found in the front photograph is mapped to the texture map coordinate to generate the texture image. Finally, the deformed face model is covered with a texture image, and the 3D modeled image is displayed to the user.

• 한국농업기계학회:학술대회논문집
• /
• 한국농업기계학회 1996년도 International Conference on Agricultural Machinery Engineering Proceedings
• /
• pp.834-841
• /
• 1996

Egg production in Korea is becoming automated with a large scale farm. Although many operations in egg production have been and cracks are regraded as a critical problem. A computer vision system was built to generate images of a single , stationary egg. This system includes a CCD camera, a frame grabber board, a personal computer (IBM PC AT 486) and an incandescent back lighting system. Image processing algorithms were developed to inspect egg shell and to sort eggs. Those values of both gray level and area of dark spots in the egg image were used as criteria to detect holes in egg and those values of both area and roundness of dark spots in the egg and those values of both area and roundness of dark spots in the egg image were used to detect cracks in egg. Fro a sample of 300 eggs. this system was able to correctly analyze an egg for the presence of a defect 97.5% of the time. The weights of eggs were found to be linear to both the projected area and the perimeter of eggs v ewed from above. Those two values were used as criteria to sort eggs. Accuracy in grading was found to be 96.7% as compared with results from weight by electronic scale.

• 대한교통학회지
• /
• 제14권4호
• /
• pp.107-129
• /
• 1996

Recently, surveillance system designed to collect various trsffic information are becoming new areas of development . Among these, the image detector is a ayatem which can measure the travel time and speed in realtime and this is emerging as the most effcient tool to be available in future related areas. But in measuring wide-area information in realtime, the image detector are yet full of problem in its accuracy. The aim of this ahesis is to develop an algorithms which can collect wide-area information such as travel time and travel speed in urban networks and freeways in realtime. The information on wide-area such as travel time and travel speed is important in accomplishing strategic function in traffic control. The algorithm developed from this study is based on the image tracking model which tracks a moving vehicle form image datas collected continuously, and is constructed to perform realtime measurement. To evaluate the performance of the developed algorithms, 600 ind vidual vehicles in total were used as data for the study, and this evaluation was carried out with the differenciation of day and night condition at the access roads in front of AJou University, In the statistical analysis results, the error rate was recorded as 5.69% and it has proved to be applicable on the field in both day and noght conditions.

• 한국레이저가공학회지
• /
• 제12권1호
• /
• pp.19-24
• /
• 2009

Because of its fast and precise welding performance, laser welding is becoming a new excellent welding method. However, the precise focusing and robust seam tracking are required to apply laser welding to the practical fields. In order to laser weld a type of T joint like a circular pipe on a square pipe, which could be met in the three dimensional structure such as an aluminum space frame, a visual sensor system was developed for automation of focusing and seam tracking. The developed sensor system consists of a digital CCD camera, a structured laser, and a vision processor. It is moved and positioned by a 2-axis motorized stage, which is attached to a 6 axis robot manipulator with a laser welding head. After stripe-type structured laser illuminates a target surface, images are captured through the digital CCD camera. From the image, seam error and defocusing error are calculated using image processing algorithms which includes efficient techniques handling continuously changed image patterns. These errors are corrected by the stage off-line during welding or teaching. Laser welding of a circular pipe on a square pipe was successful with the vision tracking system by reducing the path positioning and de focusing errors due to the robot teaching or a geometrical variation of specimens and jig holding.

• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2014년도 추계학술대회
• /
• pp.252-255
• /
• 2014

영상처리는 디지털 시대를 맞이하여 영상복원 기술의 수요가 급증함에 따라 여러 분야에서 대중화되고 있다. 그러나 영상 데이터를 획득, 전송, 처리하는 과정에서 salt and pepper 잡음에 의해 영상이 훼손된다. 기존의 영상을 복원하는 대표적인 방법들은 SMF(standard median filter), CWMF(center weighted median filter), SWMF(switching weighted median filter) 등이 있으며, 이러한 필터들은 잡음 제거 및 에지 보존 특성이 다소 미흡하다. 따라서 제안한 알고리즘은 훼손된 영상을 복원하기 위해 변형된 메디안 필터를 제안하였다.

• 방송공학회논문지
• /
• 제26권1호
• /
• pp.88-98
• /
• 2021

CCTV를 사용하여 저조도와 같은 열악한 환경에서도 범죄 예방 및 특정 대상을 정확히 확인하는 것이 최근 더욱 중요해지고 있다. 저조도 환경하의 CCTV 응용에서는 눈에 거슬리지 않는 근적외선 조명을 이용하여 영상을 획득하는데, 이 경우, 비록 사람 눈에는 어두운 저조도 환경이지만 근적외선 조명을 사용하기 때문에 영상의 상세 텍스처 정보를 얻을 수 있는 장점은 있지만, CCTV 영상내의 물체 판별이나 인물 확인을 위하여 매우 요긴한 정보인 색상 정보는 얻기 힘들다는 단점이 있다. 본 논문에서는 저조도 환경에서 근적외선 조명을 사용하여 얻은 CCTV 영상으로부터 DCGAN을 사용하여 색상정보를 획득하는 방법과 이때 재구성된 색상 영상에 생기는 색상 잡음을 제거하는 방법을 제시한다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제14권11호
• /
• pp.4426-4442
• /
• 2020

Since scene segmentation is becoming a hot topic in the field of autonomous driving and medical image analysis, researchers are actively trying new methods to improve segmentation accuracy. At present, the main issues in image semantic segmentation are intra-class inconsistency and inter-class indistinction. From our analysis, the lack of global information as well as macroscopic discrimination on the object are the two main reasons. In this paper, we propose a Densely connected residual Attention network (DA-Res2Net) which consists of a dense residual network and channel attention guidance module to deal with these problems and improve the accuracy of image segmentation. Specifically, in order to make the extracted features equipped with stronger multi-scale characteristics, a densely connected residual network is proposed as a feature extractor. Furthermore, to improve the representativeness of each channel feature, we design a Channel-Attention-Guide module to make the model focusing on the high-level semantic features and low-level location features simultaneously. Experimental results show that the method achieves significant performance on various datasets. Compared to other state-of-the-art methods, the proposed method reaches the mean IOU accuracy of 83.2% on PASCAL VOC 2012 and 79.7% on Cityscapes dataset, respectively.

• International Journal of Computer Science & Network Security
• /
• 제23권5호
• /
• pp.53-64
• /
• 2023

Early detection of endometrial carcinoma in uterus is essential for effective treatment. Endometrial carcinoma is the worst kind of endometrium cancer among the others since it is considerably more likely to affect the additional parts of the body if not detected and treated early. Non-invasive medical computer vision, also known as medical image processing, is becoming increasingly essential in the clinical diagnosis of various diseases. Such techniques provide a tool for automatic image processing, allowing for an accurate and timely assessment of the lesion. One of the most difficult aspects of developing an effective automatic categorization system is the absence of huge datasets. Using image processing and deep learning, this article presented an artificial endometrium cancer diagnosis system. The processes in this study include gathering a dermoscopy images from the database, preprocessing, segmentation using hybrid Fuzzy C-Means (FCM) and optimizing the weights using the Whale Optimization Algorithm (WOA). The characteristics of the damaged endometrium cells are retrieved using the feature extraction approach after the Magnetic Resonance pictures have been segmented. The collected characteristics are classified using a deep learning-based methodology called Long Short-Term Memory (LSTM) and Bi-directional LSTM classifiers. After using the publicly accessible data set, suggested classifiers obtain an accuracy of 97% and segmentation accuracy of 93%.

• Journal of Information Processing Systems
• /
• 제18권6호
• /
• pp.813-821
• /
• 2022

Intelligent monitoring, life entertainment, medical rehabilitation, and other fields are only a few examples where visual image technology is becoming increasingly sophisticated and playing a significant role. Recognizing Wushu, or martial arts, movements through the use of visual image technology helps promote and develop Wushu. In order to segment and extract the signals of Wushu movements, this study analyzes the denoising of the original data using the wavelet transform and provides a sliding window data segmentation technique. Wushu movement The Wushu movement recognition model is built based on the hidden Markov model (HMM). The HMM model is trained and taught with the help of the Baum-Welch algorithm, which is then enhanced using the frequency weighted training approach and the mean training method. To identify the dynamic Wushu movement, the Viterbi algorithm is used to determine the probability of the optimal state sequence for each Wushu movement model. In light of the foregoing, an HMM-based martial arts movements recognition model is developed. The recognition accuracy of the HMM model increases to 99.60% when the number of samples is 4,000, which is greater than the accuracy of the SVM (by 0.94%), the CNN (by 1.12%), and the BP (by 1.14%). From what has been discussed, it appears that the suggested system for detecting martial arts acts is trustworthy and effective, and that it may contribute to the growth of martial arts.

• 전자공학회논문지CI
• /
• 제48권3호
• /
• pp.91-101
• /
• 2011

정보통신 기술의 급격한 발전으로 인해 인터넷이 대중화됨에 따라 인터넷을 이용한 사이버 공간상에 정보의 상호교환, 전자 상거래, 인터넷뱅킹 등의 사회 활동이 활발해지고 있다. 하지만, 인터넷 사용의 편리함을 추구하는 경향에 의해 개인식별용 증명서(주민등록증, 운전면허증, 여권, 학생증 등)들이 전자적인 매체로 표현되어 인터넷상에서 노출되는 경우가 빈번하게 발생하고 있다. 따라서 본 연구에서는 인터넷상에 노출된 개인정보가 포함된 이미지들을 효율적으로 검색하기 위한 방안을 제안한다. 제안한 방안은 이미지의 색상과 질감, 그리고 모양 특징정보들 중에서 개인식별정보가 포함된 이미지들에서 고유한 특징정보들을 분석하여 추출한 후 이를 이용하여 개인식별정보가 포함된 이미지들을 검색한다. 제안한 방안을 실험한 결과, 전체 개인 식별정보가 포함된 이미지들 중에서 약 89%이상의 검색 성공률과 이미지 파일 당 수행시간은 약 0.17초가 소모되었다. 이러한 결과를 바탕으로 실제 인터넷상에서 개인식별정보가 포함된 이미지 파일들의 검색과 노출 여부 판단을 위한 시스템에 효과적으로 적용할 수 있다.