• Annual Conference of KIPS
• /
• 2012.04a
• /
• pp.665-668
• /
• 2012

멀티미디어 콘텐츠는 디지털 데이터의 특성상 수정과 복제가 쉽고, 이에 따라 불법 유통과 위 변조 등의 문제가 발생할 수 있으므로 콘텐츠 보안의 중요성이 높아지고 있고 가역 워터마킹 방법은 하나의 솔루션이다. 본 논문에서는 JPEG 압축 표준에서 사용할 수 있는 가역 워터마킹 기법을 제안한다. JPEG 압축 과정 중 비손실 과정에서 유지되는 양자화 DCT 블록 데이터에 대하여 인접 블록의 계수간 차이값의 히스토그램을 계산하고, 히스토그램 쉬프팅을 통하여 가역 워터마크를 삽입한다. 디코딩 과정에서 영상에 삽입된 워터마크를 추출함과 동시에 원본 JPEG 표준 영상의 품질을 복원한다. 제안 알고리즘의 성능 평가를 위해 다양한 영상을 대상으로 영상 품질과 압축률을 평가하였다. 삽입된 영상의 PSNR과 압축률은 32.33dB과 89.69%로, 표준 JPEG 압축된 영상 대비 각 3.24dB의 PSNR 차이와 2.58%의 압축률 차이를 보였다.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• fall
• /
• pp.314-317
• /
• 2021

하이라이트 영상은 원본 영상의 중요한 장면들을 짧은 시간 안에 감상할 수 있게 도와준다. 특히나 경기 시간 긴 축구나 야구 그리고 e-스포츠의 시청자들에게 있어, 하이라이트 영상의 효용성은 더욱 증가한다. 하이라이트 영상 추출의 자동화로 방송사나 온라인 플랫폼은 비용 절감과 시간 절약의 이점을 얻을 수 있다. 따라서 본 논문에서는 스포츠 영상에서 자동으로 하이라이트 구간을 추출하는 모델을 제안한다. 제안하는 모델은 멀티 헤드 어텐션 매커니즘과 LSTM 네트워크의 결합으로 구성된다. 해당 매커니즘의 여러 헤드를 통해 어텐션을 다양한 관점에서 진행한다. 이로 인해 영상의 전체적인 맥락과 장면 간의 유기적 관계를 다양한 관점에서 파악할 수 있다. 또한 오디오와 이미지 정보를 함께 이용하여 모델을 학습한다. 학습한 모델의 평가는 e-스포츠 경기 영상을 이용하여 평가한다.

• The Journal of the Korea Contents Association
• /
• v.13 no.8
• /
• pp.275-283
• /
• 2013

The patient radiation dose is different depending on selection of Ion chamber when taking Chest PA which using AEC. In this paper, we studied acquiring the best diagnostic images according to selection of Ion chamber on AEC mode as well as minimizing patient radiation dose. Experimental methods were selection of Ion chamber and change of sensitivity under the same conditions as Chest PA projection. At AEC mode, two upper ion chambers sensors and one lower ion chamber sensor were divided into 7 cases according to selection of on/off. after measuring five times respectively, we obtained average value and calculated exposure dose. Image assessment was done with measured Modulation Transfer Function, Peak Signal to Noise Ratio, Root Mean Square, Signal to Noise Ratio, Contrast to Noise Ratio, Mean to Standard deviation Ratio respectively. In exposure assessment results, selection of two upper chambers was the lowest. In resolution assessment results, image of two upper chambers had the second high spatial frequency at sensitivity at 625(High) was 1.343 lp/mm. RMS value of image selecting two upper chambers was low secondly. SNR, CNR, MSR were the high value secondly. As the sensitivity was increased, radiation dose was decreased but better image could be obtained on image quality. In order to obtain the best medical images while minimizing the dose, usage of two upper ion chambers is considered to be clinically useful at sensitivity 625(High).

• Journal of radiological science and technology
• /
• v.32 no.3
• /
• pp.277-283
• /
• 2009

It is important to get clinical ultrasonic images of good quality for accurate diagnosis. In this study, it observed the change of ultrasonic images against setting frequency, dynamic range(DR) and type of probes on ultrasonic scanner. In the experiment it evaluated image of LCS (Low Contrast Sensitivity) targets(-15, -6, -3, +3, +6, +15 dB) of a standard ultrasonic test phantoms(539,551, ATS, USA) similar to solid and cystic lesions. Its imaged from convex (C3-7IM) and linear probe (L5-12IM) on SA-9900 (Medison Ltd, Korea) scanner. The images obtained altering the setting parameters which are frequency(gen, pen, res, harmonic) mode and DR($40{\sim}100\;dB$). The quality of images evaluated compare with the nominal LCS value of target and measured LCS value. The results show that there was no significant changing of quality images altering DR 40, 60, 80, 100 dB against frequency in Convex probe but the image being the highest in LCS target at DR 60 dB, harmonic of frequency mode in the -15 dB target close to cystic lesion. In Linear probe, DR 40 dB, harmonic mode at -15 dB LCS target close to nominal value. It discussed necessity of evaluation about ROC(Receiver Operating Characteristic) from the psychological viewpoint and limit of evaluation from quantified images.

• Korean Journal of Remote Sensing
• /
• v.30 no.1
• /
• pp.1-12
• /
• 2014

The purpose of this paper is to provide a compact overview of the state-of-art image mosaic algorithms in commercial softwares and to propose objective assessment method of that. Among them, several algorithms, widely used and high quality, result in the mosaic image by applying to seven different kinds of seasons of KOMPSAT-2 images and then consequently each result is analyzed visually. Moreover, quality index is suggested to assess the similarity with colors regarding adjacency images and then it is performed by comparing and analyzing the visual and quantitative results. Consequently, we found out the suggested quality index is feasible.

• Proceedings of the Korea Contents Association Conference
• /
• 2014.11a
• /
• pp.233-234
• /
• 2014

하지전신계측검사(low extremity teleography)에서 자세 변화에 따른 중요장기의 방사선량을 측정하고 영상을 비교 분석하여 검사방법에 따른 유용성을 알아보고자 하였다. 대상은 하지전신계측검사를 시행한 성인남자 10명을 대상하였고 촬영조건은 관전압 73 kVp, 관전류량 32 mAs, SID 180 cm로 설정하였다. 방사선량 측정은 란도 팬텀을 이용하여 수정체, 갑상선, 생식선 부위에 유리선량계(ion chamber)를 부착한 후 전후방향자세와 후전방향자세를 각각 5번씩 시행하여 부위별로 방사선량을 측정한 후 Paired T-test로 비교 분석 하였다. 영상평가는 전후방향자세와 후전방향자세를 시행한 영상을 blind test를 실시하여 5점 척도로 평가하였다. 결과적으로 전후방향자세검사에 비해 후전방향자세검사가 수정체 약 6%, 갑상선 약 6%, 생식선에 미치는 방사선량을 약 27% 감소시킬 수 있으며 영상평가에서도 두 그룹 간에 큰 차이가 없어, 하지전신계측검사에서 전후방향자세검사보다 후전방향자세검사가 유용할 것으로 사료된다.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.18 no.3
• /
• pp.61-68
• /
• 2008

In this paper, we propose a new method to assess fingerprint image quality. In the proposed method, analysis of local variance of image's gray values, local orientation, minutiae density, size and position is applied. Especially by using position information of inputted fingerprint images, partial fingerprint images are filtered and recognition performance is improved. In the experimental results, quality threshold value for improving performance can be decided by analysis of correlation between image quality and recognition rate.

• Journal of the Korean Society of Radiology
• /
• v.10 no.4
• /
• pp.247-253
• /
• 2016

Ultrasound devices diagnose many disease, which is widely used, can not be standardized quantitative evaluated in order to evaluate sonography image of quality. Therefore, in this papers, aims to get correct image in order to accurate diagnosis by figuring out the appropriate parameter based on each target by measuring distortion which results in the analyzation of the sensitivity of SNR and the histogram of signal by manipulating parameter of 8 mm target in ATS-539 multipurpose phantom. Equipment using Acuson sequoia 512, convex probe and utilizes multi-objective phantom. experiment method is that first you put the phantom on the flat and acquire 85 sheets of image, changing frequency(2,3,4 MHz, harmonic 3, 4, 4.5 MHz), Focus(2, 4, 6 unit), and Dynamic Range(58, 68, 78, 88, 98) for a 8 mm structure. through the Image J program. The sensitivity angle of 8mm target through Image J program is gauged by each separate target SNR and the distorted angle subtract and measure Histogram of background from Histogram of signal and take top 40% from the given result value above. According to parameter variation we found out proper parameter by acquiring SNR of sensitivity and distortion data for aspect of transition. The more this findings have Focus, the lower distortion value and at 4 MHz frequency this result have high SNR and low distortion value. Dynamic Range got an appropriate image on 88 and 98. It is considered on the basis of the experimental data, the probability of disease diagnosis will get higher.

• Journal of KIISE
• /
• v.43 no.12
• /
• pp.1356-1364
• /
• 2016

This paper proposes a novel aesthetic photo recomposition method using a deep convolutional neural network (DCNN). Previous recomposition approaches define the aesthetic score of photo composition based on the distribution of salient objects, and enhance the photo composition by maximizing the score. These methods suffer from heavy computational overheads, and often fail to enhance the composition because their optimization depends on the performance of existing salient object detection algorithms. Unlike previous approaches, we address the photo recomposition problem by utilizing DCNN, which shows remarkable performance in object detection and recognition. DCNN is used to iteratively predict cropping directions for a given photo, thus generating an aesthetically enhanced photo in terms of composition. Experimental results and user study show that the proposed framework can automatically crop the photo to follow specific composition guidelines, such as the rule of thirds.

• Proceedings of the Korea Contents Association Conference
• /
• 2006.11a
• /
• pp.47-51
• /
• 2006

As the recent announcement of the functional cosmetics law, the need of studies on efficacy evaluation of skin related cosmetics and medical supplies has grown. Especially to identify skin improvement, we have to compare the exact parts of the skin, however up to now it is compared only by image matching that is appeared to the human eye. This study proposes the automatical image matching system for improving the accuracy of evaluation a skin improvement. Firstly we define the feature of the skin pores and wrinkles, and extract anticipation region from skin images. And then, we calculate moments for each extracted regions and classify them as pores and wrinkles. After that, we calculate the vector by computing centroids between each regions. Through this process, we compare the vector similarities and perform the matching between existing image and reference image. To verify the efficiency of the algorithm several experiments are conducted.