• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2019.06a
• /
• pp.9-12
• /
• 2019

지난 수 년 동안 계속해서 일반 실상 카메라를 이용한 영상분석기술에 대한 연구가 활발히 진행되고 있다. 최근에는 딥러닝 기술을 적용한 지능형 영상분석기술로 발전해 왔으며 국방기지방호, CCTV, 사용자 얼굴인식, 머신비전, 자동차, 드론 산업이 활성화되면서 많은 시너지를 효과를 일으키고 있다. 그러나 어두운 밤과 안개, 날씨, 연기 등 다양한 여건에서 따라서 카메라의 영상분석 정확성 감소와 오류가 수반될 수 있으며 일반적으로 딥러닝 기술을 활용하기 위해서는 고사양의 GPU를 필요로 하기 때문에 다른 추가적인 시스템이 요구된다. 이에 본 연구에서는 열적외선 영상의 객체 검출에 적용하기 위해 SSD(Single Shot MultiBox Detector) 기반의 경량적인 MobilNet 네트워크로 재구성하여, 모바일 기기 등 낮은 사양의 낮은 임베디드 시스템에서도 활용 할 수 있는 방법을 제안한다. 모의 실험결과 제안된 방식의 모델은 적외선 열화상 카메라에서 객체검출과 학습시간이 줄어든 것을 확인 할 수 있었다.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2009.11a
• /
• pp.111-114
• /
• 2009

멀티미디어 동영상의 부호화 알고리즘은 기본적으로 손실 부호화(lossy coding)이기 때문에 압축된 동영상에는 불가피하게 화질의 열화가 포함된다. 압축의 정도가 심하지 않거나 압축을 하더라도 영상이 단순하고 움직임이 적은 경우에는 이러한 열화의 영향이 인간의 눈에는 보이지 않지만, 압축률이 너무 높거나 원본 영상에 포함된 움직임이 복잡하거나 공간적으로 세밀한 영역이 많은 경우에는 손실이 심하여 시각적으로 거슬리는 다양한 잡음이 발생한다. 화면 떨림(flickering)은 시간 영역에서의 대표적인 부호화 잡음으로서 프레임율(frame rate)이 너무 낮은 경우에 움직임이 자연스럽지 않고 불연속적으로 튀거나 부호화 비트율(bit rate)을 제어하기 위해서 양자화 파라미터를 변동하는 과정에서 연속되는 프레임들의 화질에 차이를 보이면서 나타나는 현상이다. 본 논문에서는 DMB 환경에서 사용되는 주기적으로 I 프레임이 삽입되는 GOP(Group of Picture) 구조에서 화면 떨림 현상의 원인을 파악하고 멀티미디어 동영상의 시공간적 특성에 따라 특징적으로 나타나는 화면 떨림 현상을 분석한다. 실험 영상으로서 시공간적으로 다양한 특성을 보이며 H.264/AVC방식으로 부호화된 10개의 동영상을 선택하였으며, 양자화 파라미터를 낮은 화질의 영상에서부터 높은 화질의 영상까지 변화시키면서 화면 떨림 현상의 존재 여부 및 정도를 살펴보았다. 실험의 결과로 화면 떨림 현상은 영상의 화질이 너무 좋거나 너무 나쁘지 않은 중간 영역에서 중요하게 고려해야 할 부호화 잡음이며, 프레임 내의 세밀함의 정도, 움직임의 정도, 객체의 크기, 카메라의 움직임 등에 영향을 받음을 알 수 있다.

• Proceedings of the KIEE Conference
• /
• 2015.07a
• /
• pp.1310-1310
• /
• 2015

대단위 태양광 발전소 또는 고층건물에 설치된 태양광 모듈의 결함을 분석하는데 있어 열화상 카메라를 통한 온도로써 태양광 모듈의 결함을 검출하는 방식이 대두되고 있다. 본 논문에서는 열화상 카메라로 얻은 영상을 온도로 표현하는데 필요한 영상처리를 각각의 태양광 모듈들을 셀 단위로 분류하고 해당 셀을 기준으로 행 이미지를 ROI로 잡은 후 이미지 저장을 하는 방법을 제안한다.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.20 no.4
• /
• pp.840-846
• /
• 2016

Recently, with the rapid demand expansion on the devices of digital image processing, the excellent quality of the images is required. However, the degradation of the images occurs in the processes of data acquisition, processing, and transmission by various external causes and the noise has been well known as the major cause of image degradation. There are a variety of noises to be added on the images, with typical example of AWGN. Hence, in this article, we suggested average filter algorithm processed by the threshold values using them applying standard deviation of local mask under the AWGN environment in this study. As the result of the simulation, The proposed algorithm shows a high PSNR of 24.56[dB] for Barbara images that had been damaged of AWGN(${\sigma}=15$), compared to the existing MF, CWMF and AWMF there were improvements by 3.34[dB], 2.57[dB], and 3.32[dB], respectively.

• Journal of Broadcast Engineering
• /
• v.16 no.3
• /
• pp.552-560
• /
• 2011

In this paper, a new no-reference sharpness measure using frequency domain coefficients is proposed. Although most existing sharpness measures used pixel intensity to compute the blur degree, the proposed sharpness measure computes the sharpness using frequency coefficients. To assess the perceived sharpness of a given image, the image is re-blurred by a Gaussian low pass filter and a new quality measure function was defined using the frequency domain coefficients of the given image and the re-blurred image. To evaluate the proposed algorithms, TID2008 quality assessment database was used. Experimental results show that the proposed quality assessment method showed high correlation with the subjective scores.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2009.05a
• /
• pp.657-660
• /
• 2009

An image interpolation is a basis technique for various image processing and is required to minimize approaches for image quality deterioration. In this paper, we propose an improved bilinear interpolation using loss information estimation. In the proposed algorithm, we estimate loss information of low resolution image using down-sampling and interpolation of acquisition low resolution. The estimated loss information is utilized interpolated image, and it decrease image quality deterioration. Our experiments obtained the average PSNR 0.97~1.79dB which is improved results better than conventional method for sensitive image quality. Also, subjective image quality with edge region is more clearness. The proposed method may be helpful for applications in various multimedia systems such as image resolution enhancement and image restoration.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.31 no.2
• /
• pp.157-164
• /
• 2011

This work is devoted to the technique application of lock-in infrared Thermography in the shipbuilding and ocean engineering industry. For this purpose, an exploratory study to find the optimized test conditions is carried out by the design of experiments. It has been confirmed to be useful method that the phase contrast images were quantified by a reference image and weighted by defect hole size. Illuminated optical intensity of lower or medium strength give a good result for getting a phase contrast image. In order to get a good phase contrast image, lock-in frequency factors should be high in proportion to the illuminated optical intensity. The integration time of infrared camera should have been inversely proportional to the optical intensity. The other hand, the difference of specimen materials gave a slightly biased results not being discriminative reasoning.

• Journal of Broadcast Engineering
• /
• v.13 no.4
• /
• pp.440-451
• /
• 2008

With the advent of IP-based multimedia service based on IP network, there is a rapidly increasing demand for IPTV. Unlike the previous coaxial cable based TV, IPTV provides a variety of convergence services based on IP newark. However, since the IPTV service quality is a lot affected by the network degradation such as packet loss and jitter, it may not be guaranteed. In this paper, we propose an objective measure for various degradations of IPTV-based videos considering subjective assessment. To this end, we first determine QoE(Quality of Experience) indicators, which can affect human visual perception. Then we develop the video quality metric for each QoE indicator. Subjective assessment based on MOS is conducted and used to construct mapping relationship between each measure and perceived visual quality. Experiments are performed on various videos to confirm the efficiency and robustness of the proposed method and show high correlation with subjective assessment.

• Journal of Broadcast Engineering
• /
• v.23 no.1
• /
• pp.146-153
• /
• 2018

This paper presents a superresolution demosaicing technique that can restore high-resolution color image from differently blurred low resolution images in Bayer domain. The proposed superresolution demosaicing algorithm uses an aperture mask wheel to get differently blurred low resolution images, so we just need to estimate point spread function at each frame. In addition, it does not require image registration because there is no translational motion between low resolution images. By using a rotatable aperture mask wheel, consecutive captured images provide sufficiently exclusive information for superresolution. Therefore, the proposed method can reduce the registration error between the low-resolution image as well as the calculation amount for superresolution restoration. The existing lens system of the camera can be extended to obtain a superresolution image by only adding an rotatable aperture mask wheels. Finally, in order to verify the performance of the proposed system, experimental results are performed. The proposed method showed the significant improvements in the sense of spatial and color resolution.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.21 no.1
• /
• pp.130-136
• /
• 2017

Recently, as demand of video processor unit rapidly increases, excellent quality of the video has been required. However, generally, video data occurs the quick flame of video due to various external causes in process of acquisition, treatment, and transmission, and major cause of the quick flame of the video is known as the noise. There are various kinds of noise, which are added to the video, AWGN is a typical one. Thus, this thesis suggested algorithm that treats in three methods by scale of the edge through using edge information of local masks. In case that edge pixel is big, it applied spatial weighting according to equation of straight line about direction of edge pixel. In case that edge pixel is middle, it suggested algorithm with spatial weighting filter and average filter, and for the smooth territory, it suggested algorithm that treats with average filter.