• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.17 no.2
• /
• pp.167-176
• /
• 1999

The study about quantitative or qualitative analysis of object using digital image is being progressed actively with the development of the image medium and image process technique. But, it is very high that the dependency about image acquisition system of high resolution for image analysis of high accuracy and it is a equipment of high-price. In this study, I extracted the optimum condition of image enhancement by analyzing and enhancing the multiple images which were acquired by system of low-price. And I carried out the analysis of 3D accuracy by being applied the optimum condition of image enhancement. In the result of analysis of average 3D positioning error using law image and enhanced image which is acquired by applying the optimum condition of image enhancement, I could obtain the progressed accuracy about 10％ on the enhanced image.

• Tunnel and Underground Space
• /
• v.34 no.3
• /
• pp.208-217
• /
• 2024

There is so much research effort for developing and implementing deep learning-based surveillance systems to manage health and safety issues in construction sites. Especially, the development of deep learning-based object detection in various environmental changes has been progressing because those affect decreasing searching performance of the model. Among the various environmental variables, the accuracy of the object detection model is significantly dropped under low illuminance, and consistent object detection accuracy cannot be secured even the model is trained using low-light images. Accordingly, there is a need of low-light enhancement to keep the performance under low illuminance. Therefore, this paper conducts a comparative study of various deep learning-based low-light image enhancement models (GLADNet, KinD, LLFlow, Zero-DCE) using the acquired construction site image data. The low-light enhanced image was visually verified, and it was quantitatively analyzed by adopting image quality evaluation metrics such as PSNR, SSIM, Delta-E. As a result of the experiment, the low-light image enhancement performance of GLADNet showed excellent results in quantitative and qualitative evaluation, and it was analyzed to be suitable as a low-light image enhancement model. If the low-light image enhancement technique is applied as an image preprocessing to the deep learning-based object detection model in the future, it is expected to secure consistent object detection performance in a low-light environment.

• Proceedings of the Korean Information Science Society Conference
• /
• 2005.11b
• /
• pp.853-855
• /
• 2005

현재 표준화가 진행중인 SVC(Scalable Video Codec)에는 기존의 FGS방법이 아닌 Cyclic-FGS를 사용하여 영상을 강화하고 있다. 이 Cyclic-FGS 블록간에 Stocking Effect를 줄일 수 있고 넓은 영역을 강화할 수 있다는 장점이 있다. 하지만 널은 영역을 강화하기 때문에 기존의 FGS와 달리 ROI를 강화하는데는 적합하지 않다. 따라서 본 논문에서는 Cyclic-FGS에 적합한 새로운 Ordering 방법을 제안한다. 이 방법을 사용하면 기존의 FGS에서 사용한 Bit-shift방법을 사용하지 않고도 비슷한 효과를 낼 수 있으며, 우리가 원하는 ROI를 강화시킬 수 있다. ROI를 중점적으로 강화를 하다 보면 전체 영상에 대한 화질은 떨어질 수 있다. 그러나 두 가지 모드를 두어서 중점강화 또는 전체영상과 비교해 화질열화가 거의 없는 강화를 할 수 있게 하였다.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.15 no.1
• /
• pp.165-170
• /
• 2015

The unsharp mask technique emphasize the boundary of the image by adding the boundary of the original image. This technique improves quality by emphasize its boundaries but produce rough image from image noise. The multi channel unsharp mask is possible to enhance entire contrast of the image by applying at least two channels of unsharp mask. However, There is limitations to strengthen boundaries even if the scale strongly applies the multi channel unsharp mask technique. To solve this problem, linear scaling to nonlinear scaling by applying exponential function to existing multi channel unsharp mask technique. Experimental results show enhanced contrast for desired area because of control scaling in details compared with existing unsharp mask technique.

• Journal of the Korean earth science society
• /
• v.26 no.1
• /
• pp.58-65
• /
• 2005

Homomorphic filtering improves image by enhancing high components and reducing low components in the Sequency domain based on FFT, as one of useful digital image processing techniques. In this study, the application program f3r homomorphic filtering was developed. Using this program, satellite imageries and geophysical image such as magnetic image data were processed and their results were analyzed. In case of applying to other techniques suck as histogram equalization and kernel-based masking f3r the same purpose. they often cause the slight distortion of boundary or overall change of brightness values on the whole image. Whereas. homomorphic filtering has ability to enhance selectively detailed components in a target image. Therefore. this technique can be effectively used for extraction or separation of complex types of characteristics contained in the satellite imagery. In addition, this technique would be applicable to investigate anomalous zone in various geophysical image data.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2016.11a
• /
• pp.117-118
• /
• 2016

최근 UHDTV(ultra high definition television)가 가정에 보급이 많이 되고 있는 추세지만, UHD급 콘텐츠가 매우 부족한 실정이다. 따라서 저해상도 FHD(full high definition) 영상을 고해상도 영상으로 변환시켜 재활용할 수 있는 초해상화(super-resolution, SR) 기술의 필요성이 커졌다. 그 중, 다층의 레이어로 구성된 다층 선형 매핑(multi-layer linear mappings, MLLM)을 기반으로 하는 제안된 초해상화 기법은 상대적으로 낮은 복잡도로 좋은 품질의 고해상도 영상을 복원할 수 있었다. 최근에는 강화 예측법을 추가하여 복원된 고해상도 영상의 품질을 더 향상시키는 기법이 등장하였는데, 이를 바탕으로 본 논문에서는 제안했었던 MLLM 기법을 위한 강화 예측법 기법을 새롭게 제안한다. 제안하는 초해상화 기법은 기존 MLLM 기법과 딥러닝 기반 초해상화 기법보다 높은 품질의 고해상도 영상을 생성하는 것을 확인하였다.

• Journal of the Korean Society of Radiology
• /
• v.14 no.7
• /
• pp.991-1001
• /
• 2020

Supervised deep learning technologies for improving the image quality of computed tomography (CT) need a lot of training data. When input images have different characteristics with training images, the technologies cause structural distortion in output images. In this study, an imaging model based on the deep reinforcement learning (DRL) was developed for overcoming the drawbacks of the supervised deep learning technologies and reducing noise in CT images. The DRL model was consisted of shared, value and policy networks, and the networks included convolutional layers, rectified linear unit (ReLU), dilation factors and gate rotation unit (GRU) in order to extract noise features from CT images and improve the performance of the DRL model. Also, the quality of the CT images obtained by using the DRL model was compared to that obtained by using the supervised deep learning model. The results showed that the image accuracy for the DRL model was higher than that for the supervised deep learning model, and the image noise for the DRL model was smaller than that for the supervised deep learning model. Also, the DRL model reduced the noise of the CT images, which had different characteristics with training images. Therefore, the DRL model is able to reduce image noise as well as maintain the structural information of CT images.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• fall
• /
• pp.33-36
• /
• 2021

In this paper, we propose a new method for improving compression efficiency of immersive video using reinforcement learning. Immersive video means a video that a user can directly experience, such as 3DOF+ videos and Point Cloud videos. It has a vast amount of information due to their characteristics. Therefore, lots of compression methods for immersive video are being studied, and generally, a method, which projects an 3D image into 2D image, is used. However, in this process, a region where information does not exist is created, and it can decrease the compression efficiency. To solve this problem, we propose the reinforcement learning-based filling method with considering the characteristics of images. Experimental results show that the performance is better than the conventional padding method.

• The Bulletin of The Korean Astronomical Society
• /
• v.35 no.1
• /
• pp.31.2-31.2
• /
• 2010

본 연구의 목적은 태양 코로나 구조를 분석하기에 적합한 영상처리 기법을 찾는 것이다. 이를 위하여 우리는 IDL(Interactive Data Language)에 내장된 여러 가지 영상처리방법을 SOHO EIT 영상에 적용하였다. 우리는 영상처리를 위하여 단일 영상처리 방법과 2단계 영상 처리 방법을 사용하였다. 단일 영상처리 방법으로 히스토그램 평활화(Equalization), 주파수 필터링, 경계선 강화기법(Sobel, Robert) 등을 사용하였다. 2단계 영상처리 방법은 단일 영상 처리 방법에서 효과적이었던 방법들을 두 가지 이상 순차적으로 적용하는 것이다. 본 연구를 통하여 우리는 2단계 영상처리 방법(예, 저주파 필터 + Sobel + 히스토그램 평활화)이 단일 영상처리 방법 보다 코로나 루프 구조를 잘 보여주는 것을 확인하였다. 이 연구 결과가 태양 코로나 구조 연구에 유용하게 사용될 수 있을 것으로 기대된다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2021.05a
• /
• pp.335-337
• /
• 2021

안저 영상 촬영기술이 발달되며 진단에 사용되는 안저 영상에는 시각적으로 많은 변화가 일어났다. 새로운 촬영 기법인 초광각 안저 영상은 기존 영상에 비해 넓은 범위의 영상을 생성할 수 있다. 촬영 범위가 넓어짐에 따라 이미지에는 왜곡이 발생하고, 이로 인해 안저 영상을 통한 황반 부위 진단에 어려움을 야기하기도 한다. 본 논문에서는 이러한 왜곡을 보정하고 초광각 안저 영상을 기존 안저 영상의 영역으로 변환하는 시스템을 강화학습을 통해 구축한다. 제안하는 방법은 A3C 강화학습법을 사용하며 실험 결과는 제안 방법을 통해 안저 영상을 자동으로 변환할 수 있음을 보여준다.