• 한국정보디스플레이학회:학술대회논문집
• /
• 한국정보디스플레이학회 2007년도 7th International Meeting on Information Display 제7권2호
• /
• pp.1490-1493
• /
• 2007

We proposed the method to take 3D image having correct depths to the front and rear directions when the stereogram was displayed to an observer through an optical system. Since the magnified stereogram by lenses was not given correct depth to an observer despite having the same magnified disparity. Consequently, we achieved our goal by relations of compensated disparities to both directions with magnification of lenses, viewing distance and base distance of viewer in AFIS.

• 대한임베디드공학회논문지
• /
• 제13권6호
• /
• pp.313-319
• /
• 2018

We propose a post-processing algorithm to improve the quality of the RGB image generated by deep learning based colorization from the gray-scale image of an infrared camera. Wavelet fusion is used to generate a new luminance component of the RGB image luminance component from the deep learning model and the luminance component of the infrared camera. PSNR is increased for all experimental images by applying the proposed algorithm to RGB images generated by two deep learning models of SegNet and DCGAN. For the SegNet model, the average PSNR is improved by 1.3906dB at level 1 of the Haar wavelet method. For the DCGAN model, PSNR is improved 0.0759dB on the average at level 5 of the Daubechies wavelet method. It is also confirmed that the edge components are emphasized by the post-processing and the visibility is improved.

• 한국인터넷방송통신학회논문지
• /
• 제8권4호
• /
• pp.75-88
• /
• 2008

본 논문은 입력 이미지 블록의 클래스 조건부 확률 밀도 함수의 커널 추정에 기반한 공간 영역에서의 다중초점 이미지 융합 기법을 제안한다. 이미지 융합 문제를 시험 패턴으로부터 추정된 유사 밀도 함수에 의해 사후 클래스 확률, P($w_{i}{\mid}B_{ikl}$),을 계산하는 분류 임무로 접근하였다. C개의 입력 이미지 $I_{i}$에 대하여 제안한 방법은 i 클래스 $w_{i}$를 정의하고 베이즈 결정 원리에 기초하여 판별 함수를 최대화하는 PxQ 블록 $B_{ikl}$의 집합에 의해 표현되는 결정 지도로 부터 융합 이미지 Z(k,l)를 형성한다. 출력 화질의 척도로서 RMSE 와 상호 정보량인 MI를 사용하여 제안한 기법의 성능이 평가되었다. 커널 함수의 폭 ${\sigma}$ 도 변화시키고, 다른 종류의 커널과 블록 크기를 변화시켜 가며 성능평가를 수행하였다. 제안한 가법은 C=2 와 C=3에 대하여 시험하였고 시험 결과는 좋은 성능을 보였다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제15권12호
• /
• pp.4439-4455
• /
• 2021

Compared with two-stage object detection algorithms, one-stage algorithms provide a better trade-off between real-time performance and accuracy. However, these methods treat the intermediate features equally, which lacks the flexibility to emphasize meaningful information for classification and location. Besides, they ignore the interaction of contextual information from different scales, which is important for medium and small objects detection. To tackle these problems, we propose an image pyramid network based on dual attention mechanism (DAIPNet), which builds an image pyramid to enrich the spatial information while emphasizing multi-scale informative features based on dual attention mechanisms for one-stage object detection. Our framework utilizes a pre-trained backbone as standard detection network, where the designed image pyramid network (IPN) is used as auxiliary network to provide complementary information. Here, the dual attention mechanism is composed of the adaptive feature fusion module (AFFM) and the progressive attention fusion module (PAFM). AFFM is designed to automatically pay attention to the feature maps with different importance from the backbone and auxiliary network, while PAFM is utilized to adaptively learn the channel attentive information in the context transfer process. Furthermore, in the IPN, we build an image pyramid to extract scale-wise features from downsampled images of different scales, where the features are further fused at different states to enrich scale-wise information and learn more comprehensive feature representations. Experimental results are shown on MS COCO dataset. Our proposed detector with a 300 × 300 input achieves superior performance of 32.6% mAP on the MS COCO test-dev compared with state-of-the-art methods.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제11권5호
• /
• pp.2555-2575
• /
• 2017

This paper presents a dual exposure fusion method for image enhancement. Images taken with a short exposure time usually contain a sharp structure, but they are dark and are prone to be contaminated by noise. In contrast, long-exposure images are bright and noise-free, but usually suffer from blurring artifacts. Thus, we fuse the dual exposures to generate an enhanced image that is well-exposed, noise-free, and blur-free. To this end, we present a new scale-space patch-match method to find correspondences between the short and long exposures so that proper color components can be combined within a proposed dual non-local (DNL) means framework. We also present a residual filtering method that eliminates the structure component in the estimated noise image in order to obtain a sharper and further enhanced image. To this end, the entropy is utilized to determine the proper size of the filtering window. Experimental results show that our method generates ghost-free, noise-free, and blur-free enhanced images from the short and long exposure pairs for various dynamic scenes.

• 전기학회논문지
• /
• 제66권1호
• /
• pp.121-126
• /
• 2017

Tracking system is essential for Image Guided Surgery(IGS). Optical Tracking System(OTS) is widely used to IGS for its high accuracy and easy usage. However, OTS doesn't work when occlusion of marker occurs. In this paper sensor data fusion with OTS and Inertial Navigation System(INS) is proposed to solve this problem. The proposed system improves the accuracy of tracking system by eliminating gaussian error of the sensor and supplements the disadvantages of OTS and IMU through sensor fusion based on Kalman filter. Also, sensor calibration method that improves the accuracy is introduced. The performed experiment verifies the effectualness of the proposed algorithm.

• 대한방사선기술학회지:방사선기술과학
• /
• 제35권1호
• /
• pp.59-64
• /
• 2012

최근에는 SPECT/CT의 보급으로 융합영상을 여러 질환에 적용하고 있지만, 폐질환에 대한 적용은 널리 이용되고 있지 않다. 특히, CT 결과에서 폐색전증을 의심하여 SPECT를 시행하는 경우가 발생하며, 보다 정확한 진단을 위해 SPECT/CT가 중요하다. 하지만, SPECT/CT를 보유하지 않은 병원에서는 적용에 한계가 발생하며, SPECT/CT를 보유한 병원에서 검사를 시행한다 하더라도 이미 CT검사 이후 SPECT/CT를 진행하는 경우가 많다. CT검사에서 발생하는 피폭선량 외에도 추가적인 SPECT/CT는 이중으로 피폭선량을 발생하는 경우 해당되며, 불필요한 피폭이 발생하는 경우에 속한다. 그러므로 본 연구에서는 불필요한 피폭을 방지하고, SPECT/CT를 보유하지 않은 의료기관에서 SPECT와 CT의 각각 획득된 영상을 융합하고, 폐질환에서의 영상의 유용성을 평가하였다. 팬텀실험은 NEMA $Phantom^{TM}$ (NU2-2001)으로 SPECT와 CT에서 획득된 영상을 융합하였으며, 임상적용은 10명의 환자(남자 7명, 여자3명, 평균나이: 65.3세${\pm}$12.7세)를 대상으로 각각의 영상을 융합하여 분석하였다. 팬텀실험과 임상적용에서 모두 SPECT, CT의 각각의 영상보다 fusion image에서 높은 점수를 나타내었으며, SPECT와 CT영상을 fusion하여 각각의 장점을 극대화 할 수 있었다. CT영상으로부터 획득한 폐혈관 image가 SPECT로부터 얻은 기능적인 영상을 fusion함으로써 pulmonary embolism이 폐실질에 나타내는 영향을 더욱 잘 묘사할 수 있었다. SPECT/CT가 가장 이상적이라 할 수 있지만, 아직 보급되어 있지 않은 경우 이와 같은 프로토콜을 이용하여 진단에 사용한다면 보다 정확한 진단에 도움이 되리라 사료된다.

• 패션비즈니스
• /
• 제13권4호
• /
• pp.60-76
• /
• 2009

Expression of woman characters in fusion traditional dramas plays a role of making the progress of dramas not bigoted and new. Especially, woman characters have a high weight as a heroine by an increase in their image in fusion traditional dramas. Expression of characters harmonizing modern with tradition has also given a help in reflecting a trend of the present times. Hair style and face makeup of woman characters in fusion traditional dramas are in the process of fusionization getting an effect from postmodernism. They are expressing the hair style of symbolization modern elements of hair style to traditional hair styles. They also expressed a neutral image with faded hair styles in the shaggy cut style and dye of neoplaticism. Neo-hippie style was changed into the style of naturalism and nationalism and the hair style braided in various branches as the one of Indians was changed into a primitive and national feature. It is producing a new style by permitting even a long-hair permanent wave hair style. Expression of straight hair style, a long-hair shaggy & bulging wave style and a hair style of neoplaticism, was distinguished. In the side of face makeup, they expressed its luxurious and splendid style by attaching great importance to its luster and are exposing images of characters by a smoky makeup emphasizing eye lines. Their face makeup was almost never separated from present dramas as using pearl shadow in a glossy lips makeup and color, which made it possible to express it more dramatically in fusion traditional dramas than in present dramas. In the event of the makeup element of fusion traditional dramas permitting diversity, the character makeup of fusion traditional dramas made a foundation to show people diverse elements of makeup by mix & match a present elements and past elements of historical research, which made it possible to express a unique makeup or a special makeup. Diverse makeup expressions were limited by reflection of illumination even in the existing videos. Therefore, 'Fusion' made it possible to express it more freely in fusion traditional dramas than in present dramas.

• 한국군사과학기술학회지
• /
• 제12권4호
• /
• pp.524-531
• /
• 2009

In this paper, we propose an automatic image registration method for multi-sensor image fusion such as visible and infrared images. The registration is achieved by finding corresponding feature points in both input images. In general, the global statistical correlation is not guaranteed between multi-sensor images, which bring out difficulties on the image registration for multi-sensor images. To cope with this problem, mutual information is adopted to measure correspondence of features and to select faithful points. An update algorithm for projective transform is also proposed. Experimental results show that the proposed method provides robust and accurate registration results.

• 대한원격탐사학회:학술대회논문집
• /
• 대한원격탐사학회 2003년도 Proceedings of ACRS 2003 ISRS
• /
• pp.1316-1318
• /
• 2003

The Bernstein polynomial is one of the classic algorithms of panoramic images mosaic for shading into process applying in Virtual Reality modeling. Nevertheless, it is proven that the algorithm has its own limitation and weakness in applications. This paper was given the improved algorithm using Sinusoidal function for image mosaic. In order to put the new algorithm into image processing software as a flexible and general tool, it was further developed an extension for graduation image fusion and multi-images mosaic.