• 전자공학회논문지B
• /
• 제31B권8호
• /
• pp.181-191
• /
• 1994

A new method for enhancing blurred images using fuzzy logic concepts is proposed. Blurred images contain blurred boundaries which make it difficult to detect edges and segment areas in images. In order to sharpen blurred edges local contrast information of an image and erosion/dilation properties of local min/max operations are used in which local min/max operations are fuzzy logic operations. so that given images are transformed to fuzzy images and then these operations are applied on them. In this method the sharpening operation can be iteratively applied to the image to get better deblurring effect and gray-scale "salt-and-pepper" noises are suppressed. the efficiency of our algorithm is demonstrated through experimental results obtained with artificially-made blurred images and real blurred images.

• 전기의세계
• /
• 제31권3호
• /
• pp.204-208
• /
• 1982

The objective of this paper is to present a digital method for improving the lateral resolution of the B-scan images in the medical applications of ultrasound. The method is based upon a mathematical model of the lateral blurring caused by the finite beam width of the transducers. This model provides a simple method of applying a recursive scheme for image restoration with fast computation time. The point spread function (P.S.F.) can be measured by the reflective signals after scanning the small pins located along the depth of interest. From the measured P.S.F., one can compute the coefficient matrices of the inverse discrete-time dynamic state variable equation of the blurring process. Then, a recursive scheme for deblurring is applied to the recorded B-scan to improve the lateral resolution. One major advantage of the present recursive scheme over the transform method is in its applicability for the space-variant imaging, such as in the case of the rotational movement of transducer.

• 대한전기학회논문지
• /
• 제40권4호
• /
• pp.394-401
• /
• 1991

B-spline function is generally used for an image interpolation because of its smoothness and continuity, but it accompanies a large amount of blurring effect. In this paper, a space-variant B-spline interpolation function is proposed through deblurring process followed by de-aliasing process. The proposed function has parametric expression and performs smoothing and edge-enhancement adaptively in the interpolation process according to local property of the image. Application of this function to image enlargement, rotation, and curve representation producted good results. Even in the presence of noise, noise smoothing effect as well as edge-enhancement were observed in the image interpolation process.

• 한국멀티미디어학회논문지
• /
• 제22권10호
• /
• pp.1149-1159
• /
• 2019

We present a novel deep learning architecture for obtaining a latent image from a single blurry image, which contains dynamic motion blurs through object/camera movements. The proposed architecture consists of two sub-modules: blur image restoration and optical flow estimation. The tasks are highly related in that object/camera movements make cause blurry artifacts, whereas they are estimated through optical flow. The ablation study demonstrates that training multi-task architecture simultaneously improves both tasks compared to handling them separately. Objective and subjective evaluations show that our method outperforms the state-of-the-arts deep learning based techniques.

• 한국방송∙미디어공학회:학술대회논문집
• /
• 한국방송∙미디어공학회 2020년도 하계학술대회
• /
• pp.680-683
• /
• 2020

오래 전부터 모델 기반 최적화 방법이 이미지 디블러링을 위해 널리 사용되어 왔고, 최근에는 학습 기반 기술이 영상 디블러링에서 좋은 성과를 보이고 있다. 본 논문은 ADMM과 깊은 합성곱 신경망 잡음 제거기 이미지 prior를 이용하여 모델 기반 최적화 방법의 장점과 학습 기반 방법의 장점을 모두 활용할 수 있는 방법을 제안한다. 본 방법을 이용하여 기존 방법보다 더 좋은 디블러링 성능을 얻을 수 있었다.

• 한국방송∙미디어공학회:학술대회논문집
• /
• 한국방송∙미디어공학회 2022년도 하계학술대회
• /
• pp.235-238
• /
• 2022

본 논문은 인식이 어려운 조명 환경에도 강인한 seven-segment 문자 인식을 위해서, 영상 내에 다양한 조명 연출이 가능하도록 합성 데이터 셋을 생성하고 학습할 수 있는 OCR 방법을 제안한다. 기존 연구에서는 deblurring 과 같이 영상 이미지의 해상도를 높여 문자 인식의 정확도를 향상시키는 것에 초점을 두었으나, 여러 조명 환경에 대비할 수 있는 OCR 관련 연구들은 부족하다. 이를 해결하기 위해 본 논문에서는 문자가 포함된 자연스러운 배경 영상에, seven-segment 문자를 합성시킨 후 relighting 을 적용함으로써 실제 환경과 유사한 장면을 연출해 새로운 합성 데이터 셋을 생성한다. 그리고 생성된 데이터 셋을 딥러닝 기반 학습시켜 다양한 조명에도 강인한 문자 인식기를 만들고자 한다. 합성 데이터 셋의 사용여부와 일반적인 데이터 augmentation 기법의 사용 여부를 비교하여, 본 논문에서 제안한 방법의 효과를 확인할 수 있었다. 이를 통해서 seven-segment 문자 인식 뿐만 아니라, 다양한 문자에 대해서도 적용될 수 있는 초석이 될 것으로 기대된다.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2023년도 춘계학술발표대회
• /
• pp.473-475
• /
• 2023

자동차 번호판 인식은 영상 내 검출한 차량의 번호판의 문자열을 인식하여 차량을 식별하고 추적하는 기술로 주변 환경에 의한 잡음, 왜곡과 차량의 움직임으로 발생한 흐림, 영상 입력 장치와의 물리적 거리 등에 강인해야 한다. 본 논문에서는 차량 움직임으로 발생한 흐림이 있는 저해상도 영상에 대한 번호판 인식 성능의 향상을 위해 디블러링 모델과 초해상화 모델을 이용한 영상 복원 방법을 제안한다. 실험을 통해 디블러링 모델과 초해상화 모델을 결합하여 흐림이 있는 저해상도 국내 번호판 영상에서의 인식 성능을 개선하였다.

• 대한원격탐사학회:학술대회논문집
• /
• 대한원격탐사학회 2004년도 Proceedings of ISRS 2004
• /
• pp.115-118
• /
• 2004

Images with high resolution are desired and often required in many visual applications. When resolution can not be improved by replacing sensors, either because of cost or hardware physical limits, super resolution image reconstruction method is what can be resorted to. Super resolution image reconstruction method refers to image processing algorithms that produce high quality and high resolution images from a set of low quality and low resolution images. The method is proved to be useful in many practical cases where multiple frames of the same scene can be obtained, including satellite imaging, video surveillance, video enhancement and restoration, digital mosaicking, and medical imaging. The method can be either the frequency domain approach or the spatial domain approach. Much of the earlier works concentrated on the frequency domain formulation, but as more general degradation models were considered, later researches had been almost exclusively on spatial domain formulations. The method in spatial domains has three stages: i) motion estimate or image registration, ii) interpolation onto high resolution grid and iii) deblurring process. The super resolution grid construction in the second stage was discussed in this paper. We applied the Maximum A­Posteriori(MAP) reconstruction method that is one of the major methods in the super resolution grid construction. Based on this method, we reconstructed high resolution images from a set of low resolution images and compared the results with those from other known interpolation methods.

• 대한전기학회논문지:시스템및제어부문D
• /
• 제51권8호
• /
• pp.360-367
• /
• 2002

In this paper, genetic algorithm based adaptive image enhancement filtering scheme is proposed and Implemented on FPGA board. Conventional filtering methods require a priori noise information for image enhancement. In general, if a priori information of noise is not available, heuristic intuition or time consuming recursive calculations are required for image enhancement. Contrary to the conventional filtering methods, the proposed filter system can find optimal combination of filters as well as their sequent order and parameter values adaptively to unknown noise types using structured genetic algorithms. The proposed image enhancement filter system is mainly composed of two blocks. The first block consists of genetic algorithm part and fitness evaluation part. And the second block consists of four types of filters. The first block (genetic algorithms and fitness evaluation blocks) is implemented on host computer using C code, and the second block is implemented on re-configurabe FPGA board. For gray scale control, smoothing and deblurring, four types of filters(median filter, histogram equalization filter, local enhancement filter, and 2D FIR filter) are implemented on FPGA. For evaluation, three types of noises are used and experimental results show that the Proposed scheme can generate optimal set of filters adaptively without a pioi noise information.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제9권11호
• /
• pp.4604-4622
• /
• 2015

It is not easy to reconstruct the geometrical characteristics of the distorted images captured by the devices. One of the most popular optimization methods is fast iterative shrinkage/ thresholding algorithm. In this paper, to deal with its approximation error and the turbulence of the decrease process, an adaptive proximal conjugate gradient (APCG) framework is proposed. It contains three stages. At first stage, a series of adaptive penalty matrices are generated iterate-to-iterate. Second, to trade off the reconstruction accuracy and the computational complexity of the resulting sub-problem, a practical solution is presented, which is characterized by solving the variable ellipsoidal-norm based sub-problem through exploiting the structure of the problem. Third, a correction step is introduced to improve the estimated accuracy. The numerical experiments of the proposed algorithm, in comparison to the favorable state-of-the-art methods, demonstrate the advantages of the proposed method and its potential.