Earticle

본 연구에서는, 덴탈 파노라마 영상의 선예도 향상을 위해 새로운 유형의 적응형 파노라마 기술을 제안했다. 제안하는 기법은 동심원 운동에 대한 FBP 방법을 사용하여 재구성하는 디지털 토모신테시스를 기반으로 한다. 해당 지오메트리에 서는 불완전한 샘플링으로 인한 재구성 후 흐려짐 현상을 억제하기 위해 적합한 필터를 사용하여 필터링 프로세스를 구현했다. 제안한 방법의 타당성을 검증하기 위해 체계적인 시뮬레이션 및 재구성 영상의 특성을 조사하였다. 우리의 결과는 사전에 정의된 관심 평면에서 선명한 영상을 획득할 수 있음을 나타냈다. 결과적으로 제안하는 알고리즘은 영상 의 선예도가 명확한 결과를 가져오며 이는 덴탈 재구성 영상의 판독성을 증가시켜 진단에 도움이 될 것으로 예상한다.

In this study, we proposed a new type adaptive panoramic technique for improving the image sharpness. This imaging method is based upon the digital tomosynthesis using a filtered-backprojection (FBP) method for an iso-centric circular motion. We implemented a filtering process using the apodizing filter to suppress the blurring artifact due to the incomplete sampling in this geometry. To verify the feasibility of the suggested technique, we have performed systematic simulation works and investigated the image characteristics. Our results indicate that predefined planes-of-interest can be well focused with definite image sharpness and that the position of image layer center can be adjusted precisely with proper amounts of shift in the reconstruction.

디지털 X-선 영상에서는 불완전한 영상시스템에 의해 왜곡된 정보가 반영되어 나타나며 이는 영상 진단에 있어 위양 성을 발생시키는 원인이 된다. 특히 유한한 X-선 튜브의 초점크기와 디지털 디텍터의 유한한 픽셀크기 등으로 인해 발생되는 시스템 열화함수 (point-spread function, PSF)는 영상을 블러링 시키는 주된 원인이 된다. 본 연구에서는 이러한 PSF를 영상 시스템에서 측정하여 제한된 역필터링 기법을 통해 블러링을 제거함으로써 영상 복원을 수행하였다. 본 연구에서의 결과는 노이즈와 같은 방해요소가 있는 영상에서도 제안하는 제한된 역필터링 기법이 효과적임을 확인할 수 있다.

In this study, we proposed a new image restoration method using the limited inverse direct filtering method in digital radiography for improving image quality. The obtained image from the imaging system is imperfect due to infinite focal spot size and detector pixel size mainly. It is influenced to the diagnosis toward the false. We implemented the image restoration focused on the inverse filtering using the point spread-function (PSF) using the proposed method. The results of this study show that the limited inverse direct filtering interferences such as noise is effective.

MRI 영상에 있어 노이즈를 제거하기 위한 새로운 방법을 제안한다. MRI영상은 실영상과 허상영상으로부터 계산되어진 영상과의 의존적인 Rician 노이즈가 포함되어져 만들어진다. NLM 필터는 추가적인 노이즈에 대해 효과적인 것으로 증명되어진다. 비슷하게도 디노이징 기반 웨이브렛 변환은 더 좋은 노이즈 추정을 한다. 제안된 알고리듬은 기능적 3단계로 디노이징 과정을 수행한다. 첫 번째로 노이즈 영상은 웨이블렛 변환을 적용에 의한 다수의 서브밴드에 나타난다. NLM 필터는 웨이브렛 필터 뱅크를 사용한 신호 분해의 저주파 성분의 서브밴드에 적용되어진다. 실 노이즈 영상에 있어 제거될 노이즈에 매우 효과적으로 판명되어질 다중해상도 NLM 필터는 새로운 디노이징 영상 프레임워크를 만들기 위해 웨이브렛 임계와 결합되어진다. 실험은 알고리듬이 영상 질적 통계에 의해 효과적으로영상 노이즈를 줄일 수 있는 것으로 나타난다.

We propose a new method for the reduction of noise present in the magnetic resonance (MR) images. Magnetic resonance imaging (MRI) is corrupted by Rician noise, which is image dependent and computed from both real and imaginary images. Rician noise makes image-based quantitative measurement difficult. The non-local means (NLM) filter has been proven to be effective against additive noise. Similarly, Wavelet transform (WT) based denoising produces a better noise estimation. The proposed algorithm performs denoising in three functional steps. First, the noisy image is decomposed into multiple subbands by using the wavelet transform. NLM filter is applied to the approximation (low-frequency) subbands of a signal decomposed using a wavelet filter bank. The multiresolution NLM filter is combined with wavelet thresholding to form a new image denoising framework, which turns out to be very effective in eliminating noise in real noisy images. Experiments show that the algorithm can reduce image noise effectively in terms of image quality metrics.

The process of extracting external boundary of an object is a very important process for recognizing an object in the image. The proposed extraction method consists of two processes: External Boundary Extraction and Thinning. In the first step, external boundary extraction process separates the region representing the object in the input image. Then, only the pixels adjacent to the background are selected among the pixels constituting the object to construct an outline of the object. The second step, thinning process, simplifies the outline of an object by eliminating unnecessary pixels by examining positions and interconnection relations between the pixels constituting the outline of the object obtained in the previous extraction process. As a result, the simplified external boundary of object results in a higher recognition rate in the next step, the object recognition process.

Conventional techniques for solving the noise problem have problems to generate different results, depending on the image size and weight values of the used masks, and they require many operations by using a complex formula. In this paper, we propose an image enhancement algorithm to solve the noise problem in a simple, yet easy-to-use way. For this purpose, we determined the difference between the noise of the two adjacent pixels for the horizontal and vertical, and for the two diagonal directions that each of the noise problem occurred, and then we got the average value of these pixel values. Then, we solve the noise problem by using the optimal average value in accordance with occurrence of the noise in the horizontal and vertical, and two adjacent pixels in a diagonal direction. As a result, we got the result that the noise solution in a simple, yet easy-to-use method to obtain a resultant image.

Aiming at the problems that the luminance information is not enough and the edge information can not be preserved easily in the process of the image fusion, an effective image fusion method based on nonsubsampled contourlet transform (NSCT) and adaptive guided image filtering is proposed in this paper. Firstly the source images are decomposed by NSCT with multi-scale, multi-direction and shift-invariant properties. The fusion rule of the low-frequency subband coefficients employs the local correlation energy to improve the energy and information of the subband coefficients. To acquire good effects with edge-preserving and noise reduction, the adaptive guided image filtering is introduced to the high-frequency subband coefficients as the fusion rule for the first time, and it can make use of halo-free edge slope enhancement in the fusion process. Finally, the fused image is reconstructed by the inverse NSCT. Experimental results demonstrate that the comprehensive performances of the proposed method are improved in the fusion information, edge and luminance.

In advanced image processing, aerial image processing plays an important role in object extraction such as building extraction, road detection etc. The aerial images captured are usually bound to suffer from Gaussian noise, salt and pepper noise, speckle noise etc. Therefore obtaining of aerial image with high accuracy is very difficult task. A flawless aerial image is inevitable for further object extraction process. There are a number of filtering techniques to detach the noise for preserving the integrity of captured aerial image. In this paper we have applied mean filter, median filter, wiener filter, wavelet transform and curvelet transform for removal of various level of Gaussian noise, salt and pepper noise and speckle noise added separately in an aerial image. The performance of both the transforms and filtering methods are compared in terms of Peak Signal to Noise Ratio (PSNR), Mean Square Error (MSE).

image sequences. Our proposed algorithm uses adaptive weights based on a triangular membership function: Symmetrical, Continuous Function. In this algorithm median filter is used to suppress noise. Experimental results show when the images are corrupted by highdensity Salt and Pepper noise, our fuzzy based algorithm for noise filtering of image sequences, is much more effective in suppressing noise and preserving edges than the previously reported algorithms such as [1-13]. Indeed, assigned weights to noisy pixels are very adaptive so that it well makes use of correlation of pixels. On the other hand, the motion estimation methods are erroneous and in high-density noise they may degrade the filter performance. Therefore, our proposed fuzzy algorithm doesn’t need any estimation of motion trajectory. The proposed algorithm admissibly removes noise without having any knowledge of Salt and Pepper noise density.

Background subtraction is important process in many automatic video content analysis applications. extracting foreground moving objects from video sequences is a crucial step and also a hot topic in computer vision and image processing . This paper presents a new approach in image processing for detection crack in video. This method involves two main steps: First, we use based on Visual Background Extractor (Vibe) to detection moving crack in video. Second steps: Using a suitable threshold in a binary image and classifies all pixels two groups’ background and foreground. And use filter area and changes the area if less than the specific number to back; and using filtering to elimination of residual noise. This paper describes a method for detection crack in video we use Image Processing Techniques. The advantage of this method is clearly and accurate detection of cracks in video. Experimental work shows that our method is improved relatively to the other widely used techniques.

Techniques of generating a single panoramic image by using multiple images are being widely studied in a number of areas, such as computer vision and computer graphics. Generating a panoramic image is a good way of overcoming the limitations of the images obtained from one single camera (e.g., those of picture angles, resolutions, information amounts, etc.) and may be applied in a variety of fields such as virtual reality and robot vision that require wide-angle images. A panoramic image has a great significance in that it can provide a greater sensation of immersion compared to a single image. Currently, there are a variety of techniques of producing panoramic images, but most of them commonly use a method of detecting feature points and matching points in each of the panoramic images they generate. In addition, they use the method of converting images after obtaining homography matrix using the RANSAC (Random Sample Consensus) algorithm that uses matching points. The SURF (Speeded Up Robust Features) algorithm used in this study utilizes the black-and-white and local space information of images when detecting their feature points and is widely used because it provides an outstanding performance in detecting the viewpoints and the changes of the image sizes and is faster than SIFT (Scale Invariant Features Transform) algorithm. However, the SURF algorithm also has its weak point of detecting wrong matching points, which may slow down the performance speed of the RANSAC algorithm and thus increases CPU usage occupation rates. The errors in detecting matching points serve as essential elements of lowering the accuracy and resolutions of panoramic images. In order to minimize these errors, this paper went through an intermediate filtering process of removing wrong matching points using the RGB values of 3×3 region around their coordinates and then presented analysis and evaluation results related to improvements in panoramic image construction & processing and CPU usage occupation rates and the decreasing rates and accuracy of the extracted matching points.

As the state-of-art denoising method, BM3D is capable of achieving good denoising performance by exploiting both the non-local characteristics and sparsity prior knowledge of images. Nevertheless, experimental results show that the dissimilarity measurement defined in BM3D sometimes results in grouping patches with distinct structure. Inspired by the fact about the different impact of noise on patches with various structures, we propose a structure-adaptive image denoising method with 3D collaborative filtering by optimizing the block matching procedure. In our method, the similarity in the variance between patches is incorporated in block matching procedure. Besides, based on the prior knowledge of correlation among patches in the same neighborhood, the spatial distance between the reference patch and the candidate is also taken into account when measuring patches’ dissimilarity. Several numerical experiments demonstrate that the proposed approach achieve better results in PSNR and visual effect than original BM3D.

Digital Image Processing is basically the implementation of a set of computer algorithms for processing digital images. Digital image processing has probative advantages over Analog image processing. In this content, noise like, Gaussian, Salt and Pepper, Speckle and Poisson, is added to an image and then the original edges are restored using various filters and tools. These tools have remarkable alteration on the image and hence they are widely employed. Some of them are wavelet transform, median filter, Weiner filter and many more.

The fusion algorithm of multiple images of which focal lengths are different is necessary to generate visually enhanced composite image. Also, it can be employed as a pre-processing step for edge detection or feature extraction in order to improve the performances. In this paper, we suggest a pixel- based approach for the fusion of multi-focused image. The focus measure of SML(Sum of Modified- Laplacian) is slightly modified by adopting double thresholds. We firstly calculate 3-level SML comparison map which contains intermediate level for ambiguous pixels. And the final 2-level map is generated after reallocating these pixels to get more reliable result and median filtering to reduce the effect of isolated noises. The performances are evaluated by computer simulation which showed that the visual quality of composite image was successfully improved not only subjectively but objectively by the proposed fusion algorithm.

This paper describes a detailed description of segmentation procedure for fingerprint area detection in a digital fingerprint image. Purpose of this procedure is to extract very precisely the fingerprint area and to separate it from the image background. The precise fingerprint area detection is important not only for vendors of minutiae extraction algorithms but also for semantic conformance testing for finger minutiae data in the newly created international standard. Our segmentation procedure was evaluated for real-world scenario, so the used fingerprints were scanned from real dactyloscopic fingerprint cards. These fingerprints were taken from Ground Truth Database of fingerprints (used subset of GTD originally belongs to NIST SD14 and SD29 databases). Our procedure had to deal with specific problems and properties of these images such as handwritten or printed characters, drawings or specific noise in the background or spread over the fingerprint itself. Our approach was compared with three other methods and yields significantly better results than the best of the benchmarked methods.

Adaptive filters are commonly used to remove disturbances corrupting a signal, particularly when a reference signal correlated to the disturbance is available. Accompanying random noise, sinusoidal disturbance or other undesired signals can be removed from the desired signal. When applied to images, the term adaptive filter commonly refers to a filter which changes its features based on the local statistical characteristics of the image and noise (within each image segment). It has also been proposed previously to use the LMS adaptive filter for 2-D image filtering, mostly for random noise removal. In this paper, it is offered to extend the use of the 2-D LMS adaptive filter to other image disturbances, such as a varying frequency sinusoidal disturbance, and the removal of an undesired added image from the desired one. Moreover, it is proposed here that in some cases it may be useful to convert the two-dimensional image to a one-dimensional signal (by chaining its row or columns), and apply a 1-D LMS adaptive filter. The 2-D LMS adaptive filter is demonstrated to work well in restoring an image corrupted by a random noise, a sinusoidal disturbance or a disturbance of an undesired image added to the desired one. In some particular conditions an adaptive filter with no reference signal may also be used to remove the undesired disturbance from the image.

In this paper, on the basis of the anisotropic diffusion mechanism, analyzes emphatically represented by P - M model of diffusion filter principle of several kinds of nonlinear diffusion model, and their respective characteristics and problems. In-depth analysis of the nonlinear diffusion model, the threshold and termination mechanism of combining image geometric structure feature and visual information (gradient, brightness, contrast, structural information), in view of the existing nonlinear diffusion filtering model, the diffusion coefficient depends on the gradient and the problem that the susceptible to noise interference, presents a fidelity term used in image denoising and restoration contain nonlinear wavelet diffusion model, the theoretical analysis and experimental results show that this method is compared with other diffusion model while denoising can keep image edges and details characteristics, image visual effect is better.

In this paper, on the basis of the anisotropic diffusion mechanism, analyzes emphatically represented by P - M model of diffusion filter principle of several kinds of nonlinear diffusion model, and their respective characteristics and problems. In-depth analysis of the nonlinear diffusion model, the threshold and termination mechanism of combining image geometric structure feature and visual information (gradient, brightness, contrast, structural information), in view of the existing nonlinear diffusion filtering model, the diffusion coefficient depends on the gradient and the problem that the susceptible to noise interference, presents a fidelity term used in image denoising and restoration contain nonlinear wavelet diffusion model, the theoretical analysis and experimental results show that this method is compared with other diffusion model while denoising can keep image edges and details characteristics, image visual effect is better.

This paper conducts a research on information loss of local feature existing in the image denoising process and puts forward the method of image sparse denoising of redundant dictionary based on filtering guidance. This method utilizes bias noise (additional noise and image errors after denoised image and the corresponding additional noise deviation) for image sparse expression, and extracts the feature information of bias noise to improve the effectiveness of de-noising. In the first place, based on filtering guidance, the method carries out aftertreatment to bias noise still existing after denoise the image. And then, the method, in the basis of this bias noise, designs a new dictionary training method, and obtains redundant dictionary for image processing through self-adaption. Finally, the method extracts featured texture from bias noise image based on the dictionary mentioned above. And it takes advantage of filtering guidance in combination with featured texture extracting information and denoising image to realize image restoration. According to emulated data, the performance of proposed algorithm should be better than the selected comparing algorithm and be equipped with a better visual recovery effect.