Edge Preserving Interpolation of Digital Images Using Fuzzy Inference

This paper presents a novel edge preserving interpolation method for digital images. This new method reduces drastically the blurring and jaggy artifacts at the high-contrast edges, which are generally found in the interpolated images using conventional methods. This high performance is achieved by two proposed operations: a fuzzy-inference based edge preserving interpolator and a highly oblique edge compensation scheme developed based on an edge orientation detector. The former synthesizes the interpolated pixels to match the image local characteristics. Hence, edge sharpness can be retained. However, due to the small footage of the fuzzy interpolation method, it cannot avoid edge jaggedness along the highly oblique edges that have very sharp angles against one of the coordinates. Therefore, a segment matching technique is developed to identify precisely the orientation of the highly oblique edges. Combining these two techniques, we improve significantly the visual quality of the interpolated images, particularly at the high-contrast edges. Both the synthesized images (such as letters) and the natural scenes (captured by camera) have been tested and the results are very promising.     

Image super-resolution by curve fitting in the threshold decomposition domain

A new curve-fitting scheme is proposed in this paper to produce super-resolution images from a single low-resolution source image. The most unique feature of this method is that the threshold decomposition is performed on the given source image to obtain multiple binary images so that the curve-fitting applied on each resulted binary image can be made very efficient and accurate, thus allowing us to focus on tiny objects and thin structures so as to achieve rather nice visual results even when a large up-scaling factor is used. Two novel techniques are further proposed to improve the visual quality: (1) a spreading technique (applied on some significant pixels detected in each threshold decomposed binary image) is used to remove ladder-like false edges that often appear visually in super-resolution images, and (2) an edge correction (guided by the edge information extracted from the original source image) is used to sharpen all inherent edges. Our results are compared with those achieved by using the state-of-arts techniques, showing the ability of our algorithm to achieve a better visual quality in smooth areas as well as for sharp edges and small objects.     

Similarity relation matrix-based color edge detection

A novel edge detection algorithm for color images was described in this paper. In the proposed method, smoothness of each pixel in color image is firstly calculated by means of similarity relation matrix and is normalized to maximum gray level. In other words, color image in three-dimensional color spaces is mapped into one dimension. Accordingly the edges are performed in such a way that pixels lower than thresholds are assigned to be edge. Thus with proposed method, edge pixels in a color image are detected simultaneously without any complex calculations such as gradient, Laplace and statistical calculations.     

基于图像自身信息的图像边缘检测阈值自动设定方法

从利用边缘像素空间关系和抑制伪边缘的角度出发,提出了一种基于图像自身信息的图像边缘检测阈值自动设定方法,并阐述了其物理意义。本文方法利用图像边缘连续度衡量边缘像素的空间关系,其极大值蕴含着边缘像素增长模式的转换;利用边缘段增量识别伪边缘的出现。对比实验结果表明,本文方法具有良好的稳定性和可靠性。     

冲激信号SAR成像的方位分辨率分析

本文在分析冲激信号SAR成像特点的基础上,在发射和接收均为"超宽带信号"、"大方位积累角"的情况下,推导出了冲激信号SAR方位分辨率的解析表达式,并通过仿真实验验证了其正确性.     

基于统计特征的彩色图像快速插值方法

本文首先阐述了基于统计特征的图像插值方法,该方法通过提取待插入像素所在区域的协方差矩阵和协方差向量,得出适应于边缘位置和方向的插值权重.为了把基于统计特征的图像插值方法应用于彩色图像插值领域,本文提出了以下措施以提高计算速度:仅对Y图像估计插值权重,并同时应用到R、G、B三个分量的插值;对边缘像素应用基于统计特征的图像插值方法,而对非边缘像素应用简单的双线性插值,即混合图像插值方法.这些措施提高了计算速度,并保证了图像质量.实验表明了该算法在计算速度和插值图像质量方面的有效性.     

应用特征点的彩色图像边缘检测算法

传统的边缘检测技主要通过全局图像扫描,寻找大于设定阈值的点,然后进行伪边缘点的筛除以及断裂边缘的连接。本文提出了一种基于图像特征点检测与边缘生长相结合的边缘检测算法。该方法将边缘看成一种特殊的区域,使用图像分割中区域生长的原理来生成边缘。首先在图像中寻找特征点作为边缘生长的种子点,然后以边缘梯度响应和区域相似度为生长规则,以层序遍历方式得到图像边缘。仿真结果显示,本文提出的算法可以减少参与比较的像素个数,去孤立的边缘,保证边缘的连续性和单一像素宽度。      

基于哈夫变换的图像边缘连接

图像边缘的检测可以得到图像中处于边缘上的像素点,由于受到噪声等干扰,一组边缘像素很少能完整地描绘一条边缘。利用哈夫(Hough)变换可以将边缘像素连接成有意义的边缘。现有文献对哈夫变换在极坐标中的应用,存在不同的形式和论述,容易造成概念混淆。详细叙述哈夫变换的基本原理,及在直线检测中的应用。尤其是对极坐标下直线的标准方程,进行详细地推导和论述,从而对哈夫变换的应用进行有益的补充。     

A robust algorithm for reduction of truncation artifact in chemical shift images

A new iterative technique to reduce the ringing artifacts in chemical shift images due to the truncation of the high spatial frequency is presented. In this approach the authors extrapolate the high spatial frequency data guided by the edge information obtained from a high resolution anatomic image of the region of interest. The fact that the edge information obtained from the anatomic image can be off by a few pixels (due to factors such as chemical shift artifact, error in edge detection or misregistration) is taken into account by assuming a confidence interval of several pixels around the anatomic edges. The algorithm is validated on simulated and in vivo data, and excellent results were obtained.     

A new kernel development algorithm for edge detection using singular value ratios

The perceptual quality of an image is very sensitive to the degradation of the edge information which is usually caused by many video signal applications such as super-resolution and denoising. Hence, it is very important to detect and enhance the edge information of the image. In this research work, new sets of kernels for edge detection using ratios of singular values of an image are proposed, which results in more detailed detection of edges in the original image. The parameters, which are the elements of kernel matrices and the threshold value used for producing binary image after convolving the kernels with the image of the proposed method, are optimised to achieve more detailed edge detection of the image. The experimental results show that more detailed edges are detected by the proposed method compared to the conventional edge detection techniques.     

Combining spatial and scale-space techniques for edge detection to provide a spatially adaptive wavelet-based noise filtering algorithm

New methods for detecting edges in an image using spatial and scale-space domains are proposed. A priori knowledge about geometrical characteristics of edges is used to assign a probability factor to the chance of any pixel being on an edge. An improved double thresholding technique is introduced for spatial domain filtering. Probabilities that pixels belong to a given edge are assigned based on pixel similarity across gradient amplitudes, gradient phases and edge connectivity. The scale-space approach uses dynamic range compression to allow wavelet correlation over a wider range of scales. A probabilistic formulation is used to combine the results obtained from filtering in each domain to provide a final edge probability image which has the advantages of both spatial and scale-space domain methods. Decomposing this edge probability image with the same wavelet as the original image permits the generation of adaptive filters that can recognize the characteristics of the edges in all wavelet detail and approximation images regardless of scale. These matched filters permit significant reduction in image noise without contributing to edge distortion. The spatially adaptive wavelet noise-filtering algorithm is qualitatively and quantitatively compared to a frequency domain and two wavelet based noise suppression algorithms using both natural and computer generated noisy images.     

Adaptive bilateral filter for sharpness enhancement and noise removal.

In this paper, we present the adaptive bilateral filter (ABF) for sharpness enhancement and noise removal. The ABF sharpens an image by increasing the slope of the edges without producing overshoot or undershoot. It is an approach to sharpness enhancement that is fundamentally different from the unsharp mask (USM). This new approach to slope restoration also differs significantly from previous slope restoration algorithms in that the ABF does not involve detection of edges or their orientation, or extraction of edge profiles. In the ABF, the edge slope is enhanced by transforming the histogram via a range filter with adaptive offset and width. The ABF is able to smooth the noise, while enhancing edges and textures in the image. The parameters of the ABF are optimized with a training procedure. ABF restored images are significantly sharper than those restored by the bilateral filter. Compared with an USM based sharpening method-the optimal unsharp mask (OUM), ABF restored edges are as sharp as those rendered by the OUM, but without the halo artifacts that appear in the OUM restored image. In terms of noise removal, ABF also outperforms the bilateral filter and the OUM. We demonstrate that ABF works well for both natural images and text images.     

OSED: Object-specific edge detection

Object-specific edge detection (OSED) aims to detect object edges in an image along with classify the edge into object or non-object. It prunes edges which are not belonging to the object class for following processing, such as, feature matching for object detection, localization and three-dimensional reconstruction. In this paper, an OSED method that combines region proposal detectors with deep supervision nets to identify object-specific edges is proposed. It minimizes errors of object proposal by learning from hidden layers. Additionally, it combines features from different scales to detect object edges. In order to evaluate the performance of the OSED, we present two datasets which are captured in real scenes. The OSED method demonstrates a high accuracy of 90% and a high speed of 0.5 s for an image whose size is 512 × 448 pixels on the proposed datasets.     

Texture-based image indexing in the process of lossless data compression

The authors describe a new approach for content-based image indexing and retrieval by extracting texture features from the process of image compression via JPEG-LS. Since the compression technique adopted incorporates local edge detection to formulate predictive values for pixels being encoded, the texture features extracted by the proposed algorithms are also capable of describing image content in terms of edges and shapes of local objects without adding any significant complexity to the original JPEG-LS. While lossless data compression helps in saving storage space automatically for image databases, the extensive experiments also show that this type of feature extraction produces better retrieval results in comparison with existing similar indexing techniques which are carried out without data compression.     

Morphological Image Enlargements

This paper presents a method based in mathematical morphology to enlarge images. It does not make the low pass assumption which is common to all linear interpolation methods and which does not often hold for images. Pixels in smooth areas are properly interpolated by linear methods while those at the edges are not. The method begins with a linear interpolation and a gradient computation. The gradient serves as a measure of confidence about the linear interpolation. Then, the proposed algorithm processes the pixels in a certain order: first pixels with high confidence (smooth zones) of the image and those with a low one (edges) at the end. By doing so, it preserves both slow variations and sharp edges. The method can be applied to other image processing problems, such as edge enhancement or motion vector estimation, where there is an image and confidence information about each pixel.     

Boundary detection using simulation of particle motion in a vectorimage field

This paper introduces a novel approach in image processing based on a vector image model. A major advantage of the model is that it allows vector operations to be performed on an image. An example of a vector operation is the computation of mechanical moments for detecting inhomogeneities in an object or equivalently edges in an image. A new edge operator derived from a vector image model yields an edge vector field analogous to the Hamiltonian gradient field of the image. The distinct feature of the edge vector field is that edge vectors form current loops encompassing the objects. This feature is exploited to develop a new boundary extraction algorithm based on particle motion in a force field. The edge vector field forces a particle to move along the edges while an orthogonal normalized Laplacian gradient vector field guarantees that the particle will not drift away from the edges. The object boundary can be obtained from the convergent path of the particle trajectory. Using a fine stepping factor, the extracted boundary can achieve subpixel accuracy. The proposed algorithm has major advantages over the conventional edge-detection, edge-thinning, and edge-linking techniques in that it effectively utilizes both direction and magnitude of edges. The algorithm is simple, robust and performs very well even on high curvature objects.     

Subpixel edge detection and estimation with amicroprocessor-controlled line scan camera

A microprocessor-controlled line scan camera system for measuring edges and lengths of steel strips is described, and the problem of subpixel edge detection and estimation in a line image is considered. The edge image is assumed to change gradually in its intensity, and the true edge location may be between pixels. Detection and estimation of edges are based on measurement of gray values of the line images at a limited number of pixels. A two-stage approach is presented. At the first stage, a computationally simple discrete-template-matching method is used to place the estimated edge point to the nearest pixel value. Three second-stage methods designed for subpixel estimation are examined. The modified Chebyshev polynomial and the three-point interpolation method do not require much knowledge on the shape of the edge intensity. If the functional form of the edge is known, a least-square estimation method may be used for better accuracy. In the case of nonstationary Poisson noise, a recursive maximum-likelihood method for the first-stage edge detection, followed by subpixel estimation, is proposed     

Edge Detection of Highly Distorted Images Suffering from Impulsive Noise

A new edge detection method for highly distorted images, which suffer from impulsive noise, is introduced. The proposed method comprises three main stages; analysis for the impulsive behavior of the image pixels, restoration of the pixels which have impulsive behavior and finally, estimation of the edges. The simulation results reveal that the proposed method shows better performance than the other methods mentioned in this paper in the cases of preserving the details and detecting the edges correctly and continuously, especially when the noise ratio is very high.     

基于边缘定向增强的各向异性扩散抑噪方法

本文提出了一个新的边缘定向增强扩散模型.针对现有各向异性扩散方程中,边缘增强扩散模型不能正确地对边缘定向,而相干增强扩散模型易在光滑处产生虚假边缘的缺点,本文的模型采用基于非线性光滑算子的边缘定向算子对边缘定向,并根据边缘的位置和方向设置扩散张量的特征根,使其在光滑区域沿边缘方向和垂直边缘方向均具有较大值,而在边缘区域垂直边缘方向值小,沿边缘方向值大,从而达到既保护边缘又去除噪声的目的,在整幅图像上均具有较好的去噪效果.理论分析和数值计算结果均表明,本文方法具有比现有扩散去噪方法更好的去噪效果,同时在峰值信噪比和边缘保护指数方面具有显著优势.