Edge Texture Research Articles

Efficient Combination of Texture and Color Features in a New Spectral Clustering Method for PolSAR Image Segmentation

In this paper, an efficient combination of texture and color features is proposed for polarimetric synthetic aperture radar (PolSAR) image segmentation. In proposed method, the PolSAR image is first segmented using the mean-shift method. Then, in each segment obtained from the mean shift, the first moment of color for the three color components of L* a* b* color space is obtained. On the other hand, a texture feature vector for each pixel of the image is formed corresponding to the texture edge energy at different directions with Gabor filter. Then, a new spectral clustering method is used to combine texture and color features. The results show that the proposed method is effective in PolSAR image segmentation.

Depth dithering based on texture edge-assisted classification

Development in low-cost depth acquisition devices pushes the relevant applications forward, which requires higher quality of the depth maps. In this paper, a novel depth de-noising algorithm is proposed, which dithers the depth map according to its spatial and temporal depth classification around the extracted texture edges. After roughly filling the holes in the depth map by joint bilateral filter, the depth map is partitioned into edge region and non-edge region according to the extracted texture edges. The depth values on each side of the mapped texture edges are dithered adapting to depth classification. Joint spatio-temporal dithering approach is proposed to align the depth edges with the texture edges and to suppress the depth artifacts. Experimental results demonstrate that compared with existing depth de-noising approaches, the proposed algorithm provides much sharper and clearer depth edges and it improves the spatial and temporal consistency of the processed depth, which further benefits the applications requiring the depth of high quality, e.g., 3D rendering.

Watermarking technique based on three-coefficient comparison in DCT domain

In order to hide information, Yu et al. (Appl Res Comput 26(3):1082---1083, 1103, 2009) presented a method by using the numerical value relation of three pixel values in spatial domain. Their algorithm is very easy to realize and has a big hiding capacity. But its visual performance is not very good, because it is based on spatial domain and cannot control the modification value effectively in edge region or texture region of image. Our work embedded watermark by comparing the numerical value relation of three neighboring DCT coefficients with Zig-Zag order of block image to enhance the visual performance. First, each block of the host image was transformed by DCT. Then the DCT coefficient matrix was scanned by Zig-Zag scanning order to ensure the continuity of the neighboring coefficients. And then through modifying the numerical value relation of three Zig-Zag neighboring DCT coefficients, the three corresponding watermark bits were embedded in. To further improve its performance, the algorithm also employed redundant encoding, watermark scrambling encryption techniques, etc. The experiments show that the algorithm can guarantee a better quality of watermarked image, and has stronger robust to general image processing operations than the method proposed by Yu et al. (Appl Res Comput 26(3):1082---1083, 1103, 2009).

A novel adaptive image zooming scheme via weighted least-squares estimation

A critical issue in image interpolation is preserving edge detail and texture information in images when zooming. In this paper, we propose a novel adaptive image zooming algorithm using weighted least-square estimation that can achieve arbitrary integer-ratio zoom (WLS-AIZ) For a given zooming ratio n, every pixel in a low-resolution (LR) image is associated with an n × n block of high-resolution (HR) pixels in the HR image. In WLS-AIZ, the LR image is interpolated using the bilinear method in advance. Model parameters of every n×n block are worked out through weighted least-square estimation. Subsequently, each pixel in the n × n block is substituted by a combination of its eight neighboring HR pixels using estimated parameters. Finally, a refinement strategy is adopted to obtain the ultimate HR pixel values. The proposed algorithm has significant adaptability to local image structure. Extensive experiments comparing WLS-AIZ with other state of the art image zooming methods demonstrate the superiority of WLS-AIZ. In terms of peak signal to noise ratio (PSNR), structural similarity index (SSIM) and feature similarity index (FSIM), WLS-AIZ produces better results than all other image integer-ratio zoom algorithms.

Machined Surface Quality of Pre-Sintered Hardenable PM Steel

The benefit of pre-sintered machining is to avoid machining difficulty of sintered parts especially hardenable PM steels. Pre-sintering treatments of green PM part at temperatures lower than the normal sintering temperature of 1120 °C result in green strength improvement high enough for machining. In this study, the influences of various pre-sintering temperatures and several machining conditions on machined surface quality of pre-sintered PM samples were investigated. The pre-sintered samples were machined by a turning process using a carbide cutting insert with varied cutting speeds at a fixed feed rate and depth of cut without cutting lubricant. Chromium alloyed PM steel (Astaloy® CrM) powder samples with (0.5 wt. %C) and without graphite (0 wt. % C) additions mixed with 1 wt. % of zinc stearate were prepared as green parts by cold compaction in a cylindrical die with diameter of 30 mm. Green density was about 7.00 g/cm3 and height of each sample was controlled by hydraulic pressure and powder weight of 80 g/sample. The green samples were treated by pre-sintering treatment before machining testing. Surface quality of each machined sample was evaluated by average surface roughness and surface texture by SEM analysis and the appearance of outlet edge breakout. The experimental results revealed that the pre-sintered samples with graphite addition showed better surface quality in terms of surface roughness and surface texture and small outlet edge breakout appearance. Moreover, at high pre-sintering temperatures of 900 and 1,000°C, the samples showed similar average surface roughness under the same turning conditions. The obtained surface textures were better than those of the samples pre-sintered at 700 and 800°C. The outlet edge breakout could not be found in the graphite-added samples pre-sintering at 1,000°C.

Text detection and recognition in natural scene with edge analysis

Text plays an important role in daily life because of its rich information, thus automatic text detection in natural scenes has many attractive applications. However, detecting and recognising such text is always a challenging problem. In this study, the authors propose a method which extends the widely‐used stroke width transform by two steps of edge analysis, namely candidate edge recombination and edge classification. A new method that recognises text through candidate edge recombination and candidate edge recognition is also proposed. In the step of candidate edge recombination, they use the idea of over‐segmentation and region merging. To separate text edge from background, the edge of the input image is first divided into small segments. Then, neighbour edge segments are merged, if they have similar stroke width and colour. Through this step, each character is described by one candidate boundary. In the step of boundary classification, candidate boundaries are aggregated into text chains, followed by chain classification using character‐based and chain‐based features. To recognise text, the grey image is extracted based on the location of each candidate edge after the step of candidate edge recombination. Then, histogram of gradient features and a classifier are used to recognise each character. To evaluate the effectiveness of their method, the algorithm is run on the ICDAR competition dataset and Street View Text database. The experimental results show that the proposed method provides promising performance in comparison with the existing methods.

A feature-matching method for side-scan sonar images based on nonlinear scale space

We report a novel feature-matching method for side-scan sonar images. The method uses nonlinear diffusion filtering to build a nonlinear scale space. The noise-reduction performance is enhanced via nonlinear diffusion filtering, and the improved Perona–Malik diffusion equation results in a more distinct edge and line texture in the side-scan sonar image. The modified feature descriptor reduces the dimensionality of the feature vector so that the computational expense is reduced. Experimental results show that the method provides improved noise-reduction performance and better accuracy than SIFT, SURF, and other state-of-the-art feature-matching algorithms.

An Effective Edge and Texture Based Approach towards Curved Videotext Detection and Extraction

In present day video text greatly helps video indexing and retrieval system as they often carry significant semantic information. Video text analysis is challenging due to varying background, multiple orientations and low contrast between text and non-text regions. Proposed approach explores a new framework for curved video text detection and recognition where from the observation that curve text regions can be well defined by edges size and uniform texture, Probable curved text edge detection is accomplished by processing wavelet sub bands followed by text localization by utilizing fast texture descriptor LU-transform. Binarization is achieved by maximal H-transform. A Connected Component filtering method followed by B-Spline curve fitting on centroid of each character vertically aligns each oriented character. The aligned text string is recognized by optical character recognition (OCR). Experiments on various curved video frames shows that proposed method is efficacious and robust in detecting and recognizing curved videotext.

Boundary detection using double-opponency and spatial sparseness constraint.

Brightness and color are two basic visual features integrated by the human visual system (HVS) to gain a better understanding of color natural scenes. Aiming to combine these two cues to maximize the reliability of boundary detection in natural scenes, we propose a new framework based on the color-opponent mechanisms of a certain type of color-sensitive double-opponent (DO) cells in the primary visual cortex (V1) of HVS. This type of DO cells has oriented receptive field with both chromatically and spatially opponent structure. The proposed framework is a feedforward hierarchical model, which has direct counterpart to the color-opponent mechanisms involved in from the retina to V1. In addition, we employ the spatial sparseness constraint (SSC) of neural responses to further suppress the unwanted edges of texture elements. Experimental results show that the DO cells we modeled can flexibly capture both the structured chromatic and achromatic boundaries of salient objects in complex scenes when the cone inputs to DO cells are unbalanced. Meanwhile, the SSC operator further improves the performance by suppressing redundant texture edges. With competitive contour detection accuracy, the proposed model has the additional advantage of quite simple implementation with low computational cost.

A NOVEL APPROACH FOR EDGE DETECTION OF LOW CONTRAST SATELLITE IMAGES

Abstract. Discrimination between texture edges and geometrical edges is very difficult in low contrast images. Satellite images are low contrast images. It is important to extract the edges that are not clearly visible in case of Satellite images. The present work encompasses a new edge detection algorithm using newly constructed differentiator. Chebyshev polynomial based fractional order differentiator has been used for filtering operation on an image. High pass and Low pass filters are designed with the concept of Quadrature Mirror Filter (QMF). Pre-processing has been performed by using this filter. Sobel edge detection method has been applied on this pre-processed image. The algorithm has been tested with two different satellite images.

A Novel Image Inpainting Method Based on Sparse Decomposition

Redundant Discrete Wavelet Transform (RDWT) and Wavelet Atomic Transform (WAT) are proposed in this paper as a couple of dictionaries to get the structure and texture basing on morphological component decomposition. Then, basing on the fact that the structure and texture have different characteristics, in this paper we use curvature driven diffusion model and Criminisi texture synthesis method to inpaint the structure and texture respectively. At last, compound the inpainted structure and texture and get the inpainting result. The experiment results show the new method can not only decompose the image very well, but also inpaint the image with strong and fairing edge, complete and clear texture .This method shows better results in image inpainting compared to the classical ones.

Fast Text Processing for Interlaced Sports or News Videos

Interlaced scanning has been widely used as a trade-off solution between picture quality and transmission bandwidth since the invention of television. During the past decades, various interlaced-to-progressive conversion algorithms have been proposed to improve subjective quality or coding efficiency. However, almost all the researchers concentrate on general cases, without making full use of specific application scenarios. Based on extensive investigations, eliminating visual artifacts in areas of subtitles and station captions for interlaced sports and news videos is still an unsolved problem, which will be addressed in this paper. Firstly, motion estimation is performed between field pictures. Secondly, text edge detection is proposed for sports and news videos. Finally, different processing strategies are applied to text regions and non-text regions. Experimental results show that the proposed method can generate much better text content than existing algorithms. In addition, it is quite stable for non-text parts.

A New Efficient Text Detection Method for Image Spam Filtering

Detection of text in images plays an important role in many situations such as video retrieval, annotation, indexing, and content analysis. In information security to filter image spam, one main feature can be used is text contents in image. Extracting text features from image spam needs efficient text detection. Obfuscating techniques used by spammers such as noisy background, wavy text and text with different colors pose challenges to the text detection process. In this paper, we present a text detection method that addresses these challenges. The contribution of this research consists of two parts: a) a new edge operator can specifically be used to detect text edges, and b) proposing of text detection method for image spam filtering that can detect obfuscated text. The proposed method Accumulated Text Extraction (ATE) works for detecting horizontal and vertical lines and intersecting them, then rules are used to determine the text area and reduce non text area. The approach focuses on using non-machine learning methods with simple calculations. ATE shows encouraging results which can be efficiently used in image spam filtering. Besides its robustness against obfuscating methods in image spam, ATE shows efficient performance when used for scene text detection.

English

Text detection in images and videos is quite challenging task. Text detection is a method of detecting text character from images and videos whose size, orientation, repeating pattern and other variations makes it difficult to detect. Many different methods for text detection are proposed based on their variation in text and properties of text. For text detection, first step is to Initialization of Text Model. To initialize, extract region of text or single character from image or video frame by edge detection. Second step is to describe the features of region containing text object based on different variation of text such as geometry of text, color of text, motion of text object and edge of text. Last step is to remove of non text region from the image or video frames by comparing with threshold values of different text features. In this review work, different types of variation and problems of text associated in the Images and videos are explained, and different method for text detection in images and videos are studied.

Extracting Corresponding Point Based on Texture Synthesis for Nearly Flat Textureless Object Surface

Since the image feature points are always gathered at the range with significant intensity change, such as textured portions or edges of an image, which can be detected by the state-of-the-art intensity based point-detectors, there is nearly no point in the areas of low textured detected by classical interest-point detectors. In this paper we describe a novel algorithm based on affine transform and graph cut for interest point detecting and matching from wide baseline image pairs with weakly textured object. The detection and matching mechanism can be separated into three steps: firstly, the information on the large textureless areas will be enhanced by adding textures through the proposed texture synthesis algorithm TSIQ. Secondly, the initial interest-point set is detected by classical interest-point detectors. Finally, graph cuts are used to find the globally optimal set of matching points on stereo pairs. The efficacy of the proposed algorithm is verified by three kinds of experiments, that is, the influence of point detecting from synthetic texture with different texture sample, the stability under the different geometric transformations, and the performance to improve the quasi-dense matching algorithm, respectively.

Age Group Classification of Facial Images Using Rank Based Edge Texture Unit (RETU)

Human beings can easily categorize a person's age group from a facial image where as this ability has not been promising in the computer vision community. To address this problem very important area of research, the present paper proposes a novel scheme of age classification system using features derived from co-occurrence parameters using Rank based Edge Texture Unit (RETU). The Co-occurrence Matrix (CM) on RETU characterizes the relationship between the neighbouring edge values, while preserving local information. The novelty of the proposed RETU is it classifies the age of human into seven categories i.e. in the age groups of 1-10, 11-20, 21-30, 31-40, 41-50, 51-60, and greater than 60. The TU of the proposed RETU ranges from 0 to 17 and thus reduces overall complexity in evaluating features from CM. The co occurrence features extracted from the RETU provide complete facial image information for age classification purpose. The RETU reduces each 3×3 sub image into 2×2 sub image while preserving the texture features and thus reduces the overall dimensionality of the image.

Study of anisotropic diffusion model based on pulse coupled neural network and image entropy

In image processing, most of the anisotropic diffusion models based on partial differential equation use gradient information to detect image edge. If the image edge is seriously polluted by noise, these methods would not be able to detect image edge, so the edge features cannot be retained. Pulse coupled neural network (PCNN) has the property that similar input neurons can generate pulse at the same time; this property is used to process the noisy image, and we can get an image entropy sequence. The image entropy sequence which will be used as an edge detecting operator is introduced into the diffusion equation, and this will not only reduce the defects produced when the gradient is used as an edge detecting operator so it is easily affected by the noise, but the area image information can also retain more completely. Then, we will use the rule of minimum cross entropy to search for a minimum threshold, which would satisfy the condition that the information difference between noisy image and denoised image is the minimum. The optimal threshold designed will control diffusion intensity reasonably, and the anisotropic diffusion model based on pulse coupled neural network and image entropy (PCNN-IEAD) can be established. Analysis and simulation results show that the proposed model preserves more image information than the classical ones. It removes the image noise and at the same time protects the edge texture details of the image; the proposed model retains the area image information more completely, the performance indexes can also confirm the superiority of the new model. In addition, the operating time of the proposed model is shorter than that of the classical models, therefore, the proposed model may be the ideal one.

Collective edge modes near the onset of a graphene quantum spin Hall state

Graphene subject to a strong, tilted magnetic field exhibits an insulator-metal transition tunable by tilt-angle, attributed to the transition from a canted antiferromagnetic (CAF) to a ferromagnetic (FM) bulk state at filling factor zero. We develop a theoretical description for the spin and valley edge textures in the two phases, and the implied evolution in the nature of edge modes through the transition. In particular, we show that the CAF has gapless neutral modes in the bulk, but supports gapped charged edge modes. At the transition to the FM state the charged edge modes become gapless and are smoothly connected to the helical edge modes of the FM state. Possible experimental consequences are discussed.

Improved sampling for gradient-domain metropolis light transport

We present a generalized framework for gradient-domain Metropolis rendering, and introduce three techniques to reduce sampling artifacts and variance. The first one is a heuristic weighting strategy that combines several sampling techniques to avoid outliers. The second one is an improved mapping to generate offset paths required for computing gradients. Here we leverage the properties of manifold walks in path space to cancel out singularities. Finally, the third technique introduces generalized screen space gradient kernels. This approach aligns the gradient kernels with image structures such as texture edges and geometric discontinuities to obtain sparser gradients than with the conventional gradient kernel. We implement our framework on top of an existing Metropolis sampler, and we demonstrate significant improvements in visual and numerical quality of our results compared to previous work.

Image Signal Enhancement based on Fractional Differential Technologies

To describe the ability of image fractional differential in texture detail enhancement and to study the lateral inhibition principle, multiform masks for digital image fractional differential and their operation rules are discussed. The theoretical basis of fractional differential in modulation and demodulation is also analyzed. Through the derivation on the relation between fractional differential and signal time-frequency analysis, we can acquire the separability of two-dimensional fractional differential under certain conditions. Mach’s phenomenon generated in image texture detail is studied by us from two aspects: optic nerve model and signal processing, to propose novel masks. Then its operating rules for digital image processing based on fractional differential are proposed. The experiments show that our scheme can effectively reserve better information of edge texture detail during the denoising process, especially for weak texture and gray with little change. The total enhancing effect is obviously superior to integer order differential operator