Edge Texture Research Articles

Image Processing and Sparse Resolution for Underwater Imaging

Super-resolution is useful for weak-target image reconstruction. The technique, however, is not yet suitable for underwater imaging implementation due to the complexity of underwater environment. In this paper, an image processing and sparse super-resolution is proposed to make it more suitable for underwater images. In this paper, image processing and image super-resolution are combined. It firstly used histogram equalization to increase local gray value and extent the gray value range to mate the details of the image texture range more distinct. By this we can raise the attention of the human eyes; then use Canny edge enhancement operator to make the edge of the image more clear; finally, by combining classical underwater point spread function (PSF) and the semi image blind restoration method, the image quality are further improved. The experimental results showed that the image recovered by this method had shaper edge and clearer textures.

Variational infrared image enhancement based on adaptive dual-threshold gradient field equalization

Infrared images are characterized by low signal to noise ratio (SNR) and fuzzy texture edges. This article introduces the variational infrared image enhancement algorithm based on gradient field equalization with adaptive dual thresholds. Firstly, we transform the image into gradient domain and get the gradient histogram. Then, we do the gradient histogram equalization. By setting adaptive dual thresholds to qualify the gradients, the image is prevented from over enhancement. The total variation (TV) model is adopted in the reconstruction of the enhanced image to suppress noise. It is shown from experimental results that the image edge details are significantly enhanced, and therefore the algorithm is qualified for enhancement of infrared images in different applications.

The Contourlet Transform with Multiple Cycles Spinning for Catenary Image Denoising

In the catenary images, noise and artifacts are introduced due to the acquisition techniques and systems, which may influence the judgement of catenary quailty and working states. In this paper, the contourlet transform (CT) with performances of multi-scale, multi-resolution and anisotropy is proposed, which can be effectively applied to image denoising. However, the CT hasn’t the spinning invariance, which will lead to the Gibbs-like phenomena. In this paper, the CT with multiple cycle spinning is introduced for image denoising, which can effectively eliminate the visual artifacts due to the lack of translational invariance. Meanwhile, the different Laplacian pyramid (LP) filters and directional filter banks (DFB) are proposed to test noisy images. Finally, test the influence of cycle spinning numbers for denoising effects by using different cycle spinning times. The experiment results show that the proposed method has excellent denoising performance in terms of the signal-to-noise ratio (SNR) and the visual effects, which is also superior to some other existing methods in overcoming the Gibbs-like phenomena, background smoothing and preservation of edge sharpness and texture. DOI : http://dx.doi.org/10.11591/telkomnika.v12i5.4446

Haze Removal for a Single Image Using Adaptive Template Dark Channel Prior

In this paper, we propose an adaptive template method based on the dark channel prior. The method combines with the haze imaging model to haze removal for a single image. This method can effectively remove haze from a single input image. According to the characteristics of the image itself and the haze removal effect of the different template we divide the input image into flat region, edge region and texture region. Then, select the lager size template dispose the flat region and use midrange or minitype template dispose the edge region and texture area. Experimental results demonstrate that the proposed algorithm has very good performance for fog removal and retains the image details more effectively.

An Image Fusion Method Based on the Combination of Lifting Wavelet and Median Filter

At present, image fusion universally exists problem that fuzzy edge, sparse texture. To solve this problem, this study proposes an image fusion method based on the combination of Lifting Wavelet and Median Filter. The method adopts different fusion rules. For the low frequency coefficient, the low frequency scale coefficients have had the convolution do the square respectively to get enhanced edge of the image fusion. Then the details information of original image is extracted by measuring region characteristics. For high frequency coefficient, the high frequency parts are denoised by the Median Filter, and then neighborhood spatial frequency and consistency verification fusion rule is adopted to the fusion of detail sub-images. Compared with Weighted Average and Regional Energy , experimental results show that edge and texture information are the most. Method in study solves the fuzzy edge and sparse texture in a certain degree,which has strong practical value in image fusion.

Contour detection based on horizontal interactions in primary visual cortex

Neurons in the primary visual cortex (V1) respond differently to a simple visual element presented in isolation from when it is embedded within a cluttered scene. The contextual influences are mediated by horizontal connections within V1. Inspired by these visual cortical mechanisms, a contour detection model is proposed. Different from previous researches, the facilitation and the suppression are unified modulated by surrounding elements in the image without separating regions of excitatory and inhibitory lateral connections specifically. The proposed model can selectively extract object contours while reducing non-meaningful texture edges and process angular visual features such as corners and T-junctions effectively. The experiments demonstrate the effectiveness of the proposed model.

1P1-W07 テンプレートマッチングに基づくテクスチャ境界検出(ロボットビジョン)

We proposes an image processing algorithm for texture edge detection. In the proposed method, multiple template images are obtained from each texture area which consists of periodic pattern, and are used for segregating the texture in the image. This algorithm detect not only texture edge but also defects in the target image. We examined the performance of the algorithm in an image inspection of mask patterns used for manufacturing industrial products.

Text Extraction of Vehicle Number Plate and Document Images Using Discrete Wavelet Transform in MATLAB

Text Extraction from colour images is a challenging task in computer vision. The concept of text extraction is derived from the vehicle plate recognization and their characters extractions individually. Some examples of the applications are automatic image indexing, visual impaired people assistance or optical character reading, keyword searching in a document image. The continuous research has been done majorly that focuses on the text detection and localization in images. The paper focuses on the colour images text extraction using Two Dimensional Discrete Wavelet Transform (2D-DWT). The transform used for DWT is HAAR DWT with mathematical morphological concept. The speed of Haar DWT operates the fastest among all wavelets because its coefficients are either 1 or -1. It is one of the reasons that Haar DWT is used to detect edges of candidate text regions. Image sub bands contain both text edges and non-text edges. The work is carried out in MATLAB 12.0 image processing tool and simulated results are tested on vehicle plate text extraction and the document images.

Distinguishing Texture Edges From Object Boundaries in Video

One of the most fundamental problems in image processing and computer vision is the inherent ambiguity that exists between texture edges and object boundaries in real-world images and video. Despite this ambiguity, many applications in computer vision and image processing often use image edge strength with the assumption that these edges approximate object depth boundaries. However, this assumption is often invalidated by real world data, and this discrepancy is a significant limitation in many of today's image processing methods. We address this issue by introducing a simple, low-level, and patch-consistency assumption that leverages the extra information present in video data to resolve this ambiguity. Through analyzing how well patches can be modeled by simple transformations over time, we can obtain an indication of which image edges correspond to texture edges versus object boundaries. Our approach is simple to implement and has the potential to improve a wide range of image and video-based applications by suppressing the detrimental effects of strong texture edges on regularization terms. We validate our approach by presenting results on a variety of scene types and directly incorporating our augmented edge map into existing image segmentation and optical flow applications, showing results that better correspond to object boundaries.

Automatic Allocation Algorithm of Holographic Correlation Peaks Based on Neighborhood Variance

To increase the quality of holographic correlation peaks, near stoichiometric Mg:Fe:LiNbO3crystals is used as memory and correlator due to its better uniformity and optical quality. A correlation peaks allocation algorithm based on neighborhood variance increment is presented, which is on the foundation of neighborhood entropy. This algorithm not only can distinguish target from edge texture, but also is not influenced by background intensity. With 1000 holograms stored, it takes less than 4 seconds to allocate correlation peaks accurately against correlation output image with 1024×768 pixels, and a satisfied recognition result is also achieved.

视觉感知结合学习的自然图像轮廓检测

A model for detecting contours in natural images is presented by combining the visual perceptual mechanisms of nonclassical receptive fields and machine learning. When the stimuli inside and outside the receptive field (RF) form a precise spatial configuration, the surround stimuli will enhance the response of the central stimulus, on the other hand, surround inhibition will reduce the responses to homogeneous elements. The two different visual mechanisms are used to enhance the well-organized structures and reduce the non-meaningful distractors engendering from texture fields, respectively. We approach the task of information combination as a supervised learning problem using the logistic regression model, where we will learn the combination rules from the ground truth data. Our experiments demonstrate that the model can dramatically suppress texture edges and reduce spurious contours, and meanwhile can enhance the responses to the elements with coherent spatial configurations and avoid ground-truth contours missed by the detector. Finally, we evaluate quantitatively our results with the Berkeley segmentation dataset and compare with related work. The proposed method not only provides a biologically feasible scheme for contour detection but also contributes to further understand visual mechanisms.

A Denoising Mixed Noise Method Based on Median Filter and Lifting Wavelet Technology

In image digitization and transmission, images often suffer contamination inevitably. The noises in images often consist of Gaussian noise and impulse noise. The common denoising algorithms are capable of removing single one of them. In order to remove those two types of noise, a composite algorithm is proposed. Firstly, based on median filter, an impulse noise detection algorithm is used to filter impulse noise. Secondly, adaptive directional lifting wavelet (ADL) and normal lifting wavelet is combined to suppress noise from image signal and protect the texture edge from loss simultaneously. Meanwhile an improved half-soft threshold is used for normal lifting wavelet. At last, simulations show that this technology can suppress Gaussian and impulse noise in image efficiently．

Automatic extraction of feather quill based on Normalized cut algorithm

A modified Normalized cut(Ncut) method considering the texture weight was proposed to effectively segment feather quill.The weight including texture information enhanced association on each edge of similar texture and reduced interference of the similar region.Narrow unidirectional expansion method with a region-scalable fitting term was used to optimize the initial boundary for the final result and eliminated boundary leakage of the feather quill.The experimental results show that the proposed method can realize boundary extraction of feather quill efficiently and pave the way for the next step research of crease detection.

Sketch drawing method with hierarchical sketch texture

The existing technique of computer Non-Photorealistic Rendering(NPR) on sketch drawing always has some differences from hand-drawing.In order to solve the problem,a method was designed which can generate sketch levels,and integrate with multi-pixel-wide lines to generate sketches which has sketch texture with sense of hierarchy.First,pretreat the original image by denoising and enhancement,adopt differentiation to extract multi-pixel-wide contour lines,use dilation and erosion and subtract the result to generate multi-layer bilateral of contour lines,then take the processing of motion-blur to multi-layer bilateral of contour lines to generate sketch texture,and finally,integrate edge and sketch texture,thus completing images sketching.Through lots of experiments on images of different features,it is proved that the results of the method are more close to hand-drawing style.

Image Segmentation with Texture Gradient and Spectral Clustering

For some applications the whole image cannot be processed directly because it is inefficient and impractical. Segmentation results in a set of images that cover the entire image. This work proposes a two stage segmentation method, which effectively process both the textured and non-textured regions. Dual Tree Complex Wavelet Transform, an extension of discrete wavelet transform, extracts texture feature from the image and orientation median filtering reduces the double edge effect at the texture edges. Watershed transform of Gaussian gradient of combined texture and non-texture feature give the first stage segmentation. The initial segmentation into super-pixels reduces computational burden and the second stage uses spectral clustering technique to cluster these primitive regions. General Terms Feature, segmentation, interpolation, clustering, algorithm.

Text Extraction of Colour Images using Mathematical Morphology & HAAR Transform

Digital image processing is an ever expanding and dynamic area with applications reaching out into our daily life such as digital signature, authentication, surveillance, medicine, space exploration, automated industry inspection and many others areas. Theseapplications are involved in different processes like image enhancement, object detection, features extraction, colour imaging etc. Implementation of such applications on a general purpose computer can be easier, but every time it is not efficient due to additional constraints on memory and other peripheral devices. Out of the five senses - sight, hearing, touch, smell and taste, humans use to perceive their environment. Among all,sight of images is the most powerful. More than 99% of the activity of the human brain is involved in processing images from the visual cortex. A visual image is rich in information. There is an efficient yet simple method to extract text regions from video sequences or static images. The speed of Haar discrete wavelet transform (DWT) operates the fastest among all wavelets because its coefficients are either 1 or -1. It is one of the reasons that Haar DWT is used to detect edges of candidate text regions. Image sub bands contain both text edges and non-text edges. The intensity of the text edges is also different from that of the non-text edges. Therefore, thresholdingis used to preliminary remove the non-text edges. Text regions of colour images are composed of horizontal edges, vertical edges and diagonal edges. Morphological dilation operators as AND, OR are applied to connect isolated text edges of each detail component sub-band in a transformed binary image. The simulation is carried out on MATLAB 2012 image processing tool.

Spatial Arrangement in Texture Discrimination and Texture Segregation

We investigated the role of spatial arrangement of texture elements in three psychophysical experiments on texture discrimination and texture segregation. In our stimuli, oriented Gabor elements formed an iso-oriented and a randomly oriented texture region. We manipulated (1) the orientation similarity in the iso-oriented region by adding orientation jitter to the orientation of each Gabor; (2) the spatial arrangement of the Gabors: quasi-random or regular; and (3) the shape of the edge between the two texture regions: straight or curved. In Experiment 1, participants discriminated an iso-oriented stimulus from a stimulus with only randomly oriented elements. Experiment 2 required texture segregation to judge the shape of the texture edge. Experiment 3 replicated Experiment 2 with Gabors of a smaller spatial extent in a denser arrangement. We found comparable performance levels with regular and quasi-random Gabor positions in the discrimination task but not in the segregation tasks. We conclude that spatial arrangement plays a role in a texture segregation task requiring shape discrimination of the texture edge but not in a texture discrimination task in which it is sufficient to discriminate an iso-oriented region from a completely random region.

Fault Diagnosis Based on Mathematical Morphology and Probabilistic Neural Network for Progressing Cavity Pump Well

The problem of a scarce consideration of screw pump well pump diagram graphic information affects the diagnosis technology promotion and utilization to some extent. The method, through which the shape features in pump diagram graphic state and parameter information been directly extracted, and then a method based on Mathematical morphology is also presented. Mathematical morphology filters of open-close operator to realize graphics edge texture feature extraction. After feature digitized, using a probabilistic neural network to identify fault. The practical application shows the classification accuracy rate is above 90%.

Contour detection improved by frequency domain filtering of gradient image

We propose an intermediate computational step, frequency domain filtering of gradient image, to improve contour detection performance of gradient-based edge detectors. This step is inspired by analyzing the spectrum distribution of object contours and texture edges in the frequency domain of gradient image. We illustrate the principle and effect of this step by adding it to the Canny edge detector. The resulting operator can selectively retain object contours and region boundaries, and meanwhile can dramatically reduce non-meaningful elements caused by textured background. We use several types of images to compare the proposed method and other related methods qualitatively and quantitatively. Experimental results show that the proposed method can effectively enhance the contour detection of Canny edge detector and achieves similar detection performance to two other related methods but runs faster.

LS-SVM-based image segmentation using pixel color-texture descriptors

Image segmentation remains an important, but hard-to-solve, problem since it appears to be application dependent with usually no a priori information available regarding the image structure. Moreover, the increasing demands of image analysis tasks in terms of segmentation results' quality introduce the necessity of employing multiple cues for improving image-segmentation results. In this paper, we present a least squares support vector machine (LS-SVM) based image segmentation using pixel color-texture descriptors, in which multiple cues such as edge saliency, color saliency, local maximum energy, and multiresolution texture gradient are incorporated. Firstly, the pixel-level edge saliency and color saliency are extracted based on the spatial relations between neighboring pixels in HSV color space. Secondly, the image pixel's texture features, local maximum energy and multiresolution texture gradient, are represented via nonsubsampled contourlet transform. Then, both the pixel-level edge color saliency and texture features are used as input of LS-SVM model (classifier), and the LS-SVM model (classifier) is trained by selecting the training samples with Arimoto entropy thresholding. Finally, the color image is segmented with the trained LS-SVM model (classifier). This image segmentation not only can fully take advantage of the human visual attention and local texture content of color image, but also the generalization ability of LS-SVM classifier. Experimental results show that our proposed method has very promising segmentation performance compared with the state-of-the-art segmentation approaches recently proposed in the literature.