Search

We present an algorithm for risk prediction of road surface grip where skidding and sliding occur as main road surface problems. Prediction is done by defining a fine texture classification of the properties of road aggregate. In an experimental setup, data acquisition is performed with a supervised mobile vehicle scanning system, using a vehicle equipped with a camera and temperature sensor during movement along an arterial road. Image processing is performed by testing four texture feature extraction methods: Gabor filters, wavelet transform, gray level co-occurence matrix, and edge histogram descriptor, among which the Gabor transform shows the best results. The extraction of texture feature vectors follows by statistical algorithms for measuring feature vector similarity and reference vector selection, leading to image texture classification. The algorithm itself is upgraded by incorporating simultaneous surface temperature measurements in order to create and validate the final fine surface texture classification. The roads are classified and segmented into high-, medium-, and low-risk roads according to skid danger, enabling the creation of a map of high-risk zones. We validate our risk prediction algorithm by referring to crash rate data from the Road Traffic Safety Agency of Serbia database. This algorithm enables the location and mapping of high-risk zones and can be used as a support for autonomous driving and navigation.

The existing methods—such as sieving, microscope, light scattering, sedimentation, and electrical induction for pigment size detection—require sampling or scattering the mineral pigments, which will inevitably cause damage to the films painted by mineral pigments. A new detection method based on run length texture analysis is proposed to nondestructively detect the pigment size in the mineral paint film. The films painted by mineral pigments with preknown pigment sizes are contactlessly captured by CCD microscope under diffused light. Gray transform, histogram equalization, and median filtering are implemented to preprocess the captured images, and then the run length texture parameters are extracted from the preprocessed images. A parametric relationship between the extracted parameters and the preknown size is established to predict the pigment size in mineral paint film nondestructively. Burnt carnelian is selected as the sample to verify the feasibility of the proposed method. Results show that the max detection error of the proposed method is 5.548 μm and can be applied to the size detection of the mineral pigments used in mineral paint film.

The extraction of texture features from images faces two new challenges: large-scale databases with diversified textures, and varying imaging conditions. We propose a novel method termed multiscale blob features (MBF) to overcome these two difficulties. MBF analyzes textures in both resolution scale and gray level. Proposed statistical descriptors effectively extract structural information from the decomposed binary images. Experimental results show that MBF outperforms other methods on combined large-scale databases (VisTex+Brodatz+CUReT+OuTex). Moreover, experimental results on the University of Illinois at Urbana-Champaign database and the entire Brodatz's atlas show that MBF is invariant to gray-level scaling and image rotation, and is robust across a substantial range of spatial scaling.

A superpixel segmentation algorithm called multi-density peaks clustering (MDPC) is proposed. By selecting a sufficient number of local density maximum pixels from the image as cluster centers to depict the image texture, the boundary of the object can be captured very accurately. The algorithm framework of MDPC is divided into three steps. First, the local density of pixel is defined, and the local density maximum pixels are calculated. Then, the local density maximum pixels are used as cluster centers, and the global optimal search, which is based on the path-to-point idea, is used to complete the clustering of the remaining non-cluster center pixels to realize the initial segmentation. Finally, superpixels are obtained by merging the initial segments according to the size of the segments and the distance between adjacent segments. In quantitative comparisons, MDPC is compared with 13 state-of-the-art superpixel segmentation algorithms in three image segmentation datasets. The experimental results show that MDPC achieves better performance in terms of boundary recall, boundary precision, achievable segmentation accuracy, undersegmentation error, and explained variation. And the qualitative comparisons show that the proposed algorithm has obvious advantages over other superpixel segmentation algorithms in image detail description and boundary adherence. Finally, the practicability and stability of MDPC are further demonstrated by the application of image segmentation. The source code of MDPC will be available at https://github.com/zhaojianaaa.

To effectively detect the surface cracks of subway tunnels, an automatic tunnel crack detection system based on machine vision is presented. Aiming at the problems of environmental complexity and low contrast in subway tunnels, the image texture feature is first enhanced by the methods of frequency domain filtering and spatial differencing. Then, depending on the characteristics of the tunnel cracks in question, the crack propagation method is used to extract the complete cracks. Finally, broken cracks are connected during processing, and the method of combining projection and threshold is used to determine the crack types. At the same time, characteristics such as the length, width, and area of the cracks are obtained. The experimental results show that the presented methods can effectively extract complete cracks in complex tunnel environments. The identification error of tunnel crack parameters meets the actual engineering requirements.

Compressed sensing (CS) is a signal processing framework for effectively reconstructing signal from a small number of measurements obtained by linear projections of the signal. It is an ongoing challenge for the real-time image reconstruction of the computational imaging, including single pixel imaging based on CS. We built a block-based CS (BCS) image reconstruction framework via a deep learning network with smoothed projected Landweber (SPL). A fully connected network performs both BCS linear sensing and non-linear reconstruction stages, and SPL removes the blocking artifacts due to incorporate Wiener filtering into projected Landweber (PL) method at each iteration. The sensing matrix and nonlinear prediction operator are jointly optimized, and the smoothing filtering is coalesced into the PL framework for eliminating high-frequency oscillatory blocking artifact. Experimental results reveal that the optimized scheme outperforms the approach only based on deep neural network. The reconstruction quality can be improved while being only slightly slower, especially the gain of structural similarity is significantly better than peak signal-to-noise ratio, and the reconstruction image texture details are vivid and natural. At 10% sensing rate, the structural similarity maximum (minimum) gain reaches 0.098 (0.021). The proposed approach is not only far superior to other state-of-the-art CS algorithms in terms of reconstruction time and quality but also comparable with up-to-date deep learning methods.

Local binary patterns (LBP) are well documented in the literature as descriptors of local image texture, and their histograms have been shown to be well-performing texture features. A method for texture description that is based on the α-cutting approach is presented. The presented approach combines basic definitions from the fuzzy set theory with the main concept of LBP descriptors, which resulted in powerful texture features. The general method is introduced and defined and its binary, ternary, and quinary versions evaluated in tests produced excellent results in texture classification. The performance of our method is presented by an extensive evaluation on four datasets—KTH-TIPS2b, UIUC, Virus, and Brodatz. The introduced descriptors are compared with some of the classical approaches—LBP, improved LBP, local ternary pattern, including one very promising LBP variant—median robust extended LBP (MRELBP), as well as with three non-LBP methods, based on deep convolutional neural networks approaches—ScatNet, FV-AlexNet, and fisher vector based very deep VGG. Our method effectively deals with many classification challenges and exceeds most of the other approaches. It outperforms the classical approaches on all datasets, even in its simplest binary version. It outperforms the MRELBP descriptor on the UIUC, KTH-TIPS2b, and Brodatz datasets and reaches a better classification performance than two out of the three deep learning approaches on the KTH-TIPS2b dataset.

In the field of image processing, it has been a challenging task to obtain a complete foreground that is not uniform in color or texture. Unlike other methods, which segment the image by only using low-level features, we present a segmentation framework, in which high-level visual features, such as semantic information, are used. First, the initial semantic labels were obtained by using the nonparametric method. Then, a subset of the training images, with a similar foreground to the input image, was selected. Consequently, the semantic labels could be further refined according to the subset. Finally, the input image was segmented by integrating the object affinity and refined semantic labels. State-of-the-art performance was achieved in experiments with the challenging MSRC 21 dataset.

A hybrid image allows multiple image interpretations to be modulated by the viewing distance. It can be constructed on the basis of the multiscale perceptual mechanisms of the human visual system by combining the low and high spatial frequencies of two different images. The hybrid image was introduced as an experimental tool for visual recognition study in terms of spatial frequency perception. To produce a compelling hybrid image, the original hybrid image synthesis method could only use similar shapes of source images that were aligned in the edges. If any two different images can be hybrid, it would be beneficial as a new experimental tool. In addition, there is no measure for the actual perception of spatial frequency, whether a single spatial frequency or both spatial frequencies are perceived from the hybrid stimulus. This paper describes two methods for synthesizing a hybrid image from dissimilar shape images or unaligned images; this hybrid image is known as an “edge-alignment-free hybrid image.” A noise-inserted method can be done by intentionally inserting and enhancing noises into the high-frequency image. With this method, the low-frequency blobs are covered with high-frequency noises when viewed up close. A color-inserted method uses complementary color gratings in the background of the high-frequency image to emphasize the high-frequency image when viewed up close, whereas the gratings disappear when viewed from far away. To ascertain that our approach successfully separates the spatial frequency at each viewing distance, we measured this property using our proposed assessment method. Our proposed method allows the experimenter to quantify the probability of perceiving both spatial frequencies and a single spatial frequency in a hybrid image. The experimental results confirmed that our proposed synthesis methods successfully hid the low-frequency image and emphasized the high-frequency image at a close viewing distance. At the same time, the perception of the low-frequency image was not disturbed when the image was viewed from far away.

We deal with the problem of blind parameter estimation of signal-dependent noise from mono-component image data. Multispectral or color images can be processed in a component-wise manner. The main results obtained rest on the assumption that the image texture and noise parameters estimation problems are interdependent. A two-dimensional fractal Brownian motion (fBm) model is used for locally describing image texture. A polynomial model is assumed for the purpose of describing the signal-dependent noise variance dependence on image intensity. Using the maximum likelihood approach, estimates of both fBm-model and noise parameters are obtained. It is demonstrated that Fisher information (FI) on noise parameters contained in an image is distributed nonuniformly over intensity coordinates (an image intensity range). It is also shown how to find the most informative intensities and the corresponding image areas for a given noisy image. The proposed estimator benefits from these detected areas to improve the estimation accuracy of signal-dependent noise parameters. Finally, the potential estimation accuracy (Cramer-Rao Lower Bound, or CRLB) of noise parameters is derived, providing confidence intervals of these estimates for a given image. In the experiment, the proposed and existing state-of-the-art noise variance estimators are compared for a large image database using CRLB-based statistical efficiency criteria.

This paper proposes an improved exemplar-based inpainting using image segmentation, image quilting, and spatial blending to reduce the excessively propagating texture and pathological geometric configurations. A graph-based segmentation is used to generate an initial segmented map. The segmented regions are merged by a K -means approach with texture similarity to generate a further texture-based segmented map, which reduces the candidate source space and determines whether to utilize image quilting in the iterative updating process. An improved spatial blending is introduced as a postprocess to smooth the filled image when the filling process is completed. Experimental results show that this exemplar-based inpainting approach has a good visual effect and reduces the computing time.

Texture information plays an important role in rendering true objects, especially with the wide application of image-based three-dimensional (3-D) reconstruction and 3-D laser scanning. This paper proposes a seamless texture mapping algorithm to achieve a high-quality visual effect for 3-D reconstruction. At first, a series of image sets is produced by analyzing the visibility of triangular facets, the image sets are clustered and segmented into a number of optimal reference texture patches. Second, the generated texture patches are sequenced to create a rough texture map, then a weighting process is adopted to reduce the color discrepancies between adjacent patches. Finally, a multiresolution decomposition and fusion technique is used to generate the transition section and eliminate the boundary effect. Experiments show that the proposed algorithm is effective and practical for obtaining high-quality 3-D texture mapping for 3-D reconstruction. Compared with traditional methods, it maintains the texture clarity while eliminating the color seams, in addition, it also supports 3-D texture mapping for big data application.

Thousands of pavement images are collected by road authorities daily for condition monitoring surveys. These images typically have intensity variations and texture nonuniformities that make their segmentation challenging. The automated segmentation of such pavement images is crucial for accurate, thorough, and expedited health monitoring of roads. In the pavement monitoring area, well-known texture descriptors, such as gray-level co-occurrence matrices and local binary patterns, are often used for surface segmentation and identification. These, despite being the established methods for texture discrimination, are inherently slow. This work evaluates Laws texture energy measures as a viable alternative for pavement images for the first time. k -means clustering is used to partition the feature space, limiting the human subjectivity in the process. Data classification, hence image segmentation, is performed by the k -nearest neighbor method. Laws texture energy masks are shown to perform well with resulting accuracy and precision values of more than 80%. The implementations of the algorithm, in both MATLAB® and OpenCV/C++, are extensively compared against the state of the art for execution speed, clearly showing the advantages of the proposed method. Furthermore, the OpenCV-based segmentation shows a 100% increase in processing speed when compared to the fastest algorithm available in literature.

Image oversegmentation creates small, compact, and irregularly shaped regions subject to further clustering. Consideration of texture characteristics can improve the resulting quality of the clustering process. Existing methods based on an orthogonal transform into frequency domain can extract texture features of arbitrarily shaped regions only from inscribed rectangles. We propose a method for extracting texture features of entire arbitrarily shaped image regions using orthogonal transforms. Furthermore, we introduce a mathematically correct method for unifying spectral dimensions that is necessary for accurate comparison and classification of spectra with different dimensions. The proposed method is particularly suitable for classifying areas with periodic and quasiperiodic textures. Our approach exploits the texture periodification property of certain orthogonal transforms that is based on insertion of zeros into the spectrum. We identified some of those orthogonal transforms which possess this important property and also provide mathematical proofs of our claims. Last, we show that inclusion of luminance and chrominance components into the feature vector increases the precision of the proposed method which then becomes suitable for natural scene images as well.

Image description and annotation is an active research topic in content-based image retrieval. How to utilize human visual perception is a key approach to intelligent image feature extraction and representation. This paper has proposed an image feature descriptor called the local structure co-occurrence pattern (LSCP). LSCP extracts the whole visual perception for an image by building a local binary structure, and it is represented by a color-shape co-occurrence matrix which explores the relationship of multivisual feature spaces according to visual attention mechanism. As a result, LSCP not only describes low-level visual features integrated with texture feature, color feature, and shape feature but also bridges high-level semantic comprehension. Extensive experimental results on an image retrieval task on the benchmark datasets, corel-10,000, MIT VisTex, and INRIA Holidays, have demonstrated the usefulness, effectiveness, and robustness of the proposed LSCP.

The low quality of an image is often characterized by low contrast and blurred edge details. Gradients have a direct relationship with image edge details. More specifically, the larger the gradients, the clearer the image details become. Robust image reconstruction enhancement based on Gaussian mixture model estimation is proposed here. First, image is transformed to its gradient domain, obtaining the gradient histogram. Second, the gradient histogram is estimated and extended using a Gaussian mixture model, and the predetermined function is constructed. Then, using histogram specification technology, the gradient field is enhanced with the constraint of the predetermined function. Finally, a matrix sine transform-based method is applied to reconstruct the enhanced image from the enhanced gradient field. Experimental results show that the proposed algorithm can effectively enhance different types of images such as medical image, aerial image, and visible image, providing high-quality image information for high-level processing.

Active shape model (ASM) is a statistically parametrical model. It is widely used for facial feature extraction in face images. An ASM method is proposed. First, intensity values used in original ASM cannot provide enough information for model searching, which is sensitive to lighting conditions and so on. So we use a measure which indicates the orientation of structure at each pixel instead of intensity value to represent image texture. In addition, a new method is adopted for building a local profile model, which makes full use of texture information around the landmarks. Experimental results show that the improved ASM can locate face features more accurately than other ASM methods, and our method is more robust to poses, expression, and illumination variations.

We propose an interactive object segmentation method which learns feature-specific segmentation parameters based on a single image. The first step is to design discriminative features for each pixel, which integrate four kinds of cues, i.e, the color Gaussian mixture model (GMM), the graph learning-based attribute, the texture GMM, and the geodesic distance. Then we formulate the segmentation problem as a conditional random field model in terms of fusing multiple features. While an image-specific parameter setting is practical in interactive segmentation, the efficiency of learning process highly depends on the type of user interaction and the designed features. We propose a feature-specific parameter learning strategy to learn model parameters, in which an offline training stage is not required and parameters are computed according to some sparsely labeled pixels on the basis of a single image. Extensive experiments show that the proposed segmentation model performs well for segmenting images with a weak boundary, texture, or cluttered background. Comparative experiment results demonstrate that our method can achieve both qualitative and quantitative improvements over other state-of-the-art interactive segmentation methods.

Image completion is an active subject in image and video processing, which deals with the recovery of original data. Most previous image completion techniques required extensive searches to find the most suited texture to repair the damaged area. In addition, visual artifacts tend to appear when the damaged area is large. We present a fast texture synthesis and image completion method that does not require an extensive search process. The proposed method is based on gray-level co-occurrence matrix and weighted two-side hole filling. The method gives high quality results compared with state-of-the-art methods. It reduces the time from hundreds of seconds to a few milliseconds and is able to repair large damaged areas without artifacts or shadows.

From the texture analysis of image content, we propose a steganalytic method to detect spatial domain steganography in grayscale images. First of all, based on the local linear vectors, which are selected carefully and sensitive to image texture, images are decomposed into several textural detail subbands by the local linear transform (LLT). Then the statistical distribution of the LLT coefficient is modeled by using the generalized Gaussian distribution. Finally, novel textural features of the LLT coefficient histogram and cooccurrence matrix are extracted for steganalyzers implemented by the support vector machine. Extensive experiments are performed on four diverse uncompressed image databases and seven typical spatial domain steganographic algorithms, such as the highly undetectable stego. The results reveal that the proposed scheme is universal for detecting spatial domain steganography. By comparison with other well-known feature sets, our presented feature set offers the best performance under most circumstances.

We propose the integration of color and texture cues as an improvement of a rough set-based seg- mentation approach, previously implemented using only color features. Whereas other methods ignore the infor- mation of neighboring pixels, the rough set-based approximations associate pixels locally. Additionally, our method takes into account pixel similarity in both color and texture features. Moreover, our approach does not require cluster initialization because the number of segments is determined automatically. The color cues correspond to the a and b channels of the CIELab color space. The texture features are computed using a standard deviation map. Experiments show that the synergistic integration of features in this framework results in better segmentation outcomes, in comparison with those obtained by other related and state-of-the-art methods. © The Authors. Published by SPIE under a Creative Commons Attribution 3.0 Unported License. Distribution or repro- duction of this work in whole or in part requires full attribution of the original publication, including its DOI. (DOI: 10.1117/1.JEI.23.2 .023003)

We propose a color image segmentation approach based on rough set theory elements. Main con- tributions of the proposed approach are twofold. First, by using an adaptive threshold selection, the approach is automatically adjustable according to the image content. Second, a region-merging process, which takes into account both features and spatial relations of the resulting segments, lets us minimize over-segmentation issues. These two proposals allow our method to overcome some performance issues shown by previous rough set theory-based approaches. In addition, a study to determine the best suited color representation for our segmentation approach is carried out, determining that the best results are obtained using a perceptually uniform color space. A set of qualitative and quantitative tests over a comprehensive image database shows that the proposed method produces high-quality segmentation outcomes, better than those obtained using the pre- vious rough set theory-based and standard segmentation approaches. © The Authors. Published by SPIE under a Creative Commons Attribution 3.0 Unported License. Distribution or reproduction of this work in whole or in part requires full attribution of the original publication, including its DOI. (DOI: 10.1117/1.JEI.23.1.013024)

We describe a texture description algorithm, designed in the wavelets domain, to reduce the dimension of an existing texture-based feature vector and improve on the existing texture description algorithm in terms of both effectiveness and efficiency. We first demonstrate that the estimated texture directions at different wavelet decomposition scales are very similar. Thus, a new texture description with explicit direction representation can be constructed to improve discrimination capability. Second, we propose a subband integration scheme to further improve the texture description and achieve robustness to rotation of texture patterns. Third, a range of successful texture description elements developed in the pixel domain are applied to the LL subband and added to the texture descriptor for further enhancement of the proposed algorithm. Extensive testing, benchmarked by the existing techniques, shows that the proposed algorithm not only reduces the sensitivity of retrieval to image texture rotation, but also improves the retrieval accuracy. © 2006 SPIE and IS&T.

We present an effective solution for unsupervised texture segmentation by taking advantage of the latent Dirichlet allocation (LDA) model. LDA is a generative topic model that is capable of hierarchically organizing discrete data including texts and images. We propose a new texture model by connecting texture primitives to the topic of LDA. The model is able to extract the characteristic features of a texture primitive and group them into a topic based on their frequencies of co-occurrence. Here, the feature descriptor is the connection of Haar-like features of multiple sizes. The segments of an image are finally obtained by identifying the homogeneous regions in the corresponding topic assignment map. The evaluation results for synthetic texture mosaics, remote sensing images, and natural scene images are illustrated.

Colorizing grayscale images so that the resulting image appears natural is a hard problem. Previous colorization algorithms generally use just the luminance information and ignore the rich texture information, which means that regions with the same luminance but different textures may mistakenly be assigned the same color. A novel automatic texture-map-based grayscale image colorization method is proposed. The texture map is generated with bilateral decomposition and a Gaussian high pass filter, which is further optimized using statistical adaptive gamma correction method. The segmentation of the spatial map is performed using locally weighted linear regression on its histogram in order to match the grayscale image and the source image. Within each of the spatial segmentation, a weighted color-luminance correspondence is achieved by the results of locally weighted linear regression. The luminance-color correspondence between the grayscale image and the source image can thus be used to colorize the grayscale image directly. By considering the consistency of both color information and texture information between two images, various plausible colorization results are generated using this new method.

We improved an image segmentation algorithm based on neutrosophic set (NS) and extended the modified method into color image segmentation. The original NS image segmentation approach transformed the images into NS domain, which is described using three membership sets: T, I, and F. Then two operations, α-mean and β-enhancement operations were employed to reduce the set indeterminacy. Although this method was quite successful in image segmentation application, some drawbacks still exist, such as oversegmentation and fixed α and β parameters. Thus, a new algorithm is proposed to overcome these limitations of the NS-based image segmentation algorithm. Then, the new modified method is extended into color image segmentation. The NS-based image segmentation algorithm is applied to each color channel independently. Then each channel is moved to a matrix column, respectively, to construct the input matrix to the γ-K-means clustering. Experiments are conducted on a variety of images, and our results are compared with those new existing segmentation algorithm. The experimental results demonstrate that the proposed approach can segment the color images automatically and effectively.

Successful execution of tasks such as image classification, object detection and recognition, and scene classification depends on the definition of a set of features able to describe images effectively. Texture is among the features used by the human visual system. It provides information regarding spatial distribution, changes in brightness, and description regarding the structural arrangement of surfaces. However, although the visual human system is extremely accurate to recognize and describe textures, it is difficult to define a set of textural descriptors to be used in image analysis on different application domains. This work evaluates several texture descriptors and demonstrates that the combination of descriptors can improve the performance of texture classification.

In this paper, a novel adaptive color image watermarking scheme based on the just noticeable distortion (JND) model in balanced multiwavelet domain is proposed. The balanced multiwavelet transform can achieve orthogonality, symmetry, and high order of approximation simultaneously without requiring any input prefiltering, which makes it a good choice for image processing. According to the properties of the human visual system, a novel multiresolution JND model is proposed in balanced multiwavelet domain. This model incorporates the spatial contrast sensitivity function, the luminance adaptation effect, and the contrast masking effect via separating the sharp edge and the texture. Then, based on this model, the watermark is adaptively inserted into the most distortion tolerable locations of the luminance and chrominance components without introducing the perceivable distortions. Experimental results show that the proposed watermarking scheme is transparent and has a high robustness to various attacks such as low-pass filtering, noise attacking, JPEG and JPEG2000 compression.

Moving object detection in the presence of dynamic backgrounds remains a challenging problem in video surveillance. Earlier work established that the background subtraction technique based on a covariance matrix descriptor is effective and robust for dynamic backgrounds. The work proposed herein extends this concept further, using the covariance-matrix descriptor derived from local textural properties, instead of directly computing from the local image features. The proposed approach models each pixel with a covariance matrix and a mean feature vector and the model is dynamically updated. We made extensive studies with the proposed technique to demonstrate the effectiveness of statistics on local textural properties.

Texture analysis plays an important role in content-based image retrieval and other areas of image processing. It is often desirable for the texture classifier to be rotation and scale invariant. Furthermore, to enable real-time usage, it is desirable to perform the classification efficiently. Toward these goals, we propose several enhancements to the multiresolution Gabor analysis. The first is a new set of kernels called Slit, which can replace Gabor wavelets in applications where high computational speed is desired. Compared to Gabor, feature extraction using Slit requires only 11 to 17% of the numeric operations. The second is to make the features more rotation invariant. We propose a circular sum of the feature elements from the same scale of the feature vector. This has the effect of averaging the feature vector from all orientations. The third is a slide-matching scheme for the final stage of the classifier, which can be applied to different types of distance measures. Distances are calculated at slightly different scales, and the smallest value is used as the actual distance measures. Experimental results using different image databases and distance measures show distinct improvements over existing schemes.

It is difficult to precisely detect all impulsive noise in color images due to the nonstationarity caused by edges and fine details. For many pixels, we can not absolutely classify them as noisy or noise-free, but can only describe them using the likelihood that they are corrupted by impulsive noise. Based on this consideration, we present a new filtering solution to removing impulsive noise from color images. The proposed method first utilizes the unit transforms of quaternions to represent the chromaticity difference of two color pixels, and then divides the image into noise-free and possible noisy pixels. Finally it performs adaptive weighted vector median filtering operations on only the possible noisy pixels to suppress noise. The new weighting mechanism is based on a joint spatial/quaternion-chromaticity criterion, which ensures that pixels with different contamination likelihoods have different contributions to the filter's output. The extensive simulation results indicate that the proposed method significantly outperforms some other well-known multichannel filtering techniques.

The goal of image segmentation is to partition an input image into a set of regions. In mathematical morphology, the recon- struction of images from markers has proven to be useful in morpho- logical filtering and image segmentation. The utilization of a criterion in the problem of the image reconstruction from an image marker has been partially treated elsewhere. We further investigate this idea and extend it to the problem of image reconstruction from labeled markers by proposing a new method based on the "watershed" transforma- tion as an alternative in image segmentation. The image gradient is considered as a topographic relief that is flooded (similarly as in a nor- mal watershed). However, a criterion is added in this reconstruction process that enables the flexibility to separate structures of interest. Following the flooding analogy on topographic reliefs, this flooding process is limited to certain zones to control the recovering process of structures shapes. Experimental results are provided. A compar- ison with a viscous watershed is performed to show the differences between them. The technique is applied mainly in the biomedical do- main, although the technique can generally be applied to other areas.

A novel unsupervised texture image segmentation usinga multilayer data condensation spectral clustering algorithm is pre-sented. First, the texture features of each image pixel are extractedby the stationary wavelet transform and a multilayer data condensa-tion method is performed on this texture features data set to obtaina condensation subset. Second, the spectral clustering algorithmbased on the manifold similarity measure is used to cluster the con-densation subset. Finally, according to the clustering result of thecondensation subset, the nearest-neighbor method is adopted toobtain the original image-segmentation result. In the experiments,we apply our method to solve the texture and synthetic apertureradar image segmentation and take self-tuning k-nearest-neighborspectral clustering and Nystrom methods for baseline comparisons.The experimental results show that the proposed method is morerobust and effective for texture image segmentation. © 2010 SPIEand IS&T. DOI: 10.1117/1.3455990

Computational technology has fundamentally changed many aspects of our lives. One clear evidence is the development of artificial-vision systems, which have effectively automated many manual tasks ranging from quality inspection to quantitative assessment. In many cases, these machine-vision systems are even preferred over manual ones due to their repeatability and high precision. Such advantages come from significant research efforts in advancing sensor technology, illumination, computational hardware, and image-processing algorithms. Similar to the Special Section on Quality Control by Artificial Vision published two years ago in Volume 17, Issue 3 of the Journal of Electronic Imaging, the present one invited papers relevant to fundamental technology improvements to foster quality control by artificial vision, and fine-tuned the technology for specific applications. We aim to balance both theoretical and applied work pertinent to this special section theme. Consequently, we have seven high-quality papers resulting from the stringent peer-reviewing process in place at the Journal of Electronic Imaging. Some of the papers contain extended treatment of the authors work presented at the SPIE Image Processing: Machine Vision Applications conference and the International Conference on Quality Control by Artificial Vision. On the broad application side, Liu et al. propose an unsupervised texture image segmentation scheme.more » Using a multilayer data condensation spectral clustering algorithm together with wavelet transform, they demonstrate the effectiveness of their approach on both texture and synthetic aperture radar images. A problem related to image segmentation is image extraction. For this, O'Leary et al. investigate the theory of polynomial moments and show how these moments can be compared to classical filters. They also show how to use the discrete polynomial-basis functions for the extraction of 3-D embossed digits, demonstrating superiority over Fourier-basis functions for this task. Image registration is another important task for machine vision. Bingham and Arrowood investigate the implementation and results in applying Fourier phase matching for projection registration, with a particular focus on nondestructive testing using computed tomography. Readers interested in enriching their arsenal of image-processing algorithms for machine-vision tasks should find these papers enriching. Meanwhile, we have four papers dealing with more specific machine-vision tasks. The first one, Yahiaoui et al., is quantitative in nature, using machine vision for real-time passenger counting. Occulsion is a common problem in counting objects and people, and they circumvent this issue with a dense stereovision system, achieving 97 to 99% accuracy in their tests. On the other hand, the second paper by Oswald-Tranta et al. focuses on thermographic crack detection. An infrared camera is used to detect inhomogeneities, which may indicate surface cracks. They describe the various steps in developing fully automated testing equipment aimed at a high throughput. Another paper describing an inspection system is Molleda et al., which handles flatness inspection of rolled products. They employ optical-laser triangulation and 3-D surface reconstruction for this task, showing how these can be achieved in real time. Last but not least, Presles et al. propose a way to monitor the particle-size distribution of batch crystallization processes. This is achieved through a new in situ imaging probe and image-analysis methods. While it is unlikely any reader may be working on these four specific problems at the same time, we are confident that readers will find these papers inspiring and potentially helpful to their own machine-vision system developments.« less

We propose a novel unsupervised multiresolution adaptive and progressive gradient-based color-image segmentation algorithm (MAPGSEG) that takes advantage of gradient information in an adaptive and progressive framework. The proposed methodology is initiated with a dyadic wavelet decomposition scheme of an arbitrary input image accompanied by a vector gradient calculation of its color-converted counterpart in the 1976 Commission Internationale de l'Eclairage (CIE) L*a*b* color space. The resultant gradient map is used to automatically and adaptively generate thresholds to segregate regions of varying gradient densities at different resolution levels of the input image pyramid. At each level, the classification obtained by a progressively thresholded growth procedure is integrated with an entropy-based texture model by using a unique region-merging procedure to obtain an interim segmentation. A confidence map and nonlinear spatial filtering techniques are combined, and regions of high confidence are passed from one resolution level to another until the final segmentation at the highest (original) resolution is achieved. A performance evaluation of our results on several hundred images with a recently proposed metric called the normalized probabilistic Rand index demonstrates that the proposed work computationally outperforms published segmentation techniques with superior quality.

We propose an image-understanding algorithm for iden- tifying and ranking regions of perceptually relevant content in digital images. Global features that characterize relations between image regions are fused in a probabilistic framework to generate a region ranking map (RRM) of an arbitrary image. Features are introduced as maps for spatial position, weighted similarity, and weighted ho- mogeneity for image regions. Further analysis of the RRM, based on the receiver operating characteristic curve, has been utilized to gen- erate a binary map that signifies region of interest in the test image. The algorithm includes modules for image segmentation, feature extraction, and probabilistic reasoning. It differs from prior art by using machine learning techniques to discover the optimum Baye- sian Network structure and probabilistic inference. It also eliminates the necessity for semantic understanding at intermediate stages. Experimental results indicate an accuracy rate of 90% on a set of 4000 color images that are publicly available and compare favor- ably to state-of-the-art techniques. Applications of the proposed al- gorithm include smart image and document rendering, content- based image retrieval, adaptive image compression and coding, and automatic image annotation. © 2010 SPIE and IS&T.

Image classification has become an important topic in multimedia processing. Recently, neural network-based methods have been proposed to solve the classification problem. Among them, the radial basis function neural network (RBFNN) is the most popular architecture, because it has good learning and approximation capabilities. However, traditional RBFNNs are sensitive to center initialization. To obtain appropriate centers, it needs to find significant features for further RBF clustering. In addition, the training procedure of a traditional RBFNN is time consuming. Therefore, in this work, a combination of a self-organizing map (SOM) and learning vector quantization (LVQ) neural networks is proposed to select more appropriate centers for an RBFNN, and a modular RBF neural network (MRBFNN) is proposed to improve the classification rate and to speed up the training time. Experimental results show that the proposed MRBFNN has better performance than those of the traditional RBFNN, the discrete wavelength transform (DWT)-based method, the tree structured wavelet (TWS), the discrete wavelet frame (DWF), the rotated wavelet filter (RWF), and the wavelet neural network based on adaptive norm entropy (WNN-ANE) methods.

We consider the problem of classifying textures. First, we consider images where the orientation of the texture is known. Then, we consider the classification of textures where the orientation is unknown. Last, classification in real scenes is considered. A wide variety of techniques are tested using the Outex framework. We introduce a new grayscale multiscale texture classification method based on a class of morphological filters called sieves. The method, denoted Tex-Mex because it extracts TEXture features using Morphological EXtrema filters, is shown to be among the best performing texture feature extraction methods. Tex-Mex features can be computed rapidly and are shown to be more robust and compact than the alternatives. Furthermore, they may be applied over windows of arbitrary size and orientation, a useful attribute when classifying texture in real scenes.

Although depth information of a scene is not stored while taking a picture, some depth information is retained in the captured image. A method is proposed for generating the depth map of a single image based on different scene categories. The image is first classified into a category based on color and texture information, and then the depth map of the image is generated according to the different scene categories. The depth map can be used to generate stereo binocular image pairs by left- and right-shifting the original image. Then, the stereoscopic image with three-dimensional (3-D) visual effect can be viewed through a 3-D stereo display. The experiments showed that the proposed method works well, producing a satisfactory stereoscopic effect.

An algorithm is presented for the detection of textured areas in natural images. Texture detection has potential application to image enhancement, tone correction, defect detection, content classification, and image segmentation. For example, texture detection may be useful for object detection when combined with color models and other descriptors. Sky, e.g., is generally smooth, and foliage is textured. The texture detector presented here is based on the intuition that texture in a natural image is comprised of many components. The measure we develop examines the structure of local regions of the image. This structural approach enables us to detect both structured and unstructured textures at many scales. Furthermore, it distinguishes between edges and texture, and also between texture and noise. Automatic detection results are shown to match human classification of corresponding image areas.

An efficient texture classification method is proposed that considers the effects of both the rotation and scale of texture im- ages. In our method, the Gabor wavelets are adopted to extract local features of an image and the statistical properties of its gray- level intensities are used to represent the global features. Then, an adaptive, circular orientation normalization scheme is proposed to make the feature invariant to rotation, and an elastic cross- frequency searching mechanism is devised to reduce the effect of scaling. Our method is evaluated based on the Brodatz album and the Outex database, and the experimental results show that it out- performs the traditional algorithms. © 2008 SPIE and

We present the development of an objective evaluation method based on the integration of X-illumination, morphological fractal analysis and Bayes classifier that aims at characterizing the seam-puckering appearance. The experimental results in our research demonstrate that a highly significant correlation coefficient can be achieved between the estimated grades and the technician-generated grades; the presented method is insensitive to the color/texture of fabrics, thus showing the potential use of our newly developed method to evaluate the seam-puckering appearance objectively and quantitatively.

We present an innovative way to simultaneously perform feature extraction and classification for the quality-control issue of surface grading by applying two multivariate statistical projection methods: SIMCA and PLS-DA. These tools have been applied to compress the color texture data that describe the visual appearance of surfaces (soft color texture descriptors) and to directly perform classification using statistics and predictions from the projection models. Experiments have been carried out using an extensive ce- ramic images database (VxC TSG) comprised of 14 different mod- els, 42 surface classes, and 960 pieces. A factorial experimental design evaluated all the combinations of several factors affecting the accuracy rate. These factors include the tile model, color repre- sentation scheme (CIE Lab, CIE Luv, and RGB), and compression/ classification approach (SIMCA and PLS-DA). Moreover, a logistic regression model is fitted from the experiments to compute accu- racy estimates and study the effect of the factors on the accuracy rate. Results show that PLS-DA performs better than SIMCA, achieving a mean accuracy rate of 98.95%. These results outper- form those obtained in a previous work where the soft color texture descriptors in combination with the CIE Lab color space and the k-NN classifier achieved an accuracy rate of 97.36%. © 2008 SPIE

The importance of the perceived quality measurement is fundamental for many image processing applications, such as compression, acquisition, restoration, enhancement, and reproduction. Color information is also of great importance for the perceived image quality, although perceived information is mainly represented by luminance. We present a computational and memory-efficient no-reference image quality assessment model independent of JPEG and JPEG2000 coded color images based on local regions. We also present the discrimination algorithm for these two types of coded images. The features of local regions are blockiness around the block boundary, average absolute difference between adjacent pixels within the block, and zero crossing rate within the block of the image. We validate the performance of our model on our subjective database, which shows good quality prediction performance, and the model's generalization ability is also verified on the other database.