The surface-shape operator based shading-tolerant method for multiscale image analysis

This paper introduces new features for describing image patterns. We integrate the concepts of multiscale image analysis, aura matrix to define image features, and to obtain the features having robustness with illumination variations and shading effects, we analyse images based on the topographic structure described by the surface-shape operator. Then, illustrate usefulness of the proposed features with texture classifications. Results show that the proposed features extracted from multiscale images work much better than those from a single scale image, and confirm that the proposed features have robustness with illumination and shading variations. By comparisons with the multiresolution simultaneous autoregressive features using Mahalanobis distance and Euclidean distance, the proposed multiscale features give better performances for classifying the entire Brodatz textures (1966).

Urban is composed of different critical levels of organization where interactions are stronger within levels than among levels, it is a complex system. Image analysis performed at a unique scale for urban applications is doomed to be incomplete and misleading. This paper presents a multiscale image analysis method based on wavelet transform and watershed transform for urban system. It performs image analysis on multiple scale images, and can catch comprehensive information of urban system for different applications. The procedure consists of three main steps: multiscale representation, image segmentation, post-processing. First, multiscale represent creates multiresolution images using a wavelet transform and a modified inverse wavelet transform; a smooth and multiscale presentation of original image is provided. Second, image segmentation segments the multiscale images using watershed segment algorithm. Finally, post-processing merges over-segment regions and smoothes watersheds, then, multiscale image objects are attained. This study was conducted in Nanjing, Jiangsu province, China, using IKONOS satellite data. Experimental results demonstrate that the proposed method is suitable for describing the complex urban system. Both visual and quantitative methods are applied to evaluation the experimental results, and reveal the relation of objects within and among levels.

Two dilemmas arise in inferring shape information from shading. First, depending on the rendering physics, images can change significantly with (even) small changes in lighting or viewpoint, while the percept frequently does not. Second, brightness variations can be induced by material effects-such as pigmentation-as well as by shading effects. Improperly interpreted, material effects would confound shading effects. We show how these dilemmas are coupled by reviewing recent developments in shape inference together with a role for colour in separating material from shading effects. Aspects of both are represented in a common geometric (flow) framework, and novel displays of hue/shape interaction demonstrate a global effect with interactions limited to localized regions. Not all parts of an image are perceptually equal; shape percepts appear to be constructed from image anchor regions.

In the real world, the biological visual system is more efficient than the machine visual system in analyzing visual information. Physiology theories show that this efficiency owes to the multi-layer neural network in human visual system, in which every layer accomplishes different tasks and is related with other layers. The low-level stages of the human visual system, especially the retina, can provide certain scale information for the high-level stages of visual system through using the non-classical receptive field (nCRF) mechanism. This mechanism that the nCRF size can be adjusted automatically by ganglion cell (GC) can achieve a multi-scale image analysis. The results, reflecting the distribution of the image information, can be shared by several algorithms or processes solving different visual tasks, such as contour detection and image segmentation. A model of multi-scale image analysis based on GC has been proposed in this paper, which retains the key information and reduces the redundancy information for the further stages of the visual system. Experimental results on N-cut and contour detection show that this multi-scale image analysis model provides distinctive improvement for these image processing tasks.

Most popular image classification methods mainly focus on classification ability rather than recognizing new things. However, human lay emphasis on cognition first and then classification, which is closely related to human memory system. Though many memory models have been proposed, they are studied in word list whereas the reports about natural images are still limited. This paper proposes a memory model for image recognition and classification based on convolutional neural network and Bayesian decision. First the image feature is extracted by convolutional neural network and stored in binary form. Then the representation, storage and retrieval processes of visual images are modeled. The test image feature vector is matched in parallel to the studied image vectors, and the likelihood values are calculated. Finally, the odd that the test image belongs to a new class is computed based on all likelihood values. If the odd value is greater than a certain threshold, the test image is regarded as new; otherwise, the Bayesian decision rule for image classification is performed. Experimental results on Caltech-101 and Caltech-256 datasets show that the proposed method can perform well in image recognition and classification tasks. And the hit probability of the method is higher than two typical methods, SRC and ELM, at present while the false alarm rate is far lower than them.

PurposeOne of the main reasons for increase in mortality rate in woman is breast cancer. Accurate early detection of breast cancer seems to be the only solution for diagnosis. In the field of breast cancer research, many new computer-aided diagnosis systems have been developed to reduce the diagnostic test false positives because of the subtle appearance of breast cancer tissues. The purpose of this study is to develop the diagnosis technique for breast cancer using LCFS and TreeHiCARe classifier model.Design/methodology/approachThe proposed diagnosis methodology initiates with the pre-processing procedure. Subsequently, feature extraction is performed. In feature extraction, the image features which preserve the characteristics of the breast tissues are extracted. Consequently, feature selection is performed by the proposed least-mean-square (LMS)-Cuckoo search feature selection (LCFS) algorithm. The feature selection from the vast range of the features extracted from the images is performed with the help of the optimal cut point provided by the LCS algorithm. Then, the image transaction database table is developed using the keywords of the training images and feature vectors. The transaction resembles the itemset and the association rules are generated from the transaction representation based on a priori algorithm with high conviction ratio and lift. After association rule generation, the proposed TreeHiCARe classifier model emanates in the diagnosis methodology. In TreeHICARe classifier, a new feature index is developed for the selection of a central feature for the decision tree centered on which the classification of images into normal or abnormal is performed.FindingsThe performance of the proposed method is validated over existing works using accuracy, sensitivity and specificity measures. The experimentation of proposed method on Mammographic Image Analysis Society database resulted in classification of normal and abnormal cancerous mammogram images with an accuracy of 0.8289, sensitivity of 0.9333 and specificity of 0.7273.Originality/valueThis paper proposes a new approach for the breast cancer diagnosis system by using mammogram images. The proposed method uses two new algorithms: LCFS and TreeHiCARe. LCFS is used to select optimal feature split points, and TreeHiCARe is the decision tree classifier model based on association rule agreements.

Recent earthquake events evidenced that damage of structural components in a lifeline network may cause prolonged disruption of lifeline services, which eventually results in significant socio‐economic losses in the affected area. Despite recent advances in network reliability analysis, the complexity of the problem and various uncertainties still make it a challenging task to evaluate the post‐hazard performance and connectivity of lifeline networks efficiently and accurately. In order to overcome such challenges and take advantage of merits of multi‐scale analysis, this paper develops a multi‐scale system reliability analysis method by integrating a network decomposition approach with the matrix‐based system reliability (MSR) method. In addition to facilitating system reliability analysis of large‐size networks, the multi‐scale approach enables optimizing the level of computational effort on subsystems; identifying the relative importance of components and subsystems at multiple scales; and providing a collaborative risk management framework. The MSR method is uniformly applied for system reliability analyses at both the lower‐scale (for link failure) and the higher‐scale (for system connectivity) to obtain the probability of general system events, various conditional probabilities, component importance measures, statistical correlation between subsystem failures and parameter sensitivities. The proposed multi‐scale analysis method is demonstrated by its application to a gas distribution network in Shelby County of Tennessee. A parametric study is performed to determine the number of segments during the lower‐scale MSR analysis of each pipeline based on the strength of the spatial correlation of seismic intensity. It is shown that the spatial correlation should be considered at both scales for accurate reliability evaluation. The proposed multi‐scale analysis approach provides an effective framework of risk assessment and decision support for lifeline networks under earthquake hazards. Copyright © 2009 John Wiley & Sons, Ltd.

Multi-scale stochastic dynamic response analysis of offshore risers with lognormal uncertainties

Although the existing traditional image classification methods have been widely applied in practical problems, there are some problems in the application process, such as unsatisfactory effects, low classification accuracy, and weak adaptive ability. This method separates image feature extraction and classification into two steps for classification operation. The deep learning model has a powerful learning ability, which integrates the feature extraction and classification process into a whole to complete the image classification test, which can effectively improve the image classification accuracy. However, this method has the following problems in the application process: first, it is impossible to effectively approximate the complex functions in the deep learning model. Second, the deep learning model comes with a low classifier with low accuracy. So, this paper introduces the idea of sparse representation into the architecture of the deep learning network and comprehensively utilizes the sparse representation of well multidimensional data linear decomposition ability and the deep structural advantages of multilayer nonlinear mapping to complete the complex function approximation in the deep learning model. And a sparse representation classification method based on the optimized kernel function is proposed to replace the classifier in the deep learning model, thereby improving the image classification effect. Therefore, this paper proposes an image classification algorithm based on the stacked sparse coding depth learning model-optimized kernel function nonnegative sparse representation. The experimental results show that the proposed method not only has a higher average accuracy than other mainstream methods but also can be good adapted to various image databases. Compared with other deep learning methods, it can better solve the problems of complex function approximation and poor classifier effect, thus further improving image classification accuracy.

Super-resolution guided knowledge distillation for low-resolution image classification

A multi-scale and multi-physics approach to main steam line break accidents using coupled MASTER/CUPID/MARS code

Identification and control of systems with multiple time scales is in general known to be quite challenging than that of single scale systems. The foremost and crucial step in the development of any method for data-driven multiscale analysis is to determine whether the system warrants a multiscale treatment or otherwise. Existing works in the associated literature either assume that this non-trivial knowledge is readily available or do not necessarily assess the need for a multiscale analysis. This letter is aimed at addressing this fundamental issue. The main objectives of this letter are to develop a method for: 1) detecting the presence and 2) counting the number of time scales for linear time-invariant systems solely from input-output data. The problem is formulated in the sparse optimization framework with a non-orthogonal dictionary consisting of filtered inputs resulting from finiteorder Laguerre basis filter expansions of transfer functions over a highly redundant pole grid. A clustering of the most informative poles selected by the sparse optimizer reveals the presence of multiscale nature and aids in the cardinality of time scales. The proposed method can be applied to multi-input, multi-output systems under open-loop conditions. Simulation studies demonstrate the success of the proposed methodology.

In this paper, a compact-dictionary-based sparse representation (CDSR) method is proposed for hyperspectral image (HSI) classification. The proposed dictionary in CDSR is dynamically generated according to the spatial and spectral context of each pixel. It can effectively shrink the decision range for classification, and reduce the computational burden since the compact dictionary is composed of the classes correlated with the target pixel in terms of spatial location and spectral information. In order to obtain better spatial context information, a spatial location expanding strategy is designed for spreading local explicit label information to a wider region. Experimental results demonstrate the effectiveness and superiority of the proposed method when compared with some widely used HSI classification approaches.

Community ecology and ecosystem ecology seem to have existed in different worlds. Levin (1989) suggests that the gulf between the two is the consequence of the different historical traditions in each. Community ecology, for example, emerged from basic studies, where generalized patterns were sought in the natural interactions among the biota. From the outset, the goal has been to deduce general and simple theory. On the other hand, many of the modelling approaches developed to understand ecosystem dynamics emerged from specific applied problems, where not only biotic but abiotic and human disturbances transformed ecosystem function. That tradition, therefore, is often more complete, but at the price of producing a collection of complex specific examples from which generalization is difficult.

Towards intelligent robust detection of anatomical structures in incomplete volumetric data.