Content-based image retrieval using color vector angle difference histogram

Abstract

This paper presents an innovative method to represent an image for image retrieval in Content-Based Image Retrieval (CBIR), using a color vector angle differences histogram approach (CVADH), which is different from the existing histograms, such as color histograms, gray level co-occurrences (GLCM), color distribution entropy (CDE), or just counting the frequency number of pixels or neighbors. In contrast to other methods, the CVADH takes account of the color difference and spatial information of two pixels in RGB color space. Furthermore, it directly leverages color clues to extract shape features to show differences given different backgrounds. Experimental results show that the scheme is more efficient than traditional ones that have been developed for CBIR.     

Medical image registration‐based retrieval using distance metrics

In this article, registration and retrieval are carried out separately for medical images and then registration‐based retrieval is performed. It is aimed to provide a more thorough insight on the use of registration, retrieval, and registration‐based retrieval algorithm for medical images. The purpose of this work is to deal these techniques with anatomical imaging modalities for clinical diagnosis, treatment, intervention, and surgical planning in a more effective manner. Two steps are implemented. In the first step, the affine transformation‐based registration for medical image is processed. The second step is the retrieval of medical images processed by using seven distance metrics such as euclidean, manhattan, mahalanobis, canberra, bray‐curtis, squared chord, chi‐squared, and also by using the features like mean, standard deviation, skewness, energy, and entropy. Now images registered by affine transformation are applied for retrieval. In this work, both registration and retrieval techniques in medical domain share some common image processing steps and required to be integrated in a larger system to complement each other. Experimental results, it is evident that euclidean and manhattan produces 100% precision and 35% recall found to have higher performance in retrieval. From the four anatomical modalities considered (brain, chest, liver, and limbs) brain image has better registration. It is also found that though the registration of images changes the orientation, for better performance of images in clinical evaluation it does not widely affect the retrieval performance. In the medical domain the ultimate aim of this work is to provide diagnostic support to physicians and radiologists by displaying relevant past cases, along with proven pathologies as ground truth from experimental results. © 2013 Wiley Periodicals, Inc. Int J Imaging Syst Technol, 23, 360–371, 2013     

商标检索中形状特征描述的研究

针对图像检索中的＂语义鸿沟＂问题,以商标图像为研究对象。分析了拉丁字母的物象、抽象和意象等含义,构建了一套＂字母描述子＂语言。同时还分析了商标的起源、构成和特征,并提出了用＂字母描述子＂对商标图像进行特征描述和编码的方法。经系列的检索结果和统计数据说明,用＂字母描述子＂对商标特征进行描述、编码和检索的方法具有一定的实用意义。     

Content‐based gastric image retrieval using convolutional neural networks

The endoscopy procedure has demonstrated great efficiency in detecting stomach lesions, with extensive numbers of endoscope images produced globally each day. The content‐based gastric image retrieval (CBGIR) system has demonstrated substantial potential in gastric image analysis. Gastric precancerous diseases (GPD) have higher prevalence in gastric cancer patients. Thus, effective intervention is crucial at the GPD stage. In this paper, a CBGIR method is proposed using a modified ResNet‐18 to generate binary hash codes for a rapid and accurate image retrieval process. We tested several popular models (AlexNet, VGGNet and ResNet), with ResNet‐18 determined as the optimum option. Our proposed method was valued using a GPD data set, resulting in a classification accuracy of 96.21 ± 0.66% and a mean average precision of 0.927 ± 0.006 , outperforming other state‐of‐art conventional methods. Furthermore, we constructed a Gastric‐Map (GM) based on feature representations in order to visualize the retrieval results. This work has great auxiliary significance for endoscopists in terms of understanding the typical GPD characteristics and improving aided diagnosis.     

Subband decomposition strategies for object‐based image coding

In order to attain perfect reconstruction, subband decomposition of arbitrarily shaped objects requires special strategies to deal with the decimation problem. One of the two strategies described in this paper (scheme 1) inserts pixels with amplitudes equal to the mean values of the neighboring ones belonging to the object. The requirement of transmitting (coding) these pixels reduces the coding gain and is avoided by the second strategy (scheme 2) examined in this paper. This strategy determines specific values of the pixels added to the object in such a way that they result in predefined amplitudes after decimation. For this reason, this method outperforms scheme 1. PSNR versus bit rate are presented for two objects. © 2003 Wiley Periodicals, Inc. Int J Imaging Syst Technol 13, 179–187, 2003; Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/ima.10054     

Feature extraction through generalization of histogram refinement technique for local region‐based object attributes

Content based image retrieval (CBIR) is used to retrieve digital images from large databases. However, the problem of retrieving images on the basis of the contents remains largely unsolved. The proposed method of image retrieval is based on the information provided by histogram analysis of the intensity or grayscale values of images. Some additional properties are also calculated and used that are based on regional characteristics of various objects in the image. The need to retrieve the additional regional properties arises due to the fact that the standard histograms are insensitive to small changes in images. Many images of different types can have similar histograms, because, histograms provide only a coarse characterization of an image. This is the main disadvantage of using histograms. This research is based on the concept of Histogram Refinement (Pass and Zabih, IEEE Workshop Appl Comput Vision ( 1996 ), 96–102). Distributing the grayscale image intensities by splitting the pixels using their intensity values into several classes just like the histogram refinement method can provide an estimate of the object characteristics present in an image. After the calculation of clusters using a color refinement method, the inherent features of each of the clusters is calculated based on the regional properties of the clusters. These additional region based features expound some structural information of the image. Finally, all of these features are used for image retrieval. © 2011 Wiley Periodicals, Inc. Int J Imaging Syst Technol, 21, 298–306, 2011;     

A technique of object‐adaptive vertex‐based shape coding for MPEG‐4 videos

The MPEG‐4 visual standard is the first international standard that allows the transmission of arbitrarily shaped video objects and provides technologies to view, access, and manipulate objects rather than pixels. It addresses the encoding of video objects by shape coding, motion estimation, and texture coding for interactivity, high compression, and scalability. Current binary shape‐coding techniques can be classified into two categories: bitmap based and contour based. O'Connell (1997) proposed an object‐adaptive vertex‐based shape‐coding method to improve the efficiency of shape coding. This method encodes the relative locations of a video object's vertices by adapting the representation to the dynamic range of the relative locations and by exploiting an octant‐based representation for each relative location. We propose an extension of O'Connell's method. Two relative locations of a video object's vertices are grouped and the x pairs and y pairs of the locations are encoded, respectively. Simulation results demonstrate that our method outperforms O'Connell's method. © 2001 John Wiley & Sons, Inc. Int J Imaging Syst Technol, 11, 277–282, 2000     

An assessment of the kernel‐distance‐based multivariate control chart through an industrial application

Traditional multivariate quality control charts assume that quality characteristics follow a multivariate normal distribution. However, in many industrial applications the process distribution is not known, implying the need to construct a flexible control chart appropriate for real applications. A promising approach is to use support vector machines in statistical process control. This paper focuses on the application of the ‘kernel‐distance‐based multivariate control chart’, also known as the ‘k‐chart’, to a real industrial process, and its assessment by comparing it to Hotelling's T² control chart, based on the number of out‐of‐control observations and on the Average Run Length. The industrial application showed that the k‐chart is sensitive to small shifts in mean vector and outperforms the T² control chart in terms of Average Run Length. Copyright © 2010 John Wiley & Sons, Ltd.     

Knowledge‐based algorithms in fixed‐grid GA shape optimization

Shape optimization through a genetic algorithm (GA) using discrete boundary steps and the fixed‐grid (FG) finite‐element analysis (FEA) concept was recently introduced by the authors. In this paper, algorithms based on knowledge specific to the FG method with the GA‐based shape optimization (FGGA) method are introduced that greatly increase its computational efficiency. These knowledge‐based algorithms exploit the information inherent in the system at any given instance in the evolution such as string structure and fitness gradient to self‐adapt the string length, population size and step magnitude. Other non‐adaptive algorithms such as string grouping and deterministic local searches are also introduced to reduce the number of FEA calls. These algorithms were applied to two examples and their effects quantified. The examples show that these algorithms are highly effective in reducing the number of FEA calls required hence significantly improving the computational efficiency of the FGGA shape optimization method. Copyright © 2003 John Wiley & Sons, Ltd.     

Brain tumor severity analysis using modified multi‐texton histogram and hybrid kernel SVM

Magnetic resonance image (MRI) segmentation refers to a process of assigning labels to set of pixels or multiple regions. It plays a major role in the field of biomedical applications as it is widely used by the radiologists to segment the medical images input into meaningful regions. In recent years, various brain tumor detection techniques are presented in the literature. In this article, we have developed an approach to brain tumor detection and severity analysis is done using the various measures. The proposed approach comprises of preprocessing, segmentation, feature extraction, and classification. In preprocessing steps, we need to perform skull stripping and then, anisotropic filtering is applied to make image suitable for extracting features. In feature extraction, we have modified the multi‐texton histogram (MTH) technique to improve the feature extraction. In the classification stage, the hybrid kernel is designed and applied to training of support vector machine to perform automatic detection of tumor region in MRI images. For comparison analysis, our proposed approach is compared with the existing works using K‐cross fold validation method. From the results, we can conclude that the modified multi‐texton histogram with non‐linear kernels has shown the accuracy of 86% but the MTH with non‐linear kernels shows the accuracy of 83.8%.     

Performance comparison of data‐reordering algorithms for sparse matrix–vector multiplication in edge‐based unstructured grid computations

Several performance improvements for finite‐element edge‐based sparse matrix–vector multiplication algorithms on unstructured grids are presented and tested. Edge data structures for tetrahedral meshes and triangular interface elements are treated, focusing on nodal and edges renumbering strategies for improving processor and memory hierarchy use. Benchmark computations on Intel Itanium 2 and Pentium IV processors are performed. The results show performance improvements in CPU time ranging from 2 to 3. Copyright © 2005 John Wiley & Sons, Ltd.     

Feature selection and fault‐severity classification–based machine health assessment methodology for point machine sliding‐chair degradation

In this paper, we propose an offline and online machine health assessment (MHA) methodology composed of feature extraction and selection, segmentation‐based fault severity evaluation, and classification steps. In the offline phase, the best representative feature of degradation is selected by a new filter‐based feature selection approach. The selected feature is further segmented by utilizing the bottom‐up time series segmentation to discriminate machine health states, ie, degradation levels. Then, the health state fault severity is extracted by a proposed segment evaluation approach based on within segment rate‐of‐change (RoC) and coefficient of variation (CV) statistics. To train supervised classifiers, a priori knowledge about the availability of the labeled data set is needed. To overcome this limitation, the health state fault‐severity information is used to label (eg, healthy, minor, medium, and severe) unlabeled raw condition monitoring (CM) data. In the online phase, the fault‐severity classification is carried out by kernel‐based support vector machine (SVM) classifier. Next to SVM, the k‐nearest neighbor (KNN) is also used in comparative analysis on the fault severity classification problem. Supervised classifiers are trained in the offline phase and tested in the online phase. Unlike to traditional supervised approaches, this proposed method does not require any a priori knowledge about the availability of the labeled data set. The proposed methodology is validated on infield point machine sliding‐chair degradation data to illustrate its effectiveness and applicability. The results show that the time series segmentation‐based failure severity detection and SVM‐based classification are promising.     

Shape optimization using a NURBS‐based interface‐enriched generalized FEM

This study presents a gradient‐based shape optimization over a fixed mesh using a non‐uniform rational B‐splines‐based interface‐enriched generalized finite element method, applicable to multi‐material structures. In the proposed method, non‐uniform rational B‐splines are used to parameterize the design geometry precisely and compactly by a small number of design variables. An analytical shape sensitivity analysis is developed to compute derivatives of the objective and constraint functions with respect to the design variables. Subtle but important new terms involve the sensitivity of shape functions and their spatial derivatives. Verification and illustrative problems are solved to demonstrate the precision and capability of the method. Copyright © 2016 John Wiley & Sons, Ltd.     

Insight into the flow‐condition‐based interpolation finite element approach: solution of steady‐state advection–diffusion problems

The flow‐condition‐based interpolation (FCBI) finite element approach is studied in the solution of advection–diffusion problems. Two FCBI procedures are developed and tested with the original FCBI method: in the first scheme, a general solution of the advection–diffusion equation is embedded into the interpolation, and in the second scheme, the link‐cutting bubbles approach is used in the interpolation. In both procedures, as in the original FCBI method, no artificial parameters are included to reach stability for high Péclet number flows. The procedures have been implemented for two‐dimensional analysis and the results of some test problems are presented. These results indicate good stability and accuracy characteristics and the potential of the FCBI solution approach. Copyright © 2005 John Wiley & Sons, Ltd.