Search

Segmentation of the subcortical structures in the brain such as the hippocampus, is known to be very challenging owing to its’ image characteristics. In brain MR images, the hippocampus is observed as a gray matter structure that often exhibits very weak or unclear boundary definitions at some fragments of its’ boundary. The unclear boundaries even cause the medical experts to misjudge the hippocampus boundary, especially at the head and tail. In this research, an automated segmentation approach, termed as Assembled and Weighted Coherent Point Drift is investigated to delineate the hippocampus accurately. Evaluations on public datasets produced an average Dice Similarity Coefficient of 0.8050, which appears better, in comparison to several other hippocampus segmentation approaches, especially against the well-known software program called Freesurfer. The study also revealed that the accuracy of the proposed segmentation approach seems on par with other various state-of-the-art approaches.

The Nurse Rostering Problem is an NP-hard combinatorial optimization, scheduling problem for assigning a set of nurses to shifts per day by considering both hard and soft constraints. A novel metaheuristic technique is required for solving Nurse Rostering Problem (NRP). This work proposes a metaheuristic technique called Directed Bee Colony Optimization Algorithm using the Modified Nelder-Mead Method for solving the NRP. To solve the NRP, the authors used a multiobjective mathematical programming model and proposed a methodology for the adaptation of a Multiobjective Directed Bee Colony Optimization (MODBCO). MODBCO is used successfully for solving the multiobjective problem of optimizing the scheduling problems. This MODBCO is an integration of deterministic local search, multiagent particle system environment, and honey bee decision-making process. The performance of the algorithm is assessed using the standard dataset INRC2010, and it reflects many real-world cases which vary in size and complexity. The experimental analysis uses statistical tools to show the uniqueness of the algorithm on assessment criteria.

In object class recognition, the state-of-the-art works shows using combination varies local features may produce a good performance in recognition. These local features may have a different performance on one category to other category which it depends on the richness of local features. Due to that limitation, the shape features of objects are taken into consideration to be combined with local features. In this paper, we use Fourier Descriptor (FD), Elliptical Fourier Descriptors (EFD) and Moment Invariant (MI) as a global shape feature and Scale Invariant Feature Transform (SIFT) as local features. For learning technique, boosting is used in improving the recognition objects. This approach identifies the correct and misclassified dataset iteratively. Experimental results indicate that the recognition model outperform improved the accuracy of classification by up to 10% that is comparable to or better than that of state-of-the-art approaches.

In this paper, we have addressed the issue of over-segmented regions produced in watershed by merging the regions using global feature. The global feature information is obtained from clustering the image in its feature space using Fuzzy C-Means (FCM) clustering. The over-segmented regions produced by performing watershed on the gradient of the image are then mapped to this global information in the feature space. Further to this, the global feature information is optimized using Simulated Annealing (SA). The optimal global feature information is used to derive the similarity criterion to merge the over-segmented watershed regions which are represented by the region adjacency graph (RAG). The proposed method has been tested on digital brain phantom simulated dataset to segment white matter (WM), gray matter (GM) and cerebrospinal fluid (CSF) soft tissues regions. The experiments showed that the proposed method performs statistically better, with average of 95.242% regions are merged, than the immersion watershed and average accuracy improvement of 8.850% in comparison with RAG-based immersion watershed merging using global and local features.

Segmentation of hippocampus in the brain is one of a major challenge in medical image segmentation due to its’ imaging characteristics, with almost similar intensity between another adjacent gray matter structure, such as amygdala. The intensity similarity has causes the hippocampus to have weak or fuzzy boundaries. With this main challenge being demonstrated by hippocampus, a segmentation method that relies on image information alone may not produce accurate segmentation results. Therefore, it is needed an assimilation of prior information such as shape and spatial information into existing segmentation method to produce the expected segmentation. Previous studies has widely integrated prior information into segmentation methods. However, the prior information has been utilized through a global manner integration, and this does not reflect the real scenario during clinical delineation. Therefore, in this paper, a locally integrated prior information into a level set model is presented. This work utilizes a mean shape model to provide automatic initialization for level set evolution, and has been integrated as prior information into the level set model. The local integration of edge based information and prior information has been implemented through an edge weighting map that decides at voxel level which information need to be observed during a level set evolution. The edge weighting map shows which corresponding voxels having sufficient edge information. Experiments shows that the proposed integration of prior information locally into a conventional edge-based level set model, known as geodesic active contour has shown improvement of 9% in averaged Dice coefficient.

Image annotation is an important task in computer vision. The annotated images can be very useful for indexing and retrieval applications. In this paper, we propose a generative model for image annotation based on mixtures of the exponential family of distributions. The distributions considered are the Multivariate Gaussian, Rayleigh, Poisson, Bernoulli and Centered Laplacian. The model leads to a generic algorithm that is able to perform learning in a more efficient and flexible manner. Subsequently, it can be used to evaluate the performance of each distribution in the task of image annotation. The approach is also compared to a discriminative approach (i.e. Support Vector Machine) using the LabelMe dataset for the concepts Buildings, Street, Mountains and Coasts, where promising classification results were reported for some of the concepts.

This paper aims to provide a comprehensive review of nature-inspired techniques used in image segmentation problems. We focus particularly on multi-objective clustering and classification approaches. The approaches are classified based on the various aspects of optimization, various possible problem formulations, and types of datasets modeled. In the multi-objective clustering methods, the definition of the types of representation methods, encoding techniques, and number of clusters defined (fixed/variable) are presented. In the use of multi-objective nature-inspired techniques in classification, we describe issues related to diversity measures, accuracy measures, rule manipulation, and managing uncertainties. Through our analysis of the current state of research, we hope to address important challenges and provide specific directions for future modeling of similar problems with multi-objective optimization techniques.

Automatic road network extraction based on high resolution satellite image for urban planning holds great potential for significant reduction of database development/updating cost and turnaround time. Satellite remote sensing has been recognized worldwide as an effective technology for the monitoring and mapping the urban development. Two approaches for road network extraction for an urban region have been proposed. When an image is considered in original form it is difficult and computationally expensive to extract roads due to presence of other road-like features with straight edges. Hence roads are first extracted as elongated regions by removing bright regions (that mostly represent the buildings, parking lots and other open spaces), non-linear noise segments are removed median filtering (based upon the fact that road networks constitute large number of small linear structures).The roads are then modeled as boundaries and are extracted using Level set and Normalized cuts methods .Finally The extracted roads are overlayed on the original image. The experimental results show that these approaches are efficient in extracting road segments in urban region from high resolution satellite images. Evaluation of the results carried out by comparing the level set and normalized cuts results with manually extracted reference data. The methods were applied on the high resolution IKONOS image of urban area of Hobart, Australia.

This paper introduces an approach to perform segmentation of regions in computed tomography (CT) images that exhibit intra-region intensity variations and at the same time have similar intensity distributions with surrounding/adjacent regions. In this work, we adapt a feature computed from wavelet transform called wavelet energy to represent the region information. The wavelet energy is embedded into a level set model to formulate the segmentation model called wavelet energy-guided level set-based active contour (WELSAC). The WELSAC model is evaluated using several synthetic and CT images focusing on tumour cases, which contain regions demonstrating the characteristics of intra-region intensity variations and having high similarity in intensity distributions with the adjacent regions. The obtained results show that the proposed WELSAC model is able to segment regions of interest in close correspondence with the manual delineation provided by the medical experts and to provide a solution for tumour detection.

Land use and land cover has a leading role in the part of urbanization. As the rapid urbanization led to various activities in a region and these changes generally takes place in the agricultural land. The satellite imageries LANDSAT TM (1990 and 2010) data are used. The scales are 1: 50, 000. From 1990 to 2010 covering a period of 20 years, the aerial distribution of the land use and land cover changes has been observed. The changes were identified, in which the decrease of Agricultural land, open Scrub and Water bodies and increase of Built-up land, Fallow land and current Fallow land. The land use and land cover maps are prepared by using GIS software to evaluate the changes and it showed strong variation. Image pre-processing and supervised technique analysis will be performed to classify the images into different land use categories. Thus, the information of urban growth, land use and land cover change are very useful to local government, urban and town planners, in order to have a planned city with reduction of congestion, pollution and to achieve a sustainable development of the city.

Problem statement: This study proposed multimodal integration method at the concept level to investigate information from multimodalities. The multimodal data was represented as two separate lists of concepts which were extracted from images and its related text. The concepts extracted from image analysis are often ambiguous, while the concepts extracted from text processing could be sense-ambiguous. The major problems that face the integration of the underlying modalities (image and text) were: The difference in the coverage and the difference in the granularity level. Approach: This study proposed a novel application using ontology alignment to unify the underlying ontologies. The said lists of concepts were represented in a structured form within the corresponding ontologies then the two structural lists are enriched and matched based on the alignment, this matching represent the final knowledge. Results: The difference in the coverage was solved in this study using the alignment process and the difference in the granularity level was solved using the enrichment process. Thus, the proposed integration produced accurate integrated results. Conclusion: Thus, integration of these concepts allows the totality of the knowledge be expressed more precisely.

Most supervised neural networks are trained by minimizing the mean square error (MSE) of the training set. In the presence of outliers, the resulting neural network model can differ significantly from the underlying model that generates the data. This paper outlines two robust learning methods for a dynamic structure neural network called incremental growing multi-experts network (IGMN). It is convincingly shown by simulation that by using a scaled robust objective function instead of the least squares function, the influence of the outliers in the training data can be completely eliminated. The network generates a much better approximation in the neighborhood of outliers. Thus, the two proposed robust learning methods namely robust least mean squares (RLMSs) and least mean log squares (LMLSs) are insensitive to the presence of outliers unlike the least mean squares (LMSs) cost function. Moreover, various types of supervised learning algorithms can easily adopt LMLS, which is a parameter-free method.

This paper presents two novel approaches to determine optimum growing multi-experts network (GMN) structure. The first method called direct method deals with expertise domain and levels in connection with local experts. The growing neural gas (GNG) algorithm is used to cluster the local experts. The concept of error distribution is used to apportion error among the local experts. After reaching the specified size of the network, redundant experts removal algorithm is invoked to prune the size of the network based on the ranking of the experts. However, GMN is not ergonomic due to too many network control parameters. Therefore, a self-regulating GMN (SGMN) algorithm is proposed. SGMN adopts self-adaptive learning rates for gradient-descent learning rules. In addition, SGMN adopts a more rigorous clustering method called fully self-organized simplified adaptive resonance theory in a modified form. Experimental results show SGMN obtains comparative or even better performance than GMN in four benchmark examples, with reduced sensitivity to learning parameters setting. Moreover, both GMN and SGMN outperform the other neural networks and statistical models. The efficacy of SGMN is further justified in three industrial applications and a control problem. It provides consistent results besides holding out a profound potential and promise for building a novel type of nonlinear model consisting of several local linear models.

This paper deals with a novel idea of identification of nonlinear dynamic systems via a constructivism inspired neural network. The proposed network is known as growing multi-experts network (GMN). In GMN, the problem space is decomposed into overlapping regions by expertise domain and local expert models are graded according to their expertise level. The network output is computed by the smooth combination of local linear models. In order to avoid over-fitting problem, GMN deploys a redundant experts removal algorithm to remove the redundant local experts from the network. In addition, growing neural gas (GNG) algorithm is used to generate an induced Delaunay triangulation that is highly desired for optimal function approximation. A variety of examples are taken from literature to establish the efficacy of GMN. Discrete time nonlinear dynamic system modeling and water bath temperature control have been found to give excellent results via this novel neural network.

Artificial neural networks (ANNs) have been used to construct empirical nonlinear models of process data. Because networks are not based on physical theory and contain nonlinearities, their predictions are suspect when extrapolating beyond the range of original training data. Standard networks give no indication of possible errors due to extrapolation. This paper describes a sequential supervised learning scheme for the recently formalized Growing multi-experts network (GMN). It is shown that certainty factor can be generated by GMN that can be taken as extrapolation detector for GMN. On-line GMN identification algorithm is presented and its performance is evaluated. The capability of the GMN to extrapolate is also indicated. Four benchmark experiments are dealt with to demonstrate the effectiveness and utility of GMN as a universal function approximator.

To segment an image using the random walks algorithm; users are often required to initialize the approximate lo- cations of the objects and background in the image. Due to its segmenting model that is mainly reflected by the relationship among the neighborhood pixels and its boundary conditions, random walks algorithm has made itself sensitive to the inputs of the seeds. Instead of considering the relationship between the neighborhood pixels solely, an attempt has been made to modify the weighting function that accounts for the intensity changes between the neighborhood nodes. Local affiliation within the defined neighborhood region of the two nodes is taken into consideration by incorporating an extra penalty term into the weighting function. Besides that, to better segment images, particularly medical images with texture features, GLCM variance is incorporated into the weighting function through kernel density estimation (KDE). The probability density of each pixel belonging to the initialized seeds is estimated and integrated into the weighting function. To test the performance of the proposed weighting model, several medical images that mainly made up of 174-brain tumor images are experimented. These experiments establish that the proposed method produces better segmentation results than the original random walks.

Though the need to consider all aspects of real-time in the development of a system is not new to control engineers, they are relatively new to the machine vision specialists, especially as they tend to come from various different backgrounds. In practice most vision experts tend to “tweak” their systems to match the speed of the application environment. They rarely use any formal or indeed informal methods to understand, design and analyse the real-time behavior of their systems. This paper, however, suggests that this trend needs to be reversed, and that correct timing of an operation is as important as correct logical operation. The paper starts by justifying the need to analyse the dynamic behavior of machine-vision systems. The paper then briefly reviews some of the basic concepts of real-time systems, and explores a tool available for the specification and analysis of real-time systems. Finally, the paper discusses the analysis of the real-time behavior of a machine vision system developed for the inspection of the bonding process occurring in IC fabrication.

This paper presents a new approach for hippocampus localization using pairwise non-rigid Coherent Point Drift registration method. The concept of assembled point set is introduced, which is a combination of the available training point sets into a single data space that represents its distribution. Non-rigid Coherent Point Drift is then adapted to register the assembled point set with a randomly chosen base model for hippocampus localization. The primary focus of this work is on the computational intensiveness of the localization approach, in which the proposed localization approach using assembled point set is compared with an existing groupwise non-rigid Coherent Point Drift (GCPD) approach. The computation intensiveness of the proposed approach grows at a quadratic rate as compared with GCPD that grows at a cubic rate. The proposed approach is validated with hippocampus localization task using 40-datasets. The Root Mean Square (RMS) distance between the approximated hippocampus locations and the ground truth is within an acceptable average of 0.6957-mm.

Automating the detection of lesions in liver CT scans requires a high performance and robust solution. With CT-scan start to become the norm in emergency department, the need for a fast and efficient liver lesions detection method is arising. In this paper, we propose a fast and evolvable method to profile the features of pre-segmented healthy liver and use it to detect the presence of liver lesions in emergency scenario. Our preliminary experiment with the MICCAI 2007 grand challenge datasets shows promising results of a fast training time, ability to evolve the produced healthy liver profiles, and accurate detection of the liver lesions. Lastly, the future work directions are also presented.

This paper investigates the effects of feature selection via dimensionality reduction techniques for the task of object class recognition. Two filter-based algorithms are considered namely Correlation-based Feature Selection (CFS) and Principal Components Analysis (PCA). A Support Vector Machine is used to compare these two techniques against classical feature concatenation, based on the Graz02 dataset. Experimental results show that the feature selection algorithms are able to retain the most relevant and discriminant features, while maintaining recognition accuracy and improving model building time.

Analysis of high resolution satellite images has been an important research topic for urban analysis. One of the important features of urban areas in urban analysis is the automatic road network extraction. Two approaches for road extraction based on Level Set and Mean Shift methods are proposed. From an original image it is difficult and computationally expensive to extract roads due to presences of other road-like features with straight edges. The image is preprocessed to improve the tolerance by reducing the noise (the buildings, parking lots, vegetation regions and other open spaces) and roads are first extracted as elongated regions, nonlinear noise segments are removed using a median filter (based on the fact that road networks constitute large number of small linear structures). Then road extraction is performed using Level Set and Mean Shift method. Finally the accuracy for the road extracted images is evaluated based on quality measures. The lm resolution IKONOS data has been used for the experiment.

With the boom of web and social networking, the amount of generated text data has increased enormously. Much of this data can be considered and modeled as a stream and the volume of such data necessitates the application of automated text classification strategies. Although streaming data classification is not new, considering text data streams for classification purposes has been extensively researched only recently. Before applying any classification method in text data streams, it is imperative that we apply them for existing well-known non-stream text data sets and evaluate their performance. One of the many characteristics of text data that has been pursued for research is its multi-labelity. A single text document may cover multiple class-labels at the same time and hence gives rise to the concept of multi-labelity. From classification perspective, an immediate drawback of such a characteristic is that traditional binary or multi-class classification techniques perform poorly on multi-label text data. In this paper, we extend our previously formulated SISC (Semi-supervised Impurity based Subspace Clustering) [1] approach and its multi-label variation SISC-ML [2]. We call this new algorithm H-SISC (Hierarchical SISC). H-SISC captures the underlying correlation that exists between each pair of class labels in a multi-label environment. Developing a robust multi-label classifier will allow us to apply such a model in classifying streaming text data more effectively. We have experimented with well known text data sets and empirical evaluation on these real world multi-label NASA ASRS (Aviation Safety Reporting System), Reuters and 20 Newsgroups data sets reveals that our proposed approach outperforms other state-of-the-art text classification as well as subspace clustering algorithms.

There is a lot of commercial software for working with images, from very simple editing and enhancement up to very complicated and sophisticated jobs, but we cannot find a share area that is accessible from all around the world and can respond to our needs in image processing and data visualizing. For working with the images in a distributed manner, we need a middleware which supports our system in a transparent way. For addressing this issue, we develop a framework, which uses web services for image processing and data visualization. This is the place that a general framework is being needed to respond our requirements. This paper presents different ways for developing a framework and shows that web services is one of the best solutions and gives the ability for connecting interior parts, saving and retrieving selections and annotations by using XML.

Context is a vital element in both biological as well as synthetic vision systems. It is essential for deriving meaningful explanation of an image. Unfortunately, there is a lack of consensus in the computer vision community on what context is and how it should be represented. In this paper context is defined generally as "any and all information that is not directly derived from the object of interest but helps in explaining it". Furthermore, a description of context is provided in terms of its three major aspects namely scope, source and type. As an application of context in improving object detection results a Context Verification System (ConVeS) is proposed. ConVeS incorporates semantic and spatial context with an external knowledgebase to verify object detection results provided by state-of-the-art machine learning algorithms such as support vector machine or artificial neural network. ConVeS is presented as a simple framework that can be effectively applied to a wide range of computer vision applications such as medical image, surveillance video, and natural imagery.

Image processing plays an important role in computer science, making complex image manipulation more feasible. It can be used either in a general manner or in specific domains such as for medical purposes. With the availability of the internet, image processing applications can be distributed to be available for different people, regardless of their geographical location. Although there are a handful of professional image processing solutions available, there is no well-designed framework that can be simply distributed and customized to be used in specific areas. Other than image processing tools, image storage and retrieval is another issue that needs to be addressed. This research proposes a general platform that can be used for implementing a distributed image processing framework. It provides solutions for storing and retrieving images, particularly large-size ones as well as an approach for executing image processing functions that are physically located anywhere within the distributed system.

In this paper, we propose a technique for classifying shots of playfield-based sports video into their respective view classes. Based on common broadcasting style, a shot can be classified as a far-view or a closeup-view. The technique considers the frame-wise color values of each pixel in the HSV color space, while at the same time calculating the assumed object size within the segmented playfield region. Based on our experiments, it is shown that this technique can greatly reduce the number of misclassified shots, while at the same time maintain a good level of accuracy. At the moment, we have tested our approach on soccer videos but believe that it can be applied to other playfield-based sports as well.

One major challenge faced by segmentation techniques in analyzing and visualizing individual slices of a 3D anatomical structure, is the degree of manual interaction required. To alleviate this problem, researchers have proposed the automatic incorporation of anatomical knowledge, via medical atlases to assist with the segmentation process. Some solutions include constructing specialized simple 2D, as well as complex 3D atlases. In this paper we propose a simple method that automatically transfers a prior anatomical knowledge from a simple teaching atlas of a single 2D slice to the most similar slice in the 3D volume dataset. Segmentation of the selected anatomy is then able to be propagated automatically on the remaining slices in the dataset without further manual interaction. Our experiments are conducted on abdomen CT images and we successful delineation and visualization have been demonstrated for spleen, and kidney images.

This paper presents a new segmentation method that integrates a wavelet based feature, which is able to enhance the dissimilarity between regions with low variations in intensity. This feature is integrated to formulate a new level set based active contour model that addresses the segmentation of regions with highly similar intensities in medical images, which do not have clear boundaries between them. In the first phase of this research, the strength of wavelet transform is adapted to formulate a statistical feature, named as wavelet energy. The second phase of this work is dedicated to formulate a new level set based active contour model that is suitable for segmenting regions without clear boundaries and exhibits intensity inhomogeneity within the region. The proposed segmentation method is named wavelet energy guided level set based active contour. Experiments are conducted using medical images focusing on the neck and bladder regions to validate the proposed method. The experimental results show that the proposed method is able to segment the region of interest in close correspondence with the manual delineation by the radiologists.

Overlaid text appears frequently in broadcast sports video. They provide supplementary information regarding the happenings of a particular game. Examples include important events of interest such as bookings and substitutions in a soccer match. Furthermore, overlaid-text is displayed when a particular concept of interest is happening or has happened. This paper presents a technique to automatically detect only video frames that contain valid overlaid text. Experiments have shown reliable detection, extraction and recognition, whose results have been successfully used for domain concept understanding via matching with a soccer term database.

Object class recognition is a highly challenging area in computer vision and machine learning. In this paper, we introduce a novel approach to object class recognition using Neuro Evolution of Augmenting Topologies (NEAT) to evolve artificial neural networks (ANN) capable of taking advantage of the robust SIFT feature based descriptor histograms. We claim that NEAT can produce ANN classifier which exhibits outstanding ability of learning from only afew training examples without sacrificing accuracy. Our empirical evaluations against the performance of state of the art statistical machine learning method such as support vector machine show that NEAT-evolved ANN classifier outperforms by an average of 9.96% higher accuracy when presented with very small training set proving its superior ability to generalize its learning.

Overlaid-text appears frequently in broadcast sports video. They provide a plethora of information regarding the goings-on of a particular game. Examples include important events and video segments of interest such as bookings and half-time analysis, respectively. Furthermore, it is common that overlaid text is displayed when a particular concept is happening or has happened. This paper presents a concept identification framework, based on matched keywords from overlaid-text extraction and recognition. Possible occurrences of overlaid-text in soccer programs are extracted and recognized, and then matched against a soccer-term database. Preliminary experiments show reliable character extraction, whose recognition has been successfully matched with keywords within the database.

This paper introduces Textured Renyi Entropy for image thresholding based on a novel combination mechanism. The Renyi Entropy is extended by modifying its priori, while still preserving overall functionality. An optional priori is introduced to improve accuracy. The priori modification allows adding of texture information in an efficient way, which results in more accurate thresholding. Furthermore, the proposed mechanism allows adding other features after some normalization. Finally the novel way allows adding multiple features simultaneously using the modified priors.

This paper presents a hybrid algorithm for object based clustering. The algorithm is designed based on hybrid of hierarchical and k-means clustering algorithm. For this work, we used dataset consist of natural imagery collected from PASCAL database 2006 collection and Google images. A collection of low level features image is used to validate the performance of our approach. Experimental results show that hybrid algorithm produced higher accuracy compared to k-means and hierarchical algorithm by up to 25%.

Technologies such as data warehousing, data mining, and campaign management software have made Customer Relationship Management (CRM) a new area where firms can gain a competitive advantage. It is becoming common knowledge in business that retaining existing customers is an important strategy to survive in industry. Once identified, these customers can be targeted with proactive retention campaigns in a bid to retain them. These proactive marketing campaigns usually involve the offering of incentives to attract the customer into carrying on their service with the supplier. These incentives can be costly, so offering them to customers who have no intention to defect results in lost revenue. Also many predictive techniques do not provide significant time to make customer contact. This time restriction does not allow sufficient time for capturing those customers who are intending to leave. This research aims to develop methodologies for predicting customer churn in advance, while keeping misclassification rates to a minimum.

In this paper, a novel approach to automatically segment the optic disc contour using the center point of an optic disc candidate is proposed The optic disc segmentation algorithm consists of 2 stages. The first stage involves the removal of blood vessels that obscure the optic disc. The blood vessel structures are detected using morphological operations. These detected structures are then removed by anisotropic diffusion smoothing. The second stage involves the detection of edge points belonging to the optic disc contour. A number of one dimensional intensity profiles which pass through the center point of optic disc region are then obtained at multiple angles with fixed angular intervals. The modulus maxima of each intensity profile are identified as a contour point for the optic disc. Among these contour points, some of the outliers are removed by re-positioning to a new position which complies with optic disc's shape using spline interpolation. Using these contour points, a coarse contour of the optic disc is constructed. Testing the approach on 23 colour fundus images demonstrates that the proposed algorithm is able to detect the optic disc contour to an accuracy of 92% to that drawn by a human expert.

Image segmentation is the most essential and crucial process in order to facilitate the delineation, characterization and visualization of structures of interests in medical images. In recent years, various methods of medical image segmentation have been employed depending on the type of tissues, anatomy of object of interest and imaging modality being used. In this paper, a novel problem specific segmentation method is proposed to detect secondary structures in medical images incorporating wavelet transform and watershed transformation. The proposed method consists of 5 stages: wavelet transformation, inter scale linking, feature calculation and clustering, morphological processing and watershed transformation. This paper presents the proposed method and preliminary results tested using Computed Tomography (CT) images of human anatomy. The preliminary experimental results of this research have shown encouraging results with elimination of over segmentation problem frequently faced in watershed transformation.

High quality epitaxial YBa2Cu3O7−x (YBCO) thin films have been deposited onto ferrimagnetic yttrium iron garnet (100) substrates using a double buffer layer lattice engineering scheme. The YBCO films are c‐axis oriented and show superconductive transition temperatures of 88–89 K with transition widths ≤0.5 K. The films also exhibited Jc’s of 2.7×106 A/cm2 in zero field and 1.1 ×106 A/cm2 in 2000 G at 77 K. Dielectric resonator measurements indicated unloaded Q values of 29 000 at 77 K and 24 GHz for unpatterned films. These values correspond to a surface resistance (RS) of 500 μΩ at 77 K, 10 GHz, which is the lowest obtained to date for a YBCO film on a ferrimagnetic substrate. Such films could be used for fabricating miniature‐high‐temperature superconducting microwave elements such as nonreciprocal and passive microwave devices, making possible, their integration on a single ferrimagnetic wafer.

High quality YBa2Cu3O7−δ (YBCO) films have been epitaxially grown on low dielectric constant, low loss tangent (001)MgF2 substrates using a double buffer layer scheme. The YBCO films are (001) oriented, exhibit Tc’s of 88–89 K with transition widths ≤0.5 K and Jc’s of 4×106 A/cm2 at 77 K in zero field. Dielectric resonator experiments indicate unloaded Q values in excess of 30 000 at 77 K and 24 GHz in the unpatterned films. Films patterned into meanderline resonators and operating at 10 GHz, 79 K, exhibit surface resistance (Rs)

Pose characterization refers to the unique representation of an object's pose through a set of image features. In this regard, one of the prerequisite for using visual images in representing an object's pose is a set of view-dependent image features that is unique to a single position and orientation of the object relative to the camera. Here, we propose an enhancement to the global feature extraction method. It utilizes an active illumination that projects a set of orthogonal grid patterns on the object. The deformations of the grid pattern incident on the complex target object carried important information about the surface structure of the target. This grid-encoded image can be characterized using global descriptors. Results indicate that this illumination increases the sensitivity of the global features to changes in object pose.© (1999) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.

We have performed depth profile analyses of the lattice parameters in epitaxial thin films of La1−xCaxMno3 (LCMO), where x = 0.33 or 0.3, to understand the evolution of strain relaxation processes in these materials. The analyses were done using Grazing Incidence X-ray Scattering (GIXS) on films of different thicnesses on two different substrates, (100) oriented LaAlO3 (LAO), with a lattice mismatch of ∼2% and (110) oriented NGO, with a lattice mismatch of less than 0.1%. Films grown on LAO can exhibit up to three in-plane strained lattice constants, corresponding to a slight orthorhombic distortion of the crystal, as well as near-surface and columnar lattice relaxation. As a function of film thickness, a crossover from a strained film to a mixture of strained and relaxed regions in the film occurs in the range of 700 Å. The structural evolution at this thickness coincides with a change in the resistivity curve near the metalinsulator transition. The in-plane compressive strain has a range of 0.2 – 1.5%, depending on the film thickness for filsm in the range of 400 - 1500 A.

This study reviews the state-of-the-art multiobjective optimisation (MOO) techniques with metaheuristic through clustering approaches developed specifically for image segmentation problems. The authors treat image segmentation as a real-life problem with multiple objectives; thus, focusing on MOO methods that allow a trade-off among multiple objectives. A reasonable solution to a multiobjective (MO) problem is to investigate a set of solutions, each of which satisfies the objectives at an acceptable level without being dominated by any other solution. The primary difference of MOO methods from traditional image segmentation is that instead of a single solution, their output is a set of solutions called Pareto-optimal solution. This study discusses the evolutionary and non-evolutionary MO clustering techniques for image segmentation. It diagnoses the requirements and issues for modelling MOO via MO clustering technique. In addition, the potential challenges and the directions for future research are presented.

Third Internal Visual Inspection in the IC assembly process deals with the wire bond quality inspection and is a very labor intensive stage. This paper describes a simple but comprehensive prototype model developed to automate the inspection of wedge and ball bonds. It is assumed that the die tilt is within specifications and the bond pad is automatically aligned. The defects associated with the bond quality are classified into four categories: size, shape, position and dimension. The bond is inspected sequentially for each category of defects and is rejected without further processing when any defect is detected. The procedure adopted in the prototype is as follows: A global thresholding technique automatically binarizes the intensity image of the bond. Bond limits are determined by a scanning procedure. Simple techniques such as pixel count, MER, Centroid and Median are used to verify the specifications related to size and shape. An intelligent scanning technique inspects the position related specifications in addition to identifying the wire and tail positions of the bond. The dimensions of the bond are determined by using projection information. 100% success is achieved with the experiments conducted using 100 sample images.© (1993) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.

We report charge-ordered Bi0.4Ca0.6MnO3 thin films with charge-ordering temperature near room temperature, and observation of large photoinduced resistivity changes in these films associated with melting of the charge ordering by visible light. Films grown under small compressive strain exhibit the largest photoinduced resistivity changes. The lifetime of the photoinduced low-resistance state is on the order of half a minute. These photoinduced resistivity changes in thin films of Bi0.4Ca0.6MnO3 make them very promising for photonic device application.

A simple but comprehensive prototype model developed to automate the inspection of wedge bonds in the IC assembly process is described. The defects associated with the bond quality are classified into four categories: size, shape, position, and dimension. The bond is inspected sequentially for each categoiy of defects and is rejected without further processing when any defect is detected. The procedure adopted in the prototype is as follows. A global thresholding technique automatically binarizes the intensity image of the bond. Simple features such as the pixel count, minimum enclosing rectangle, centroid, and median are used to verify the specifications related to the size and shape. An intelligent scanning technique inspects the position-related specifications in addition to identifying the wire and tall positions of the bond. The dimensions of the bond are determined using projection information. Most importantly, this model is capable of determining the wire position, which is useful in inspecting wire-related defects. Experiments conducted on actual sample ICs have shown a 100% success rate.

The shortest/optimal path generation is essential for the efficient operation of a mobile robot. Recent advances in robotics and machine intelligence have led to the application of modern optimization method such as the genetic algorithm (GA), to solve the path-planning problem. However, the genetic algorithm path planning approach in the previous works requires a preprocessing step that captures the connectivity of the free-space in a concise representation. In this paper, GA path-planning approach is enhanced with feasible path detection mechanism based on traversability vectors method. This novel idea eliminates the need of free-space connectivity representation. The feasible path detection is performed concurrently while the GA performs the search for the shortest path. The performance of the proposed GA approach is tested on three different environments consisting of polygonal obstacles with increasing complexity. In all experiments, the GA has successfully detected the near-optimal feasible...