AGEL-SVM is an extension to a kernel Support Vector Machine (SVM) and is designed for distributed computing using Approximate Global Exhaustive Local sampling (AGEL)-SVM. The dual form of SVM is typically solved using sequential minimal optimization (SMO) which iterates very fast if the full kernel matrix can fit in a computer's memory. AGEL-SVM aims to partition the feature space into sub problems such that the kernel matrix per problem can fit in memory by approximating the data outside each partition. AGEL-SVM has similar Cohen's Kappa and accuracy metrics as the underlying SMO implementation. AGEL-SVM's training times greatly decreased when running on a 128 worker MATLAB pool on Amazon's EC2. Predictor evaluation times are also faster due to a reduction in support vectors per partition.

Moisture transport properties of fabrics are of key interest within the textile industry. This paper introduces a new method to measure horizontal wicking and drying properties of a fabric using a thermal imaging machine vision system. By tracking the area of the wet region of a fabric specimen over time, five new features are measured that characterize key periods of the wicking and drying process. Experimental results show significant differences in these measurements for fabrics that have been treated with a proprietary formula known to improve wicking properties of fabrics. Furthermore, analysis shows that the repeatability of this new method shows marked improvement over current industry standards.

Due to overwhelming use of 3D models in video games and virtual environments, there is a growing interest in 3D scene generation, scene understanding and 3D model retrieval. In this paper, we introduce a data-driven 3D scene generation approach from a Maximum Entropy (MaxEnt) model selection perspective. Using this model selection criterion, new scenes can be sampled by matching a set of contextual constraints that are extracted from training and synthesized scenes. Starting from a set of random synthesized configurations of objects in 3D, the MaxEnt distribution is iteratively sampled (using Metropolis sampling) and updated until the constraints between training and synthesized scenes match, indicating the generation of plausible synthesized 3D scenes. To illustrate the proposed methodology, we use 3D training desk scenes that are all composed of seven predefined objects with different position, scale and orientation arrangements. After applying the MaxEnt framework, the synthesized scenes show that the proposed strategy can generate reasonably similar scenes to the training examples without any human supervision during sampling. We would like to mention, however, that such an approach is not limited to desk scene generation as described here and can be extended to any 3D scene generation problem.

As a result of increasing geospatial image libraries, many algorithms are being developed to automatically extract and classify regions of interest from these images. However, limited work has been done to compare, validate and verify these algorithms due to the lack of datasets with high accuracy ground truth annotations. In this paper, we present an approach to generate a large number of synthetic images accompanied by perfect ground truth annotation via learning scene statistics from few training images through Maximum Entropy (ME) modeling. The ME model [1,2] embeds a Stochastic Context Free Grammar (SCFG) to model object attribute variations with Markov Random Fields (MRF) with the final goal of modeling contextual relations between objects. Using this model, 3D scenes are generated by configuring a 3D object model to obey the learned scene statistics. Finally, these plausible 3D scenes are captured by ray tracing software to produce synthetic images with the corresponding ground truth annotations that are useful for evaluating the performance of a variety of image analysis algorithms.

Effective image segmentation of a digitized scene into a set of recognizable objects requires the development of sophisticated scene analysis algorithms. Progress in this area has been made through the development of a statistical-based deformable model that improves upon existing point distribution models (PDMs) for boundary-based object segmentation. Existing PDM boundary finding techniques often suffer from the shortcoming that global shape and gray-level information are treated independently during boundary optimization. A deformable model algorithm is under development in which the objective function used during optimization of the boundary encompasses several important characteristics. Most importantly the objective function includes both shape and gray-level characteristics, so optimization occurs with respect to both pieces of information simultaneously. This algorithm has been applied to geometric test images and a simple industrial-type scene for which results are presented.

This paper describes the application of a new probabilistic shape and appearance model (PSAM) algorithm to the task of detecting polycystic kidney disease (PKD) in X-ray computed tomography images of laboratory mice. The genetically engineered PKD mouse is a valuable animal model that can be used to develop new treatments for kidney-related problems in humans. PSAM is a statistical-based deformable model that improves upon existing point distribution models for boundary-based object segmentation. This new deformable model algorithm finds the optimal boundary position using an objective function that has several unique characteristics. Most importantly, the objective function includes both global shape and local gray-level characteristics, so optimization occurs with respect to both pieces of information simultaneously. PSAM is employed to segment the mouse kidneys and then texture measurements are applied within kidney boundaries to detect PKD. The challenges associated with the segmentation non-rigid organs along with the availability of a priori information led to the choice of a trainable, deformable model for this application. In 103 kidney images that were analyzed as part of a preclinical animal study, the mouse kidneys and spine were segmented with an average error of 2.4 pixels per boundary point. In all 103 cases, the kidneys were successfully segmented at a level where PKD could be detected using mean-of-local-variance texture measurements within the located boundary.

The concept of content-based image compression (CBIC) has far-reaching effects in the areas of archiving and telecommunications. This paper presents some pilot-study results from the application of CBIC to mammography. Unlike traditional compression approaches, CBIC first analyzes the content of the data before the compression takes place. In this approach, prior to compression, the data is pre-processed and is segmented into two non-overlapping regions: (1) focus-of-attention regions (FARs) that contain the "important" segments of the data, and (2) background regions. Subsequently, the former regions are compressed using a lossless compression technique (maintaining fidelity), while the latter regions are compressed with the aid of a lossy technique (attaining large reductions in data). The intended result is an optimal balance between data reduction and data fidelity. In this case, compression ratios 5-6 times greater than that of lossless compression alone can be reached while preserving the important information.

This paper describes a robust segmentation algorithm for the detection and localization of woven fabric defects. The essence of the presented segmentation algorithm is the localization of those events (i.e., defects) in the input images that disrupt the global homogeneity of the background texture. To this end, preprocessing modules, based on the wavelet transform and edge fusion, are employed with the objective of attenuating the background texture and accentuating the defects. Then, texture features are utilized to measure the global homogeneity of the output images. If these images are deemed to be globally nonhomogeneous (i.e., defects are present), a local roughness measure is used to localize the defects. The utility of this algorithm can be extended beyond the specific application in our work, that is, defect segmentation in woven fabrics. Indeed, in a general sense, this algorithm can be used to detect and to localize anomalies that reside in images characterized by ordered texture. The efficacy of this algorithm has been tested thoroughly under realistic conditions and as a part of an on-line fabric inspection system. Using over 3700 images of fabrics, containing 26 different types of defects, the overall detection rate of our approach was 89% with a localization accuracy of less than 0.2 inches and a false alarm rate of 2.5%.

This paper describes designing and testing systems for inspection of web materials using a development environment simulated in software. This environment comprises five sub-systems: sensing, detection, characterization, validation and decision making. Given a specific inspection problem the system is customized by incorporating specific algorithms into each of the sub-systems. In order to illustrate the customization, the paper uses an example of inspection of a nonwoven material. The inspection objective is to quantify the randomness and homogeneity across the material. This task requires using statistical modeling to characterize texture. The paper discusses a theoretical approach based on a Poissonian model, its efficient implementation, and steps required to implement this procedure online.

The article is a work-in-progress report of a research endeavor that deals with the design and development of a novel approach to computer aided diagnosis (CAD) of mammographic images. With the initial emphasis being on the analysis of microcalcifications, the proposed approach defines a synergistic paradigm that utilizes new methodologies together with previously developed techniques. The new paradigm is intended to promote a higher degree of accuracy in CAD of mammograms with an increased overall throughput. The process of accomplishing these goals is initiated by the fractal encoding of the input image, which gives rise to the generation of focus-of-attention regions (FARs), that is, regions that contain anomalies. The primary thrust of this work is to demonstrate that by considering FARs, rather than the entire input image, the performances of the ensuing processes (i.e., segmentation, feature extraction, and classification) are enhanced in terms of accuracy and speed. An experimental study is included that demonstrates the impact of FAR generation on the process of microcalcification segmentation.