Statistical shape modeling is an established technique and is used for a variety of tasks in medical image processing, such as image segmentation and analysis. A challenging task in the construction of a shape model is establishing a good correspondence across the set of training shapes. Especially for shapes of cylindrical topology, very little work has been done. This paper describes an automatic method to obtain a correspondence for a set of cylindrical shapes. The method starts from an initial correspondence which is provided by cylindrical parameterization. The quality of the obtained correspondence, measured in terms of the description length, is then improved by deforming the parameterizations using cylindrical b-spline deformations and by optimization of the spatial alignment of the shapes. In order to allow efficient gradient-guided optimization, an analytic expression is provided for the gradient of this quality measure with respect to the parameters of the parameterization deformation and the spatial alignment. A comparison is made between models obtained from the correspondences before and after the optimization. The results show that, in comparison with parameterization-based correspondences, this new method establishes correspondences that generate models with significantly increased performance in terms of reconstruction error, generalization ability, and specificity.

Novelty or anomaly detection in spatio/temporal data refers to the automatic identification of novel or abnormal events embedded in data that occur at a specific location/time. Traditional techniques used in process control to identify novelties are not robust for noise in the data set. We present an algorithm based on the support vector machine approach for domain description. This technique is intrinsicly robust for outliers in the data set but to make it work, several extensions are needed which form the contribution of this work: an extended representation of the spatio/temporal data, a tensor product kernel to separately deal with the distinct features of time and measurements, and a voting function which identifies novelties based on different representations of the time series in a robust way. Experimental results on both artificial and real data demonstrate that our algorithm performs significantly better than other standard techniques used in process control.

To obtain accurate modeling results, it is of primal importance to find optimal values for the hyperparameters in the Support Vector Regression (SVR) model. In general, we search for those parameters that minimize an estimate of the generalization error. In this study, we empirically investigate different performance measures found in the literature: k-fold cross-validation, the computationally intensive, but almost unbiased leave-one-out error, its upper bounds -radius/margin and span bound -, Vapnik's measure, which uses an estimate of the VC dimension, and the regularized risk functional itself. For each of the estimates we focus on accuracy, complexity and the presence of local minima. The latter significantly influences the applicability of gradient-based search techniques to determine the optimal parameters.

This paper introduces a precision- and range-independent tool for testing the compliance of hardware or software implementations of (multiprecision) floating-point arithmetic with the principles of the IEEE standards 754 and 854. The tool consists of a driver program, offering many options to test only specific aspects of the IEEE standards, and a large set of test vectors, encoded in a precision-independent syntax to allow the testing of basic and extended hardware formats as well as multiprecision floating-point implementations. The suite of test vectors stems on one hand from the integration and fully precision- and range-independent generalization of existing hardware test sets, and on the other hand from the systematic testing of exact rounding for all combinations of round and sticky bits that can occur. The former constitutes only 50% of the resulting test set. In the latter we especially focus on hard-to-round cases. In addition, the test suite implicitly tests properties of floating-point operations, following the idea of Paranoia, and it reports which of the three IEEE-compliant underflow mechanisms is used by the floating-point implementation under consideration. We also chech whether that underflow mechanism is used consistently. The tool is backward compatible with the UCBTEST package and with Coonen's test syntax.

The IEEE 754 and 854 standards for floating-point arithmetic are essentially a specification of a programming environment, encompassing aspects from computer hardware, operating systems, and compilers to programming languages (see especially Section 8). Parts I and II of this paper together describe a tool to test floating-point implementations of arbitrary precision and exponent range (hardware as well as software) for compliance with the principles outlined in the IEEE standards. The tool consists of a driver program, together with a very large set of test vectors encoded in a precision-independent syntax. In Part I we have convered the testing of the basic operaitons +, −, ×, / and of the square root and remainder functions. In Part II we describe the extension of the test tool to deal with conversions between floating-point formats, conversions between floating-point and integer formats, the rounding of floating-point numbers to integral values, and binary-decimal conversions. Conversions can now be tested from a floating-point format of arbitrary precision and exponent range to another arbitrary smaller (larger) floating-point format as well as to and from fixed hardware integer formats. Conversion between the bases 2 and 10 can be tested for a number of precisions ranging from sing (24 bits), double (53 bits), long double or extended (64 bits) to quadruple (113 bits) precisioon and a proper multiprecision (240 bits) format. We conclude Part II with some applications of our test tool and report on the results of testing various floating-point implementations, meaning various language-compiler-hardware combinations as well as multiprecision libraries.