Backward elimination model construction for regression and classification using leave-one-out criteria

A fundamental principle in practical nonlinear data modeling is the parsimonious principle of constructing the minimal model that explains the training data well. Leave-one-out (LOO) cross validation is often used to estimate generalization errors by choosing amongst different network architectures (M. Stone, "Cross validatory choice and assessment of statistical predictions", J. R. Stast. Soc., Ser. B, 36, pp. 117-147, 1974). Based upon the minimization of LOO criteria of either the mean squares of LOO errors or the LOO misclassification rate respectively, we present two backward elimination algorithms as model post-processing procedures for regression and classification problems. The proposed backward elimination procedures exploit an orthogonalization procedure to enable the orthogonality between the subspace as spanned by the pruned model and the deleted regressor. Subsequently, it is shown that the LOO criteria used in both algorithms can be calculated via some analytic recursive formula, as derived in this contribution, without actually splitting the estimation data set so as to reduce computational expense. Compared to most other model construction methods, the proposed algorithms are advantageous in several aspects; (i) There are no tuning parameters to be optimized through an extra validation data set; (ii) The procedure is fully automatic without an additional stopping criteria; and (iii) The model structure selection is directly based on model generalization performance. The illustrative examples on regression and classification are used to demonstrate that the proposed algorithms are viable post-processing methods to prune a model to gain extra sparsity and improved generalization.

Controls of carbon dioxide concentrations and fluxes above central London

Eddy-covariance measurements of carbon dioxide fluxes were taken semi-continuously between October 2006 and May 2008 at 190 m height in central London (UK) to quantify emissions and study their controls. Inner London, with a population of 8.2 million (~5000 inhabitants per km2) is heavily built up with 8% vegetation cover within the central boroughs. CO2 emissions were found to be mainly controlled by fossil fuel combustion (e.g. traffic, commercial and domestic heating). The measurement period allowed investigation of both diurnal patterns and seasonal trends. Diurnal averages of CO2 fluxes were found to be highly correlated to traffic. However changes in heating-related natural gas consumption and, to a lesser extent, photosynthetic activity that controlled the seasonal variability. Despite measurements being taken at ca. 22 times the mean building height, coupling with street level was adequate, especially during daytime. Night-time saw a higher occurrence of stable or neutral stratification, especially in autumn and winter, which resulted in data loss in post-processing. No significant difference was found between the annual estimate of net exchange of CO2 for the expected measurement footprint and the values derived from the National Atmospheric Emissions Inventory (NAEI), with daytime fluxes differing by only 3%. This agreement with NAEI data also supported the use of the simple flux footprint model which was applied to the London site; this also suggests that individual roughness elements did not significantly affect the measurements due to the large ratio of measurement height to mean building height.

Measuring the impact of observations on the predictability of the Kuroshio Extension in a shallow-water model

In this paper sequential importance sampling is used to assess the impact of observations on a ensemble prediction for the decadal path transitions of the Kuroshio Extension (KE). This particle filtering approach gives access to the probability density of the state vector, which allows us to determine the predictive power — an entropy based measure — of the ensemble prediction. The proposed set-up makes use of an ensemble that, at each time, samples the climatological probability distribution. Then, in a post-processing step, the impact of different sets of observations is measured by the increase in predictive power of the ensemble over the climatological signal during one-year. The method is applied in an identical-twin experiment for the Kuroshio Extension using a reduced-gravity shallow water model. We investigate the impact of assimilating velocity observations from different locations during the elongated and the contracted meandering state of the KE. Optimal observations location correspond to regions with strong potential vorticity gradients. For the elongated state the optimal location is in the first meander of the KE. During the contracted state of the KE it is located south of Japan, where the Kuroshio separates from the coast.

Deriving global flood hazard maps of fluvial floods through a physical model cascade

Global flood hazard maps can be used in the assessment of flood risk in a number of different applications, including (re)insurance and large scale flood preparedness. Such global hazard maps can be generated using large scale physically based models of rainfall-runoff and river routing, when used in conjunction with a number of post-processing methods. In this study, the European Centre for Medium Range Weather Forecasts (ECMWF) land surface model is coupled to ERA-Interim reanalysis meteorological forcing data, and resultant runoff is passed to a river routing algorithm which simulates floodplains and flood flow across the global land area. The global hazard map is based on a 30 yr (1979–2010) simulation period. A Gumbel distribution is fitted to the annual maxima flows to derive a number of flood return periods. The return periods are calculated initially for a 25×25 km grid, which is then reprojected onto a 1×1 km grid to derive maps of higher resolution and estimate flooded fractional area for the individual 25×25 km cells. Several global and regional maps of flood return periods ranging from 2 to 500 yr are presented. The results compare reasonably to a benchmark data set of global flood hazard. The developed methodology can be applied to other datasets on a global or regional scale.

K-Surfer: a KNIME extension for the management and analysis of Human brain MRI FreeSurfer/FSL data

Human brain imaging techniques, such as Magnetic Resonance Imaging (MRI) or Diffusion Tensor Imaging (DTI), have been established as scientific and diagnostic tools and their adoption is growing in popularity. Statistical methods, machine learning and data mining algorithms have successfully been adopted to extract predictive and descriptive models from neuroimage data. However, the knowledge discovery process typically requires also the adoption of pre-processing, post-processing and visualisation techniques in complex data workflows. Currently, a main problem for the integrated preprocessing and mining of MRI data is the lack of comprehensive platforms able to avoid the manual invocation of preprocessing and mining tools, that yields to an error-prone and inefficient process. In this work we present K-Surfer, a novel plug-in of the Konstanz Information Miner (KNIME) workbench, that automatizes the preprocessing of brain images and leverages the mining capabilities of KNIME in an integrated way. K-Surfer supports the importing, filtering, merging and pre-processing of neuroimage data from FreeSurfer, a tool for human brain MRI feature extraction and interpretation. K-Surfer automatizes the steps for importing FreeSurfer data, reducing time costs, eliminating human errors and enabling the design of complex analytics workflow for neuroimage data by leveraging the rich functionalities available in the KNIME workbench.

Tracking performance evaluation on PETS 2015 Challenge datasets

This paper presents a quantitative evaluation of a tracking system on PETS 2015 Challenge datasets using well-established performance measures. Using the existing tools, the tracking system implements an end-to-end pipeline that include object detection, tracking and post- processing stages. The evaluation results are presented on the provided sequences of both ARENA and P5 datasets of PETS 2015 Challenge. The results show an encouraging performance of the tracker in terms of accuracy but a greater tendency of being prone to cardinality error and ID changes on both datasets. Moreover, the analysis show a better performance of the tracker on visible imagery than on thermal imagery.

The SARS algorithm: detrending CoRoT light curves with Sysrem using simultaneous external parameters

Surveys for exoplanetary transits are usually limited not by photon noise but rather by the amount of red noise in their data. In particular, although the CoRoT space-based survey data are being carefully scrutinized, significant new sources of systematic noises are still being discovered. Recently, a magnitude-dependant systematic effect was discovered in the CoRoT data by Mazeh et al. and a phenomenological correction was proposed. Here we tie the observed effect to a particular type of effect, and in the process generalize the popular Sysrem algorithm to include external parameters in a simultaneous solution with the unknown effects. We show that a post-processing scheme based on this algorithm performs well and indeed allows for the detection of new transit-like signals that were not previously detected.

Generating Segmented Quality Meshes from Images

Techniques devoted to generating triangular meshes from intensity images either take as input a segmented image or generate a mesh without distinguishing individual structures contained in the image. These facts may cause difficulties in using such techniques in some applications, such as numerical simulations. In this work we reformulate a previously developed technique for mesh generation from intensity images called Imesh. This reformulation makes Imesh more versatile due to an unified framework that allows an easy change of refinement metric, rendering it effective for constructing meshes for applications with varied requirements, such as numerical simulation and image modeling. Furthermore, a deeper study about the point insertion problem and the development of geometrical criterion for segmentation is also reported in this paper. Meshes with theoretical guarantee of quality can also be obtained for each individual image structure as a post-processing step, a characteristic not usually found in other methods. The tests demonstrate the flexibility and the effectiveness of the approach.

Enhanced visualization of histological samples with an adjustable RGB contrast system with application for tissue used in photodynamic therapy

The analysis of histological sections has long been a valuable tool in the pathological studies. The interpretation of tissue conditions, however, relies directly on visual evaluation of tissue slides, which may be difficult to interpret because of poor contrast or poor color differentiation. The Chromatic Contrast Visualization System (CCV) combines an optical microscope with electronically controlled light-emitting diodes (LEDs) in order to generate adjustable intensities of RGB channels for sample illumination. While most image enhancement techniques rely on software post-processing of an image acquired under standard illumination conditions, CCV produces real-time variations in the color composition of the light source itself. The possibility of covering the entire RGB chromatic range, combined with the optical properties of the different tissues, allows for a substantial enhancement in image details. Traditional image acquisition methods do not exploit these visual enhancements which results in poorer visual distinction among tissue structures. Photodynamic therapy (PDT) procedures are of increasing interest in the treatment of several forms of cancer. This study uses histological slides of rat liver samples that were induced to necrosis after being exposed to PDT. Results show that visualization of tissue structures could be improved by changing colors and intensities of the microscope light source. PDT-necrosed tissue samples are better differentiated when illuminated with different color wavelengths, leading to an improved differentiation of cells in the necrosis area. Due to the potential benefits it can bring to interpretation and diagnosis, further research in this field could make CCV an attractive technique for medical applications.

Fingerprint Segmentation

In this thesis, a new algorithm has been proposed to segment the foreground of the fingerprint from the image under consideration. The algorithm uses three features, mean, variance and coherence. Based on these features, a rule system is built to help the algorithm to efficiently segment the image. In addition, the proposed algorithm combine split and merge with modified Otsu. Both enhancements techniques such as Gaussian filter and histogram equalization are applied to enhance and improve the quality of the image. Finally, a post processing technique is implemented to counter the undesirable effect in the segmented image. Fingerprint recognition system is one of the oldest recognition systems in biometrics techniques. Everyone have a unique and unchangeable fingerprint. Based on this uniqueness and distinctness, fingerprint identification has been used in many applications for a long period. A fingerprint image is a pattern which consists of two regions, foreground and background. The foreground contains all important information needed in the automatic fingerprint recognition systems. However, the background is a noisy region that contributes to the extraction of false minutiae in the system. To avoid the extraction of false minutiae, there are many steps which should be followed such as preprocessing and enhancement. One of these steps is the transformation of the fingerprint image from gray-scale image to black and white image. This transformation is called segmentation or binarization. The aim for fingerprint segmentation is to separate the foreground from the background. Due to the nature of fingerprint image, the segmentation becomes an important and challenging task. The proposed algorithm is applied on FVC2000 database. Manual examinations from human experts show that the proposed algorithm provides an efficient segmentation results. These improved results are demonstrating in diverse experiments.

Reasoning with multimedia information using symbolic projection

In many multimedia application systems, it is not the final goal to retrieve the relevant multimedia information from different multimedia information sources. Rather, post-processing of the retrieved multimedia information is needed. For example, the retrieved information is used as “known facts”. The systems will do some reasoning to obtain further conclusions based on these multimedia form “known facts”. We call this reasoning with multimedia information. Most current research work in multimedia information processing is focused on multimedia information retrieval, but post-processing the retrieved information is more or less ignored. This paper explores the way to tackle this problem by using symbolic projection. A case study of reasoning with still image information is presented. Some extensions to symbolic projection- introducing auxiliary pictorial objects in symbolic pictures that need to be processed-are discussed. We expect this paper will stimulate further research on this important but ignored topic.

Knowledge based optimisation techniques for sheet metal forming die design

A knowledge based optimism technique has been developed to predict solutions for quality issues found in an initial draw die design. Post processing of the initial design yields all the features applying forces, and major quality issues. Using the geometric relationship between the two, a knowledge-base is interrogated to determine the possible corrective actions. These actions are then passed through a fast semi-analytical model to determine the level of change required. Results from a 2D forming are presented to highlight the advantage of the new algorithms over current optimisation techniques.

A procedural, minimal input, natural terrain plug-in for Blender

This project uses methods of terrain representation, creation and realism described in literature. We find that using a combination of Fractional Brownian Motion and procedural formation of rivers via squig curves to form initial terrain, with hydraulic erosion for post processing, we have full control over the style of terrain: from jagged mountains to flat regions; and the phase of river from tightly rock controlled to flood plain regions.

Super-resolution of a 3-dimensional scene from novel viewpoints

Super-resolution is a method of post-processing image enhancement that increases the spatial resolution of video or images. Existing super-resolution techniques apply only to images captured of a planar scene. This paper aims to extend super-resolution concepts from the 2D domain to the 3D domain, drawing on ideas from both superresolution and multi-view geometry, two fields of research that until now have predominantly been studied in isolation. 2D super-resolution methods are not without their complexities and limitations. However, once multiple views of a scene are considered within a super-resolution framework, a new range of issues arise that must also be resolved. For example, when input images of a scene with variation in depth are considered, it is no longer clear how and where the images should be registered. This paper describes the use of sparse 3D reconstruction in order to ‘register’ the input images, which are then transferred to a novel image plane and combined to increase the perceived detail in the scene. Experimental results using real images captured from generally positioned input cameras are presented.

Extracting 3D mesh skeletons using antipodal points locations

Finding the skeleton of a 3D mesh is an essential task for many applications such as mesh animation, tracking, and 3D registeration. In recent years, new technologies in computer vision such as Microsoft Kinect have proven that a mesh skeleton can be useful such as in the case of human machine interactions. To calculate the 3D mesh skeleton, the mesh properties such as topology and its components relations are utilized. In this paper, we propose the usage of a novel algorithm that can efficiently calculate a vertex antipodal point. A vertex antipodal point is the diametrically opposite point that belongs to the same mesh. The set of centers of the connecting lines between each vertex and its antipodal point represents the 3D mesh desired skeleton. Post processing is completed for smoothing and fitting centers into optimized skeleton parts. The algorithm is tested on different classes of 3D objects and produced efficient results that are comparable with the literature. The algorithm has the advantages of producing high quality skeletons as it preserves details. This is suitable for applications where the mesh skeleton mapping is required to be kept as much as possible.