Accelerating taxonomic discovery through automated character extraction

This paper discusses the following key messages. Taxonomy is (and taxonomists are) more important than ever in times of global change. Taxonomic endeavour is not occurring fast enough: in 250 years since the creation of the Linnean Systema Naturae, only about 20% of Earth's species have been named. We need fundamental changes to the taxonomic process and paradigm to increase taxonomic productivity by orders of magnitude. Currently, taxonomic productivity is limited principally by the rate at which we capture and manage morphological information to enable species discovery. Many recent (and welcomed) initiatives in managing and delivering biodiversity information and accelerating the taxonomic process do not address this bottleneck. Development of computational image analysis and feature extraction methods is a crucial missing capacity needed to enable taxonomists to overcome the taxonomic impediment in a meaningful time frame. Copyright © 2009 Magnolia Press.

What patient information allows us to make accurate predictions of outcome?

Only some of the information contained in a medical record will be useful to the prediction of patient outcome. We describe a novel method for selecting those outcome predictors which allow us to reliably discriminate between adverse and benign end results. Using the area under the receiver operating characteristic as a nonparametric measure of discrimination, we show how to calculate the maximum discrimination attainable with a given set of discrete valued features. This upper limit forms the basis of our feature selection algorithm. We use the algorithm to select features (from maternity records) relevant to the prediction of failure to progress in labour. The results of this analysis motivate investigation of those predictors of failure to progress relevant to parous and nulliparous sub-populations.

Using upper bounds on attainable discrimination to select discrete valued features

Selection of features that will permit accurate pattern classification is a difficult task. However, if a particular data set is represented by discrete valued features, it becomes possible to determine empirically the contribution that each feature makes to the discrimination between classes. This paper extends the discrimination bound method so that both the maximum and average discrimination expected on unseen test data can be estimated. These estimation techniques are the basis of a backwards elimination algorithm that can be use to rank features in order of their discriminative power. Two problems are used to demonstrate this feature selection process: classification of the Mushroom Database, and a real-world, pregnancy related medical risk prediction task - assessment of risk of perinatal death.

Feature selection using expected attainable discrimination

We propose expected attainable discrimination (EAD) as a measure to select discrete valued features for reliable discrimination between two classes of data. EAD is an average of the area under the ROC curves obtained when a simple histogram probability density model is trained and tested on many random partitions of a data set. EAD can be incorporated into various stepwise search methods to determine promising subsets of features, particularly when misclassification costs are difficult or impossible to specify. Experimental application to the problem of risk prediction in pregnancy is described.

Structural damage detection method using frequency response functions

Structural damage detection using measured dynamic data for pattern recognition is a promising approach. These pattern recognition techniques utilize artificial neural networks and genetic algorithm to match pattern features. In this study, an artificial neural network–based damage detection method using frequency response functions is presented, which can effectively detect nonlinear damages for a given level of excitation. The main objective of this article is to present a feasible method for structural vibration–based health monitoring, which reduces the dimension of the initial frequency response function data and transforms it into new damage indices and employs artificial neural network method for detecting different levels of nonlinearity using recognized damage patterns from the proposed algorithm. Experimental data of the three-story bookshelf structure at Los Alamos National Laboratory are used to validate the proposed method. Results showed that the levels of nonlinear damages can be identified precisely by the developed artificial neural networks. Moreover, it is identified that artificial neural networks trained with summation frequency response functions give higher precise damage detection results compared to the accuracy of artificial neural networks trained with individual frequency response functions. The proposed method is therefore a promising tool for structural assessment in a real structure because it shows reliable results with experimental data for nonlinear damage detection which renders the frequency response function–based method convenient for structural health monitoring.

Improved Security Analysis of Fugue-256 (Poster)

We present some improved analytical results as part of the ongoing work on the analysis of Fugue-256 hash function, a second round candidate in the NIST’s SHA3 competition. First we improve Aumasson and Phans’ integral distinguisher on the 5.5 rounds of the final transformation of Fugue-256 to 16.5 rounds. Next we improve the designers’ meet-in-the-middle preimage attack on Fugue-256 from 2480 time and memory to 2416. Finally, we comment on possible methods to obtain free-start distinguishers and free-start collisions for Fugue-256.

Corner detection in images under different noise levels

Corner detection has shown its great importance in many computer vision tasks. However, in real-world applications, noise in the image strongly affects the performance of corner detectors. Few corner detectors have been designed to be robust to heavy noise by now, partly because the noise could be reduced by a denoising procedure. In this paper, we present a corner detector that could find discriminative corners in images contaminated by noise of different levels, without any denoising procedure. Candidate corners (i.e., features) are firstly detected by a modified SUSAN approach, and then false corners in noise are rejected based on their local characteristics. Features in flat regions are removed based on their intensity centroid, and features on edge structures are removed using the Harris response. The detector is self-adaptive to noise since the image signal-to-noise ratio (SNR) is automatically estimated to choose an appropriate threshold for refining features. Experimental results show that our detector has better performance at locating discriminative corners in images with strong noise than other widely used corner or keypoint detectors.

Scale adaptive tracking using mean shift and efficient feature matching

The mean shift tracker has achieved great success in visual object tracking due to its efficiency being nonparametric. However, it is still difficult for the tracker to handle scale changes of the object. In this paper, we associate a scale adaptive approach with the mean shift tracker. Firstly, the target in the current frame is located by the mean shift tracker. Then, a feature point matching procedure is employed to get the matched pairs of the feature point between target regions in the current frame and the previous frame. We employ FAST-9 corner detector and HOG descriptor for the feature matching. Finally, with the acquired matched pairs of the feature point, the affine transformation between target regions in the two frames is solved to obtain the current scale of the target. Experimental results show that the proposed tracker gives satisfying results when the scale of the target changes, with a good performance of efficiency.