Robust and unbiased estimation of the background distribution for automated quantitative imaging

Quantitative analysis of high-throughput microscopy images requires robust automated algorithms. Background estimation is usually the first step and has an impact on all subsequent analysis, in particular for foreground detection and calculation of ratiometric quantities. Most methods recover only a single background value, such as the median. Those that aim to retrieve a background distribution by dividing the intensity histogram yield a biased estimation in images in non-trivial cases. In this work, we present the first method to recover an unbiased estimation of the background distribution directly from an image and without any additional input. Through a robust statistical test, our method leverages the lack of local spatial correlation in background pixels to select a subset of pixels that accurately represent the background distribution. This method is both fast and simple to implement, as it only uses standard mathematical operations and an averaging filter. Additionally, the only parameter, the size of the averaging filter, does not require fine tuning. The obtained background distribution can be used to test for foreground membership of individual pixels, or to estimate confidence intervals in derived quantities. We expect that the concepts described in this work can help to develop a novel family of robust segmentation methods.

One stop shop IV: taxonomic update with molecular phylogeny for important phytopathogenic genera: 76–100 (2020)

This is a continuation of a series focused on providing a stable platform for the taxonomy of phytopathogenic fungi and fungus-like organisms. This paper focuses on one family: Erysiphaceae and 24 phytopathogenic genera: Armillaria, Barriopsis, Cercospora, Cladosporium, Clinoconidium, Colletotrichum, Cylindrocladiella, Dothidotthia,, Fomitopsis, Ganoderma, Golovinomyces, Heterobasidium, Meliola, Mucor, Neoerysiphe, Nothophoma, Phellinus, Phytophthora, Pseudoseptoria, Pythium, Rhizopus, Stemphylium, Thyrostroma and Wojnowiciella. Each genus is provided with a taxonomic background, distribution, hosts, disease symptoms, and updated backbone trees. Species confirmed with pathogenicity studies are denoted when data are available. Six of the genera are updated from previous entries as many new species have been described.

Locating Seismo-Conductivity Anomaly before the 2017 MW 6.5 Jiuzhaigou Earthquake in China Using Far Magnetic Stations

Changes in the underlying conductivity around hypocenters are generally considered one of the promising mechanisms of seismo-electromagnetic anomaly generation. Parkinson vectors are indicators of high-conductivity materials and were utilized to remotely monitor conductivity changes during the MW 6.5 Jiuzhaigou earthquake (103.82°E, 33.20°N) on 8 August 2017. Three-component geomagnetic data recorded in 2017 at nine magnetic stations with epicenter distances of 63–770 km were utilized to compute the azimuths of the Parkinson vectors based on the magnetic transfer function. The monitoring and background distributions at each station were constructed by using the azimuths within a 15-day moving window and over the entire study period, respectively. The background distribution was subtracted from the monitoring distribution to mitigate the effects of underlying inhomogeneous electric conductivity structures. The differences obtained at nine stations were superimposed and the intersection of a seismo-conductivity anomaly was located about 70 km away from the epicenter about 17 days before the earthquake. The anomaly disappeared about 7 days before and remained insignificant after the earthquake. Analytical results suggested that the underlying conductivity close to the hypocenter changed before the Jiuzhaigou earthquake. These changes can be detected simultaneously by using multiple magnetometers located far from the epicenter. The disappearance of the seismo-conductivity anomaly after the earthquake sheds light on a promising candidate of the pre-earthquake anomalous phenomena.

Finding binary star fractions in any distribution

Candidate visual binary systems are often found by identifying two stars that are closer together than would be expected by chance. However, in regions with non-trivial density distributions, the ‘random’ distances between stars varies because of the background distribution, as well as the presence of binaries. We show that when no binaries are present, the distribution of the ratios of the distances to the nearest and tenth nearest neighbours, d1/d10, is always well approximated by a Gaussian with mean 0.2–0.3 and variance 0.16–0.19 for any underlying density distribution. The introduction of binaries causes some (or all) nearest neighbours to become closer than expected by random chance, introducing a component to the distribution where d1/d10 is much lower than expected. We show how a simple single or double Gaussian fit to the distribution of d1/d10 can be used to find the binary fraction in any underlying density distribution quickly and simply.

Analysis of the Use of Background Distribution for Naive Bayes Classifiers

The naive Bayes classifier is a popular classifier, as it is easy to train, requires no cross-validation for parameter tuning, and can be easily extended due to its generative model. Moreover, recently it was shown that the word probabilities (background distribution) estimated from large unlabeled corpora could be used to improve the parameter estimation of naive Bayes. However, previous methods do not explicitly allow to control how much the background distribution can influence the estimation of naive Bayes parameters. In contrast, we investigate an extension of the graphical model of naive Bayes such that a word is either generated from a background distribution or from a class-specific word distribution. We theoretically analyze this model and show the connection to Jelinek-Mercer smoothing. Experiments using four standard text classification data sets show that the proposed method can statistically significantly outperform previous methods that use the same background distribution.

Salient Object Detection Based on Background Feature Clustering

Automatic estimation of salient object without any prior knowledge tends to greatly enhance many computer vision tasks. This paper proposes a novel bottom-up based framework for salient object detection by first modeling background and then separating salient objects from background. We model the background distribution based on feature clustering algorithm, which allows for fully exploiting statistical and structural information of the background. Then a coarse saliency map is generated according to the background distribution. To be more discriminative, the coarse saliency map is enhanced by a two-step refinement which is composed of edge-preserving element-level filtering and upsampling based on geodesic distance. We provide an extensive evaluation and show that our proposed method performs favorably against other outstanding methods on two most commonly used datasets. Most importantly, the proposed approach is demonstrated to be more effective in highlighting the salient object uniformly and robust to background noise.