A Largest Matching Area Approach to Image Denoising

Given the success of patch-based approaches to image denoising,this paper addresses the ill-posed problem of patch size selection.Large patch sizes improve noise robustness in the presence of good matches, but can also lead to artefacts in textured regions due to the rare patch effect; smaller patch sizes reconstruct details more accurately but risk over-fitting to the noise in uniform regions. We propose to jointly optimize each matching patch’s identity and size for gray scale image denoising, and present several implementations.The new approach effectively selects the largest matching areas, subject to the constraints of the available data and noise level, to improve noise robustness. Experiments on standard test images demonstrate our approach’s ability to improve on fixed-size reconstruction, particularly at high noise levels, on smoother image regions.

Distribution-based adaptive bounding genetic algorithm for continuous optimisation problems

A new search-space-updating technique for genetic algorithms is proposed for continuous optimisation problems. Other than gradually reducing the search space during the evolution process with a fixed reduction rate set ‘a priori’, the upper and the lower boundaries for each variable in the objective function are dynamically adjusted based on its distribution statistics. To test the effectiveness, the technique is applied to a number of benchmark optimisation problems in comparison with three other techniques, namely the genetic algorithms with parameter space size adjustment (GAPSSA) technique [A.B. Djurišic, Elite genetic algorithms with adaptive mutations for solving continuous optimization problems – application to modeling of the optical constants of solids, Optics Communications 151 (1998) 147–159], successive zooming genetic algorithm (SZGA) [Y. Kwon, S. Kwon, S. Jin, J. Kim, Convergence enhanced genetic algorithm with successive zooming method for solving continuous optimization problems, Computers and Structures 81 (2003) 1715–1725] and a simple GA. The tests show that for well-posed problems, existing search space updating techniques perform well in terms of convergence speed and solution precision however, for some ill-posed problems these techniques are statistically inferior to a simple GA. All the tests show that the proposed new search space update technique is statistically superior to its counterparts.

The exponentiated convex variable step-size (ECVSS) algorithm

For some time there is a large interest in variable step-size methods for adaptive filtering. Recently, a few stochastic gradient algorithms have been proposed, which are based on cost functions that have exponential dependence on the chosen error. However, we have experienced that the cost function based on exponential of the squared error does not always satisfactorily converge. In this paper we modify this cost function in order to improve the convergence of exponentiated cost function and the novel ECVSS (exponentiated convex variable step-size) stochastic gradient algorithm is obtained. The proposed technique has attractive properties in both stationary and abrupt-change situations. (C) 2010 Elsevier B.V. All rights reserved.

Current brood size and residual reproductive value predict offspring desertion in the burying beetle Nicrophorus vespilloides

Life-history theory suggests that offspring desertion can be an adaptive reproductive strategy, in which parents forgo the costly care of an unprofitable current brood to save resources for future reproduction. In the burying beetle, Nicrophorus vespilloides, parents commonly abandon their offspring to the care of others, resulting in female-only care, male-only care, brood parasitism, and the care of offspring sired by satellite males. Furthermore, when there is biparental care, males routinely desert the brood before larval development is complete, leaving females behind to tend their young. We attempted to understand these patterns of offspring desertion by using laboratory experiments to compare the fitness costs associated with parental care for each sex and the residual reproductive value of the 2 sexes. We also tested whether current brood size and residual reproductive value together predicted the incidence of brood desertion. We found that males and females each sustained fecundity costs as a consequence of caring for larvae and that these costs were of comparable magnitude. Nevertheless, males had greater residual reproductive value than females and were more likely than females to desert experimental broods. Our results can explain why males desert the brood earlier than females in nature and why female-only care is more common than male-only care. They also suggest that the tipping point from brood parasitism or satellite male behavior to communal breeding (and vice versa) depends on the value of the current brood relative to residual reproductive value.

Body size differs in workers produced across time and is associated with variation in the quantity and composition of larval food in Nannotrigona perilampoides (Hymenoptera, Meliponini)

Although variation in body size has been recently reported in stingless bees (Meliponini), empirical evidence evaluating possible factors related to such variation is lacking, and thus it is not clear if it may have an adaptive significance. We evaluated if variation in the body size and weight of workers of stingless bees fluctuates across a seasonal pattern and if this could be related to characteristics of the food consumed during the larval stage. The weight of larval provisions, their protein, and sugar content were evaluated in four colonies of Nannotrigona perilampoides every 2 months across 1 year. Worker-destined larvae from the same combs were allowed to develop and were sampled as callow workers to determine their weight and size using morphometric data. The weight and size of workers were highly correlated and varied across the seasons in established colonies, suggesting that size variation cycles across the year in stingless bees. An increase in the protein content and, to a lesser degree, the quantity of larval food were positively linked to variation in body weight and size; food with richer protein content resulted in larger and heavier workers. This study provides the first evidence of an effect of the quantity and composition of larval food on the size of workers in stingless bees. Although body weight and size of workers differed across seasons, they were not readily noticeable as changes seem to occur as a continuum across the year. Since size polymorphism was of a larger magnitude across time but not within age cohorts and as it was highly determined by food resources, it may not be an adaptive feature in stingless bees. However, more studies are needed to determine the role of the cyclical change in worker body size on colony performance and thus its adaptive significance in stingless bees.

Parsing parallel evolution:Ecological divergence and differential gene expression in the adaptive radiations of thick-lipped Midas cichlid fishes from Nicaragua

The study of parallel evolution facilitates the discovery of common rules of diversification. Here, we examine the repeated evolution of thick lips in Midas cichlid fishes (the Amphilophus citrinellus species complex) - from two Great Lakes and two crater lakes in Nicaragua - to assess whether similar changes in ecology, phenotypic trophic traits and gene expression accompany parallel trait evolution. Using next-generation sequencing technology, we characterize transcriptome-wide differential gene expression in the lips of wild-caught sympatric thick- and thin-lipped cichlids from all four instances of repeated thick-lip evolution. Six genes (apolipoprotein D, myelin-associated glycoprotein precursor, four-and-a-half LIM domain protein 2, calpain-9, GTPase IMAP family member 8-like and one hypothetical protein) are significantly underexpressed in the thick-lipped morph across all four lakes. However, other aspects of lips' gene expression in sympatric morphs differ in a lake-specific pattern, including the magnitude of differentially expressed genes (97-510). Generally, fewer genes are differentially expressed among morphs in the younger crater lakes than in those from the older Great Lakes. Body shape, lower pharyngeal jaw size and shape, and stable isotopes (dC and dN) differ between all sympatric morphs, with the greatest differentiation in the Great Lake Nicaragua. Some ecological traits evolve in parallel (those related to foraging ecology; e.g. lip size, body and head shape) but others, somewhat surprisingly, do not (those related to diet and food processing; e.g. jaw size and shape, stable isotopes). Taken together, this case of parallelism among thick- and thin-lipped cichlids shows a mosaic pattern of parallel and nonparallel evolution. © 2012 Blackwell Publishing Ltd.