Recognizing lines of unconstrained handwritten text is a challenging task. The difficulty of segmenting cursive or overlapping characters, combined with the need to exploit surrounding context, has ...led to low recognition rates for even the best current recognizers. Most recent progress in the field has been made either through improved preprocessing or through advances in language modeling. Relatively little work has been done on the basic recognition algorithms. Indeed, most systems rely on the same hidden Markov models that have been used for decades in speech and handwriting recognition, despite their well-known shortcomings. This paper proposes an alternative approach based on a novel type of recurrent neural network, specifically designed for sequence labeling tasks where the data is hard to segment and contains long-range bidirectional interdependencies. In experiments on two large unconstrained handwriting databases, our approach achieves word recognition accuracies of 79.7 percent on online data and 74.1 percent on offline data, significantly outperforming a state-of-the-art HMM-based system. In addition, we demonstrate the network's robustness to lexicon size, measure the individual influence of its hidden layers, and analyze its use of context. Last, we provide an in-depth discussion of the differences between the network and HMMs, suggesting reasons for the network's superior performance.
Daptomycin is a cyclic lipopeptide antibacterial agent with potent bactericidal activity against a broad range of Gram-positive organisms. In 2003, daptomycin for injection received approval from the ...US Food and Drug Administration (FDA) for the treatment of patients with complicated skin and skin structure infections (cSSSIs); in 2006, it was approved for the treatment of patients with
bacteremia, including those with right-sided infective endocarditis caused by methicillin-susceptible and methicillin-resistant isolates. In 2016, the FDA approved a new formulation of daptomycin for injection (daptomycin RF) for the same indications. The efficacy and safety of daptomycin for injection have been established in pivotal clinical trials, and the findings of nonclinical studies indicate that both formulations of daptomycin for injection are equivalent. Herein we refer to the new daptomycin formulation as daptomycin RF to distinguish it from the original formulation. Daptomycin RF provides clinicians and clinical pharmacists with a product that offers improved stability and more rapid, in-vial reconstitution with either sterile or bacteriostatic water for injection, while maintaining the same antibacterial coverage. Here we discuss the rationale for and the potential value of daptomycin RF, and briefly review the similarities and differences between the original formulation of daptomycin and daptomycin RF.
This paper investigates the generation and use of classifier ensembles for offline handwritten text recognition. The ensembles are derived from the integration of a language model in the hidden ...Markov model based recognition system. The word sequences output by the ensemble members are aligned and combined according to the ROVER framework. The addressed environment is extreme because of the existence of a large number of word classes. Moreover, the recognisers do not produce single output classes but sequences of classes. Experiments conducted on the IAM database show that the ensemble methods are able to produce statistically significant improvements in the word level accuracy when compared to the base recogniser.
This work investigates three different rejection strategies for offline handwritten sentence recognition. The rejection strategies are implemented as a postprocessing step of a hidden Markov model ...based text recognition system and are based on confidence measures derived from a list of candidate sentences produced by the recognizer. The better performing confidence measures make use of the fact that the recognizer integrates a word bigram language model. Experimental results on extracted sentences from the IAM database validate the effectiveness of the proposed rejection strategies.
In this paper we apply a novel non-uniform slant correction preprocessing technique to improve the recognition of offline handwritten text lines. The local slant correction is expressed as a global ...optimisation problem of the sequence of local slant angles. This is different to conventional slant removal techniques that rely on the average slant angle. Experiments based on a state-of-the-art handwritten text line recogniser show a significant gain in word level accuracy for the investigated preprocessing methods.
This paper compares two different methods to combine feature streams to improve the performance of offline handwritten text line recognition systems. In both methods a pixel-based and a geometric ...feature stream are combined. The first method integrates the feature streams at an early stage whereas in the second method a combination step at the decision level is applied. In the experiments, the early integration approach outperforms the decision level combination as well as recognisers built from the individual feature streams
This paper investigates the generation and use of multiple recognition results to improve the performance of an offline handwritten text line recognition system. Multiple recognition results are ...created by specific integration of a language model in the hidden Markov model based recognition system. The ROVER algorithm is applied to combine the multiple results. Experiments conducted on the IAM database show that the proposed system is able to produce statistically significant improvements in the recognition rate compared to the original system.
A novel method to reduce the lexicon size in handwritten word recognition is proposed in this paper. Due to large lexica, the computational cost of current word recognisers is often too high for ...practical applications. The proposed lexicon reduction method is based on character shape codes. We examine four different shape code mappings based on machine printed character font, on handprint, and on cursive handwriting. Experimental evaluation shows that the proposed method can reduce the computational effort while keeping the recognition rate high.
The mass of a star is the most fundamental parameter for its structure, evolution, and final fate. It is particularly important for any kind of stellar archaeology and characterization of exoplanets. ...There exist a variety of methods in astronomy to estimate or determine it. In this review we present a significant number of such methods, beginning with the most direct and model-independent approach using detached eclipsing binaries. We then move to more indirect and model-dependent methods, such as the quite commonly used isochrone or stellar track fitting. The arrival of quantitative asteroseismology has opened a completely new approach to determine stellar masses and to complement and improve the accuracy of other methods. We include methods for different evolutionary stages, from the pre-main sequence to evolved (super)giants and final remnants. For all methods uncertainties and restrictions will be discussed. We provide lists of altogether more than 200 benchmark stars with relative mass accuracies between
0.3
,
2
%
for the covered mass range of
M
∈
0.1
,
16
M
⊙
,
75
%
of which are stars burning hydrogen in their core and the other
25
%
covering all other evolved stages. We close with a recommendation how to combine various methods to arrive at a “mass-ladder” for stars.
New Cooling Sequences for Old White Dwarfs Renedo, I; Althaus, L. G; Miller Bertolami, M. M ...
The Astrophysical journal,
07/2010, Volume:
717, Issue:
1
Journal Article
Peer reviewed
Open access
We present full evolutionary calculations appropriate for the study of hydrogen-rich DA white dwarfs. This is done by evolving white dwarf progenitors from the zero-age main sequence, through the ...core hydrogen-burning phase, the helium-burning phase, and the thermally pulsing asymptotic giant branch phase to the white dwarf stage. Complete evolutionary sequences are computed for a wide range of stellar masses and for two different metallicities, Z = 0.01, which is representative of the solar neighborhood, and Z = 0.001, which is appropriate for the study of old stellar systems, like globular clusters. During the white dwarf cooling stage, we self-consistently compute the phase in which nuclear reactions are still important, the diffusive evolution of the elements in the outer layers and, finally, we also take into account all the relevant energy sources in the deep interior of the white dwarf, such as the release of latent heat and the release of gravitational energy due to carbon-oxygen phase separation upon crystallization. We also provide colors and magnitudes for these sequences, based on a new set of improved non-gray white dwarf model atmospheres, which include the most up-to-date physical inputs like the Ly{alpha} quasi-molecular opacity. The calculations are extended down to an effective temperature of 2500 K. Our calculations provide a homogeneous set of evolutionary cooling tracks appropriate for mass and age determinations of old DA white dwarfs and for white dwarf cosmochronology of the different Galactic populations.
PDF
