This article was republished on April 6, 2015, to correct errors that were introduced during the typesetting process. Fig 4 was incorrectly cropped, and we have also corrected some typographical errors in the text. The publisher apologizes for these errors. Please download this article again to view the correct version. The originally published, uncorrected article and the republished, correcte...

This paper presents an improved version of the features based on Discrete Fourier Transform (DFT) for texture description that demonstrate robustness against rotation. The features have been tested on cropped parts of textures from Brodatz collection and their rotated versions. The results show improved performance for both, recognition as well as retrieval as compared to texture features based...

This study provides food security situation in India for the next two decades. As expected, total physical land will remain more or less flat in the next two decade. But net cropped area is projected to increase through higher cropping intensity. More than 70 percent of additional area is likely to switch to horticultural and permanent crops. Among crops, oilseeds such as soybeans, rapeseed, an...

Background Andosols are generally characterized by strong resilience to degradation and high soil fertility. This may decline during long-term peanut (Arachis hypogea) monoculture, as indicated chemical biological properties. Aims The study investigated the monoculture-induced changes in environment driver for Methods In this on-farm study, soils from seven sites cropped with monoculture differ...

Cropping methods that induce the least amount of stress are key in ensuring acceptable meat quality from wild ungulates. The purpose of this study was to compare the effect of day and night cropping on the quality of Red Hartebeest (Alcelaphus buselaphus) meat. Ten animals were cropped during the day and ten at night. Emphasis was placed on ultimate pH (pHu) and rate of pH decline. All pH measu...

An end-to-end method for multi-language scene text localization, recognition and script identification is proposed. The approach is based on a set of convolutional neural nets. The method, called E2E-MLT, achieves state-of-theart performance for both joint localization and script identification in natural images and in cropped word script identification. E2E-MLT is the first published multi-lan...