Machine Learning is the study of methods for programming computers to learn. Computers are applied to a wide range of tasks, and for most of these it is relatively easy for programmers to design and implement the necessary software. However, there are many tasks for which this is difficult or impossible. These can be divided into four general categories. First, there are problems for which there exist no human experts. For example, in modern automated manufacturing facilities, there is a need to predict machine failures before they occur by analyzing sensor readings. Because the machines are new, there are no human experts who can be interviewed by a programmer to provide the knowledge necessary to build a computer system. A machine learning system can study recorded data and subsequent machine failures and learn prediction rules. Second, there are problems where human experts exist, but where they are unable to explain their expertise. This is the case in many perceptual tasks, such as speech recognition, hand-writing recognition, and natural language understanding. Virtually all humans exhibit expert-level abilities on these tasks, but none of them can describe the detailed steps that they follow as they perform them. Fortunately, humans can provide machines with examples of the inputs and correct outputs for these tasks, so machine learning algorithms can learn to map the inputs to the outputs. Third, there are problems where phenomena are changing rapidly. In finance, for example, people would like to predict the future behavior of the stock market, of consumer purchases, or of exchange rates. These behaviors change frequently, so that even if a programmer could construct a good predictive computer program, it would need to be rewritten frequently. A learning program can relieve the programmer of this burden by constantly modifying and tuning a set of learned prediction rules. Fourth, there are applications that need to be customized for each computer user separately. Consider, for example, a program to filter unwanted electronic mail messages. Different users will need different filters. It is unreasonable to expect each user to program his or her own rules, and it is infeasible to provide every user with a software engineer to keep the rules up-to-date. A machine learning system can learn which mail messages the user rejects and maintain the filtering rules automatically. Machine learning addresses many of the same research questions as the fields of statistics, data mining, and psychology, but with differences of emphasis. Statistics focuses on understanding the phenomena that have generated the data, often with the goal of testing different hypotheses about those phenomena. Data mining seeks to find patterns in the data that are understandable by people. Psychological studies of human learning aspire to understand the mechanisms underlying the various learning behaviors exhibited by people (concept learning, skill acquisition, strategy change, etc.).

Machine learning

Cause, conseguenze e strategie di mitigazione Proponiamo il primo di una serie di articoli in cui affronteremo l’attuale problema dei mutamenti climatici. Presentiamo il documento redatto, votato e pubblicato dall’Ipcc - Comitato intergovernativo sui cambiamenti climatici - che illustra la sintesi delle ricerche svolte su questo tema rilevante.

/pdf/climate-change-2007-the-physical-science-basis-1x93ttz9jt.pdf

Climate Change 2007: The Physical Science Basis.

This review concentrates on advances in nitric oxide synthase (NOS) structure, function and inhibition made in the last seven years, during which time substantial advances have been made in our understanding of this enzyme family. There is now information on the enzyme structure at all levels from primary (amino acid sequence) to quaternary (dimerization, association with other proteins) structure. The crystal structures of the oxygenase domains of inducible NOS (iNOS) and vascular endothelial NOS (eNOS) allow us to interpret other information in the context of this important part of the enzyme, with its binding sites for iron protoporphyrin IX (haem), biopterin, L-arginine, and the many inhibitors which interact with them. The exact nature of the NOS reaction, its mechanism and its products continue to be sources of controversy. The role of the biopterin cofactor is now becoming clearer, with emerging data implicating one-electron redox cycling as well as the multiple allosteric effects on enzyme activity. Regulation of the NOSs has been described at all levels from gene transcription to covalent modification and allosteric regulation of the enzyme itself. A wide range of NOS inhibitors have been discussed, interacting with the enzyme in diverse ways in terms of site and mechanism of inhibition, time-dependence and selectivity for individual isoforms, although there are many pitfalls and misunderstandings of these aspects. Highly selective inhibitors of iNOS versus eNOS and neuronal NOS have been identified and some of these have potential in the treatment of a range of inflammatory and other conditions in which iNOS has been implicated.

/pdf/nitric-oxide-synthases-structure-function-and-inhibition-1t31plqilr.pdf

Nitric oxide synthases: structure, function and inhibition

I read this book the same weekend that the Packers took on the Rams, and the experience of the latter event, obviously, colored my judgment. Although I abhor anything that smacks of being a handbook (like, \"How to Earn a Merit Badge in Neurosurgery\") because too many volumes in biomedical science already evince a boyscout-like approach, I must confess that parts of this volume are fast, scholarly, and significant, with certain reservations. I like parts of this well-illustrated book because Dr. Sj6strand, without so stating, develops certain subjects on technique in relation to the acquisition of judgment and sophistication. And this is important! So, given that the author (like all of us) is somewhat deficient in some areas, and biased in others, the book is still valuable if the uninitiated reader swallows it in a general fashion, realizing full well that what will be required from the reader is a modulation to fit his vision, propreception, adaptation and response, and the kind of problem he is undertaking. A major deficiency of this book is revealed by comparison of its use of physics and of chemistry to provide understanding and background for the application of high resolution electron microscopy to problems in biology. Since the volume is keyed to high resolution electron microscopy, which is a sophisticated form of structural analysis, but really morphology in a modern guise, the physical and mechanical background of The instrument and its ancillary tools are simply and well presented. The potential use of chemical or cytochemical information as it relates to biological fine structure , however, is quite deficient. I wonder when even sophisticated morphol-ogists will consider fixation a reaction and not a technique; only then will the fundamentals become self-evident and predictable and this sine qua flon will become less mystical. Staining reactions (the most inadequate chapter) ought to be something more than a technique to selectively enhance contrast of morphological elements; it ought to give the structural addresses of some of the chemical residents of cell components. Is it pertinent that auto-radiography gets singled out for more complete coverage than other significant aspects of cytochemistry by a high resolution microscopist, when it has a built-in minimal error of 1,000 A in standard practice? I don't mean to blind-side (in strict football terminology) Dr. Sj6strand's efforts for what is \"routinely used in our laboratory\"; what is done is usually well done. It's just that …

Methods in Enzymology.

Image classification is a complex process that may be affected by many factors. This paper examines current practices, problems, and prospects of image classification. The emphasis is placed on the summarization of major advanced classification approaches and the techniques used for improving classification accuracy. In addition, some important issues affecting classification performance are discussed. This literature review suggests that designing a suitable image-processing procedure is a prerequisite for a successful classification of remotely sensed data into a thematic map. Effective use of multiple features of remotely sensed data and the selection of a suitable classification method are especially significant for improving classification accuracy. Non-parametric classifiers such as neural network, decision tree classifier, and knowledge-based classification have increasingly become important approaches for multisource data classification. Integration of remote sensing, geographical information systems (GIS), and expert system emerges as a new research frontier. More research, however, is needed to identify and reduce uncertainties in the image-processing chain to improve classification accuracy.

/pdf/a-survey-of-image-classification-methods-and-techniques-for-5bc5ltr9dq.pdf

A survey of image classification methods and techniques for improving classification performance

. With rapid globalization, the extent of built-up areas is continuously increasing. Extraction of features for classifying built-up areas that are more robust and abstract is a leading research topic from past many years. Although, various studies have been carried out where spatial information along with spectral features has been utilized to enhance the accuracy of classification. Still, these feature extraction techniques require a large number of user-specific parameters and generally application specific. On the other hand, recently introduced Deep Learning (DL) techniques requires less number of parameters to represent more abstract aspects of the data without any manual effort. Since, it is difficult to acquire high-resolution datasets for applications that require large scale monitoring of areas. Therefore, in this study Sentinel-2 image has been used for built-up areas extraction. In this work, pre-trained Convolutional Neural Networks (ConvNets) i.e. Inception v3 and VGGNet are employed for transfer learning. Since these networks are trained on generic images of ImageNet dataset which are having very different characteristics from satellite images. Therefore, weights of networks are fine-tuned using data derived from Sentinel-2 images. To compare the accuracies with existing shallow networks, two state of art classifiers i.e. Gaussian Support Vector Machine (SVM) and Back-Propagation Neural Network (BP-NN) are also implemented. Both SVM and BP-NN gives 84.31 % and 82.86 % overall accuracies respectively. Inception-v3 and VGGNet gives 89.43 % of overall accuracy using fine-tuned VGGNet and 92.10 % when using Inception-v3. The results indicate high accuracy of proposed fine-tuned ConvNets on a 4-channel Sentinel-2 dataset for built-up area extraction.

/pdf/extraction-of-built-up-areas-using-convolutional-neural-4hd1o2f79p.pdf

Extraction of built-up areas using convolutional neural networks and transfer learning from sentinel-2 satellite images

The use of visual saliency for perceptual enhancement of images has drawn significant attention. In this paper, we explore the idea of selectively enhancing salient regions of an image. Moreover, we develop an algorithm based on adaptive bilateral filtering for this purpose. In most of the filtering based methods, detail enhancement is performed by decomposing the image into base and detail layers; the detail layer is amplified and added back to the base layer to obtain the enhanced image. The decomposition is performed using edge-preserving smoothing such as bilateral filtering. The present novelty is that we use the saliency map to locally guide the smoothing (and the enhancement) action of the bilateral filter. The effectiveness of our proposal is demonstrated using visual results. In particular, our method does not suffer from gradient reversals and halo artifacts, and does not amplify fine details in non-salient regions that often appear as noise grains in the enhanced image. Moreover, if we choose to perform the filtering channelwise, then our method can be efficiently implemented using an existing fast algorithm.

Saliency Guided Image Detail Enhancement

Motivated by the increasing availability and importance of high radiometric resolution remote sensing data, this study aims to determine whether current generation high radiometric resolution remote sensing data could be used accurately for land use/land cover classification in place of traditional moderate radiometric resolution multispectral data. A comparative study has been carried out to evaluate the utility of the simulated 12-bit LISS-III sensor compared with that of the original 7-bit LISS-III sensor for land use/land cover classification. It has been found that there is a small increase of 3% in overall accuracy by using the high radiometric resolution (12-bit) LISS-III data over the moderate radiometric resolution (7-bit) LISS-III data for land use/land cover classification. A 4-6% difference in classification accuracy of simulated LISS-III for cotton, chillies, black gram, and sugar cane indicates the usefulness of high radiometric resolution data for ground classes that are more heterogeneous within the same class. This study suggests that high radiometric resolution satellite data may not be a requisite for accurate land use/land cover mapping, whereas the spectral bandwidth, band placement, and method of classification parameters may be more important.

Evaluation of radiometric resolution on land use/land cover mapping in an agricultural area

. Forest fire has been regarded as one of the major reasons for the loss of biodiversity and dreadful conditions of environment. Global warming is also increasing the incidence of forest fire at an alarming rate. That’s why, one need to understand the complex biophysical parameters, which are responsible for this disaster. As it is difficult to predict forest fire, fire risk zone map can be useful for combating the forest fire. So the main aim of this study is to generate a Fire risk model to map fire risk zone using Remote Sensing & GIS technique. Pauri Garhwal District, located in Uttarakhand, India, has been selected for this study as it continually faces the problem of forest fire. Landsat-8 data of 18th April, 2016 have been used for land use land cover mapping. Slope and other information have been derived from topographic maps and field information. For thematic and topographic information analysis ArcGIS and ERDAS Imagine software have been used. Forest fire risk model was generated by using AHP method, where each category was assigned subjective weight according to their sensitivity to fire. Three categories of forest fire risk ranging from very high to low were derived. The generated forest fire risk model was found to be in strong agreement with actual fire-affected sites.

/pdf/modeling-of-parameters-for-forest-fire-risk-zone-mapping-3dz2nlvv1j.pdf

Modeling of parameters for forest fire risk zone mapping

We propose a simple and fast algorithm called PatchLift for computing distances between patches (contiguous block of samples) extracted from a given one-dimensional signal. PatchLift is based on the observation that the patch distances can be efficiently computed from a matrix that is derived from the one-dimensional signal using lifting; importantly, the number of operations required to compute the patch distances using this approach does not scale with the patch length. We next demonstrate how PatchLift can be used for patch-based denoising of images corrupted with Gaussian noise. In particular, we propose a separable formulation of the classical nonlocal means (NLM) algorithm that can be implemented using PatchLift. We demonstrate that the PatchLift-based implementation of separable NLM is a few orders faster than standard NLM and is competitive with existing fast implementations of NLM. Moreover, its denoising performance is shown to be consistently superior to that of NLM and some of its variants, both in terms of peak signal-to-noise ratio/structural similarity index and visual quality. (C) 2016 SPIE and IS&T

Sanjay Ghosh

Papers

Extraction of built-up areas using convolutional neural networks and transfer learning from sentinel-2 satellite images

Saliency Guided Image Detail Enhancement

Evaluation of radiometric resolution on land use/land cover mapping in an agricultural area

Modeling of parameters for forest fire risk zone mapping

Fast separable nonlocal means