Combination Of Descriptors Research Articles

Automatic Land Cover Reconstruction From Historical Aerial Images: An Evaluation of Features Extraction and Classification Algorithms.

The land cover reconstruction from monochromatic historical aerial images is a challenging task that has recently attracted an increasing interest from the scientific community with the proliferation of large-scale epidemiological studies involving retrospective analysis of spatial patterns. However, the efforts made by the computer vision community in remote-sensing applications are mostly focused on prospective approaches through the analysis of high-resolution multi-spectral data acquired by the advanced spatial programs. Hence, four contributions are proposed in this paper. They aim at providing a comparison basis for the future development of computer vision algorithms applied to the automation of the land cover reconstruction from monochromatic historical aerial images. First, a new multi-scale multi-date dataset composed of 4.9 million non-overlapping annotated patches of the France territory between 1970 and 1990 has been created with the help of geography experts. This dataset has been named HistAerial. Second, an extensive comparison study of the state-of-the-art texture features extraction and classification algorithms, including deep convolutional neural networks (DCNNs), has been performed. It is presented in the form of an evaluation. Third, a novel low-dimensional local texture filter named rotated-corner local binary pattern (R-CRLBP) is presented as a simplification of the binary gradient contours filter through the use of an orthogonal combination representation. Finally, a novel combination of low-dimensional texture descriptors, including the R-CRLBP filter, is introduced as a light combination of local binary patterns (LCoLBPs). The LCoLBP filter achieved state-of-the-art results on the HistAerial dataset while conserving a relatively low-dimensional feature vector space compared with the DCNN approaches (17 times shorter).

Toxicological and pharmacologic effects of farnesol (C15H26O): A descriptive systematic review

The objective of the present study was to perform a systematic review (SR) composed of preclinical and clinical studies which investigated the toxicological and pharmacologic effects of farnesol [Molecular formula: C15H26O; IUPAC: (3,7,11-Trimethyl-2,6,10-dodecatrien-1-ol]. This SR was performed according to PRISMA guidelines. Literature research was performed using PubMed, MEDLINE, Scopus and Web of Science databases using the descriptor combinations: “farnesol and pharmacological effect” and “farnesol and toxicology”. The inclusion criteria used were original articles from preclinical and clinical studies investigating the pharmacological and toxicological effects of farnesol, published between January 1960 and December 2017 which were written in English, Portuguese and Spanish. Primary research identified 414 articles, from which 76 articles were selected for final analysis following the inclusion criteria. After grouping, 51.32 and 22.37% of the articles investigated the antimicrobial and antitumor effect, respectively. Methodological biases have been observed both in pre-clinical studies with non-human animals and in clinical trials, mainly in group allocation and blinding. This SR is the first study developed to compile the studies concerning the pharmacological and toxicological effects of farnesol. This study concludes that farnesol possesses different pharmacological and toxicological features, which permit its use as an active or a coadjuvant drug.

RETRACTED ARTICLE: Ensemble of texture and shape descriptors using support vector machine classification for face recognition

One of the significant task in pattern recognition and computer vision along with artificial intelligence and machine learning is the Face Recognition. Most of the prevailing approaches on face recognition concentrates on the recognition of the utmost appropriate facial attributes for efficiently recognizing and differentiating amongst the considered images. In this paper, an ensemble aided facial recognition approach is suggested that performs well in wild environment using an ensemble of feature descriptors and preprocessing approaches. The combination of texture and color descriptors are mined from the preprocessed facial images and classified using support vector machine algorithm. The experimental outcome of the suggested methodology is illustrated using two databases such as FERET data samples and Labeled Faces in the Wild data samples. From the results, it is shown that, the proposed approach has good classification accuracy and combination utility of pre-processing techniques due to the usage of additional preprocessing and extracted feature descriptors. The average classification accuracies for the both the data samples are 99% and 94% respectively.

Substituted naphthalene reaction rates with peroxy-acid treatment: prediction of reactivity using PEST

ABSTRACTPersistent organic contaminants in the environment pose an environmental risk due to widespread occurrence and toxic properties. Advanced oxidation processes (AOPs) are treatment methods that have been used to successfully degrade organic contaminants in water, soil, sediments and sludge. Reaction rate constants (k) for peroxy acid treatment of 10 substituted naphthalene compounds were determined. The treatment method utilized acetic acid, hydrogen peroxide and a sulphuric acid catalyst to degrade the polyaromatic structures found in the compounds. Molecular structures of the selected compounds were derived at the B3LYP/6-31G* level of theory. Property-encoded surface translator (PEST) descriptors were calculated from B3LYP/6-31G* optimized structures and were found to have significant levels of correlation with k. Models using minimum local ionization potential (PIP.MIN) and a histogram [bin] of the gradient of the K electronic kinetic energy normal to the isosurface (DKN) were evaluated and found to agree within 10% of experimentally derived values of k in most instances. Results show that a combination of PEST descriptors could be used to predict reactivity by the peroxy-acid process. The PEST technology could prove to be a valuable asset for effective remediation design by predicting reaction outcomes for substituted naphthalene compounds and possibly other hydrophobic organic compounds (HOCs).

Machine Learning Prediction of H Adsorption Energies on Ag Alloys.

Adsorption energies on surfaces are excellent descriptors of their chemical properties, including their catalytic performance. High-throughput adsorption energy predictions can therefore help accelerate first-principles catalyst design. To this end, we present over 5000 DFT calculations of H adsorption energies on dilute Ag alloys and describe a general machine learning approach to rapidly predict H adsorption energies for new Ag alloy structures. We find that random forests provide accurate predictions and that the best features are combinations of traditional chemical and structural descriptors. Further analysis of our model errors and the underlying forest kernel reveals unexpected finite-size electronic structure effects: embedded dopant atoms can display counterintuitive behavior such as nonmonotonic trends as a function of composition and high sensitivity to dopants far from the adsorbing H atom. We explain these behaviors with simple tight-binding Hamiltonians and d-orbital densities of states. We also use variations among forest leaves to predict the uncertainty of predictions, which allows us to mitigate the effects of larger errors.

Recognition of Fingerspelling Sequences in Polish Sign Language Using Point Clouds Obtained from Depth Images.

The paper presents a method for recognizing sequences of static letters of the Polish finger alphabet using the point cloud descriptors: viewpoint feature histogram, eigenvalues-based descriptors, ensemble of shape functions, and global radius-based surface descriptor. Each sequence is understood as quick highly coarticulated motions, and the classification is performed by networks of hidden Markov models trained by transitions between postures corresponding to particular letters. Three kinds of the left-to-right Markov models of the transitions, two networks of the transition models—independent and dependent on a dictionary—as well as various combinations of point cloud descriptors are examined on a publicly available dataset of 4200 executions (registered as depth map sequences) prepared by the authors. The hand shape representation proposed in our method can also be applied for recognition of hand postures in single frames. We confirmed this using a known, challenging American finger alphabet dataset with about 60,000 depth images.

Interpretation of QSAR Models by Coloring Atoms According to Changes in Predicted Activity: How Robust Is It?

Most chemists would agree that the ability to interpret a quantitative structure-activity relationship (QSAR) model is as important as the ability of the model to make accurate predictions. One type of interpretation is coloration of atoms in molecules according to the contribution of each atom to the predicted activity, as in "heat maps". The ability to determine which parts of a molecule increase the activity in question and which decrease it should be useful to chemists who want to modify the molecule. For that type of application, we would hope the coloration to not be particularly sensitive to the details of model building. In this Article, we examine a number of aspects of coloration against 20 combinations of descriptors and QSAR methods. We demonstrate that atom-level coloration is much less robust to descriptor/method combinations than cross-validated predictions. Even in ideal cases where the contribution of individual atoms is known, we cannot always recover the important atoms for some descriptor/method combinations. Thus, model interpretation by atom coloration may not be as simple as it first appeared.

Thermodynamic solubility modelling, solvent effect and preferential solvation of naftopidil in aqueous co-solvent solutions of (n-propanol, ethanol, isopropanol and dimethyl sulfoxide)

The naftopidil solubility in aqueous co-solvent mixtures of n-propanol, ethanol, isopropanol and dimethyl sulfoxide (DMSO) at temperatures from 278.15 K to 318.15 K was reported. At fixed composition of n-propanol, ethanol, isopropanol or DMSO and temperature, the mole fraction solubility of naftopidil was highest in DMSO (1) + water (2) mixture, and lowest in isopropanol (1) + water (2) mixture. The Jouyban-Acree model and van’t Hoff-Jouyban-Acree model provided well correlation results obtaining relative average deviation lower than 2.85% and root-mean-square deviation lower than 8.89 × 10−4. Furthermore, the method of linear solvation energy relationships was performed with a suitable combination of solvent polarity descriptors to elucidate the nature of intermolecular interactions giving rise to the solubility variation in co-solvent mixtures. Quantitative values for the local mole fraction of n-propanol (ethanol, isopropanol or DMSO) and water around the naftopidil were computed by using the Inverse Kirkwood–Buff integrals method applied to the determined solubility data. Naftopidil was preferentially solvated by water for the four solvent mixtures in water-rich compositions; while in intermediate and co-solvent-rich compositions for n-propanol (isopropanol and ethanol) (1) + water (2) mixtures, naftopidil was preferentially solvated by the co-solvent. The preferential solvation magnitude of naftopidil was highest in isopropanol mixtures than in the other three co-solvent mixtures.

Selective multi-descriptor fusion for face identification

Over the last 2 decades, face identification has been an active field of research in computer vision. As an important class of image representation methods for face identification, fused descriptor-based methods are known to lack sufficient discriminant information, especially when compared with deep learning-based methods. This paper presents a new face representation method, multi-descriptor fusion (MDF), which represents face images through a combination of multiple descriptors, resulting in hyper-high dimensional fused descriptor features. MDF enables excellent performance in face identification, exceeding the state-of-the-art, but it comes with high memory and computational costs. As a solution to the high cost problem, this paper also presents an optimisation method, discriminant ability-based multi-descriptor selection (DAMS), to select a subset of descriptors from the set of 65 initial descriptors whilst maximising the discriminant ability. The MDF face representation, after being refined by DAMS, is named selective multi-descriptor fusion (SMDF). Compared with MDF, SMDF has much smaller feature dimension and is thus usable on an ordinary PC, but still has similar performance. Various experiments are conducted on the CAS-PEAL-R1 and LFW datasets to demonstrate the performance of the proposed methods.

Predictive Computational Tools for Assessment of Ecotoxicological Activity of Organic Micropollutants in Various Water Sources in Brazil.

The increase of chemical pollutants detected in different aquatic environments over the past few years has been in the focus of several studies related to their occurrence, transport, fate, and hazards, or risks to human and environmental health. In Brazil, recent studies have been conducted on the occurrence of a series of organic micropollutants (OMPs) in aquatic environments. Nevertheless, the toxicological information and environmental behavior for most of these pollutants are still difficult to evaluate. We obtained consensus models using a combination of VolSurf and Dragon descriptors and Random Forest approach to predict the toxicity against Daphnia magna and Pimephales promelas for compounds identified in different sources of water in Brazil. The models showed satisfactory performance when compared with other approaches, such as Toxicity Estimation Software Tool (T.E.S.T.), etc. Both models can be used as complementary tools to aid in the prediction of organic compounds against both organisms.

Quantum-mechanical LSERs for the concentration-dependent adsorption of aromatic organic compounds by activated carbon: Applications and comparison with carbon nanotubes

ABSTRACTCarbon nanotubes (CNTs) have taken precedence over activated carbon in various applications where adsorption is the primary process. The adsorption of chemical compounds by CNTs and activated carbon is most often predicted through linear free energy/solvation energy relationships (LFERs/LSERs). This work proposes quantum-mechanical LSER models based on a combination of quantum-mechanical descriptors and solvatochromic descriptors of LSERs for predicting the adsorption of aromatic organic compounds by activated carbon at varying adsorbate concentrations. The models are validated using state-of-the-art procedures employing an external prediction set of compounds. This work reveals the hydrogen bond donating and accepting ability of compounds to be the most influencing – but a negative – factor in the adsorption process of activated carbon. The quantum-mechanical LSERs proposed in this work are analysed and found to be equally reliable as the existing LSERs. These were further used to predict the adsorption of nucleobases, steroid hormones, agrochemicals, endocrine disruptors and pharmaceutical drugs. Notably, agrochemicals and endocrine disruptors are predicted to be adsorbed more strongly by activated carbon when compared with their adsorption by CNTs. However, quantum-mechanical LSERs predict the adsorption strength of biomolecules on activated carbon to be similar to that on the CNTs, which can be used to assess the risk associated with using carbon materials.

Equilibrium solubility investigation and thermodynamic aspects of biologically active gimeracil (form P) dissolved in aqueous co-solvent mixtures of isopropanol, N,N-dimethylformamide, ethylene glycol and dimethylsulfoxide

Solubilities of gimeracil (form P) in aqueous co-solvent mixtures of isopropanol, N,N-dimethylformamide (DMF), ethylene glycol (EG) and dimethylsulfoxide (DMSO) were investigated via the isothermal dissolution equilibrium method at (283.15–328.15) K under ambient pressure p = 101.2 kPa. Experimental solubility was increased with increasing temperature and mass fraction of each co-solvent. The largest solubility was found in neat co-solvents. The solids equilibrated with liquid phase were characterized by X-ray power diffraction, indicating no polymorphic transformation, solvate formation or crystal transition according to the spectral data. The Jouyban-Acree model was adopted to correlate the obtained solubility. The highest RAD and RMSD values were, respectively, 2.61 × 10−2 and 14.96 × 10−4. Quantitative values for the local mole fraction of DMF (EG, DMSO or isopropanol) and water around gimeracil (form P) were acquired by using the Inverse Kirkwood–Buff integrals method. The preferential solvation parameters for isopropanol were positive in the isopropanol mixtures in intermediate and isopropanol-rich compositions, which indicated that gimeracil (form P) was preferentially solvated by isopropanol. Gimeracil (form P) could act mainly as a Lewis acid interacting with proton-acceptor functional group of isopropanol. Within the same region, gimeracil (form P) was not preferentially solvated by DMF, EG and DMSO. Furthermore, the method of linear solvation energy relationships was performed with a suitable combination of solvent polarity descriptors to explain the nature of intermolecular interactions resulting in the solubility variation in the co-solvent mixtures.

Mechanistic Reactivity Descriptors for the Prediction of Ames Mutagenicity of Primary Aromatic Amines.

Pharmaceutical products are often synthesized by the use of reactive starting materials and intermediates. These can, either as impurities or through metabolic activation, bind to the DNA. Primary aromatic amines belong to the critical classes that are considered potentially mutagenic in the Ames test, so there is a great need for good prediction models for risk assessment. How primary aromatic amines exert their mutagenic potential can be rationalized by the widely accepted nitrenium ion hypothesis of covalent binding to the DNA of reactive electrophiles formed out of the aromatic amines. Since the reactive chemical species is different in chemical structure from the actual compound, it is difficult to achieve good predictions via classical descriptor or fingerprint-based machine learning. In this approach, we use a combination of different molecular and atomic descriptors that is able to describe different mechanistic aspects of the metabolic transformation leading from the primary aromatic amine to the reactive metabolite that binds to the DNA. Applied to a test set, the combination shows significantly better performance than models that only use one of these descriptors and complemented the general internal Ames mutagenicity prediction model at Bayer.

Analysis of softwares for emotion recognition in children and teenagers with autism spectrum disorder

ABSTRACT Purpose: to investigate the use of softwares for emotion recognition in children and teenagers with Autism Spectrum Disorders (ASD). Methods: an integrative review of the literature with scientific papers published from 2012 to 2017 indexed in Periodico Capes, Science Direct, and PudMed; combined descriptors: autism AND emotion AND software; autism AND emotion recognition AND software. Inclusion criterion was the use of software related to emotion recognition in children and teenagers with ASD, up to 18 years old. Review papers and those using robots were excluded. Results: ten international papers were reviewed. The most used emotional expressions were “happiness”, “fear”, “anger”, “disgust”, “sadness”, and “surprise”. Ten software programs were described: Emotion Recognition Task (1), Cambridge Mindreading Face-Voice Battery for Children (3), Mind Reading (2), Mood Maker (1), Virtual-Reality Emotion Sensitivity Test (1), FaceSay (1), Penn Emotion Recognition (1), FaceMaze Game (1), Computer Emotion Recognition Toolbox (CERT) (1), and Emotiplay (1). Conclusion: studies with software programs focused on ASD intervention allow future research efforts in the diagnosis and intervention of this disorder.

An efficient similarity search using a combination between descriptors: a case of study in human face recognition

Face recognition is one of the important fields of search in computer vision. Its principle consists to look for images that represent the similar faces to a given face image the image request. This process is done by extracting a set of features of the request image then making comparison between features generated by the request one and the others extracted from whole face image database. Recently, numerous face representation and classification methods have been proposed in the literature. Nevertheless, many issues related to indexing, combination of adequate descriptors and time computing have not yet been solved. In this paper, we deal with problems related to features combination and this, by conceiving a preformat content-based image retrieval that is mainly oriented to handle face authentication challenges. Its convivial interface allows to user the selection of appropriate weighting coefficient values associate to each feature based on human judgment in order to enhance the retrieval performance. We have tested our proposed method on ORL database by using a set of known features. The obtained results show the performance of our proposed method.

Brachiopods classification based on fusion of contour and region based descriptors

In this paper, we propose a contour-region based shape descriptor for Brachiopods classification by using a combinations of Fourier descriptors and R-transform extracted from Radon transform. Fourier descriptors is supported by the well-developed and well-understood Fourier theory and are a powerful features for the recognition of two-dimensional connected shapes. We used the stable and complete Fourier descriptors proposed by Ghorbel to present the contour information. To depict shape interior content we used the R-transform. It is advantages lies in its low computational complexity and geometric invariance. We compared the proposed descriptor with curvature scale space, R-transform and Ghorbel descriptors using city block distance measure and our Brachiopods database. We present the experimental results to reveal the performance of the proposed descriptor which is independent on the starting points and efficient.

Combining Features of Metal Oxide Nanoparticles

A poor applicability of classic 2D descriptors for representation of metal oxide nanoparticles is briefly discussed. The combination of 1D descriptors with previously calculated size-dependent descriptors is utilized to represent the structural features of nanoparticles in QSAR modeling. For this purpose, descriptors based on the fundamental characteristics of atoms (nuclear charge, oxidation level, electronegativity, ionic radius, ionic refraction etc.) were combined with those derived from structural formula (doublets –A2, AB, …; triplets – A3, A2B, ABC, … etc.) and “liquid drop model” derived size-dependent parameters. Nano-QSAR models are developed for cytotoxicity of metal oxide nanoparticles against E. coli and HaCaT cells. Two developed nano-QSAR models are discussed in terms of cluster analysis.

Brachiopods classification based on fusion of contour and region based descriptors

In this paper, we propose a contour-region based shape descriptor for Brachiopods classification by using a combinations of Fourier descriptors and R-transform extracted from Radon transform. Fouri...

A new method of combining colour, texture and shape features using the genetic algorithm for image retrieval

Semi-automatic or automatic image indexation emerged because manual image indexation is slow and tedious. Generally, this first indexation is used as part of a content-based image retrieval system (CBIR). To have a powerful CBIR system, it is necessary to be concerned with three main facets: 1) the choice of the descriptors (based on shape, colour and texture and/or a combination between them); 2) the process of indexation and finally; 3) the retrieval process. In this work, we focus mainly on an indexing based on genetic algorithm and particle swarm optimisation (PSO) algorithm. We chose an optimal combination of colour, shape and texture (PCM: powerful combination method) descriptors. The fruit of our research work is implemented in a system called image search engine (ISE) which showed a very promising performance. In fact, the performance evaluation of the PCM method of our descriptors combination showed upgrades of the average precision metric from 66.6% to 89.30% for the 'food' category colour histogram, from 77.7% to 100% concerning CCV for the 'flower' category, and from 44.4% to 87.65% concerning the co-occurrence matrix for the 'building' category using the Corel dataset. Likewise, our ISE system showed much more interesting performance compared to what was shown in previous works.

Methods of assessing body composition in people with limb amputation

Objective: Identify which methods of have been used to assess the body composition individuals with limb amputation from the literature review. Method: In order to reach the defined objective, the Pubmed and Scielo databases were used, through the descriptors (in combination with amputation) body composition, fat mass and lean mass from 1986 to 2019. Results: Found articles went through the inclusion and exclusion criteria and 17 articles were selected, all of them using anthropometric measures (body mass and stature), followed by bone densitometry (DXA) and thickness of skinfolds. It can also be observed that most of the studies are focused on the health aspect which focused on nutrition and risk factors. Conclusion: Through this review we observed that the anthropometric methods were mostly used to evaluate body composition in individuals with limb amputation followed by bone densitometry. However, the studies were scarce and many procedures were not so clear to make a more careful analysis unfeasible. In addition, there is a lack of longitudinal studies and especially in the area of sports performance in this population, indicating an important gap to be explored.