Image classification for a large number of object categories

L'increment de bases de dades que cada vegada contenen imatges més difícils i amb un nombre més elevat de categories, està forçant el desenvolupament de tècniques de representació d'imatges que siguin discriminatives quan es vol treballar amb múltiples classes i d'algorismes que siguin eficients en l'aprenentatge i classificació. Aquesta tesi explora el problema de classificar les imatges segons l'objecte que contenen quan es disposa d'un gran nombre de categories. Primerament s'investiga com un sistema híbrid format per un model generatiu i un model discriminatiu pot beneficiar la tasca de classificació d'imatges on el nivell d'anotació humà sigui mínim. Per aquesta tasca introduïm un nou vocabulari utilitzant una representació densa de descriptors color-SIFT, i desprès s'investiga com els diferents paràmetres afecten la classificació final. Tot seguit es proposa un mètode par tal d'incorporar informació espacial amb el sistema híbrid, mostrant que la informació de context es de gran ajuda per la classificació d'imatges. Desprès introduïm un nou descriptor de forma que representa la imatge segons la seva forma local i la seva forma espacial, tot junt amb un kernel que incorpora aquesta informació espacial en forma piramidal. La forma es representada per un vector compacte obtenint un descriptor molt adequat per ésser utilitzat amb algorismes d'aprenentatge amb kernels. Els experiments realitzats postren que aquesta informació de forma te uns resultats semblants (i a vegades millors) als descriptors basats en aparença. També s'investiga com diferents característiques es poden combinar per ésser utilitzades en la classificació d'imatges i es mostra com el descriptor de forma proposat juntament amb un descriptor d'aparença millora substancialment la classificació. Finalment es descriu un algoritme que detecta les regions d'interès automàticament durant l'entrenament i la classificació. Això proporciona un mètode per inhibir el fons de la imatge i afegeix invariança a la posició dels objectes dins les imatges. S'ensenya que la forma i l'aparença sobre aquesta regió d'interès i utilitzant els classificadors random forests millora la classificació i el temps computacional. Es comparen els postres resultats amb resultats de la literatura utilitzant les mateixes bases de dades que els autors Aixa com els mateixos protocols d'aprenentatge i classificació. Es veu com totes les innovacions introduïdes incrementen la classificació final de les imatges.

Prognozowanie stężeń pyłku leszczyny, olszy i brzozy w powietrzu atmosferycznym na obszarze Polski

Wydział Nauk Geograficznych i Geologicznych: Instytut Geoekologii i Geoinformacji

Advanced feature selection methods in multinominal dementia classification from structural MRI data

Recent studies showed that features extracted from brain MRIs can well discriminate Alzheimer’s disease from Mild Cognitive Impairment. This study provides an algorithm that sequentially applies advanced feature selection methods for findings the best subset of features in terms of binary classification accuracy. The classifiers that provided the highest accuracies, have been then used for solving a multi-class problem by the one-versus-one strategy. Although several approaches based on Regions of Interest (ROIs) extraction exist, the prediction power of features has not yet investigated by comparing filter and wrapper techniques. The findings of this work suggest that (i) the IntraCranial Volume (ICV) normalization can lead to overfitting and worst the accuracy prediction of test set and (ii) the combined use of a Random Forest-based filter with a Support Vector Machines-based wrapper, improves accuracy of binary classification.

Speed prediction for triggering vehicle activated signs

Accurate speed prediction is a crucial step in the development of a dynamic vehcile activated sign (VAS). A previous study showed that the optimal trigger speed of such signs will need to be pre-determined according to the nature of the site and to the traffic conditions. The objective of this paper is to find an accurate predictive model based on historical traffic speed data to derive the optimal trigger speed for such signs. Adaptive neuro fuzzy (ANFIS), classification and regression tree (CART) and random forest (RF) were developed to predict one step ahead speed during all times of the day. The developed models were evaluated and compared to the results obtained from artificial neural network (ANN), multiple linear regression (MLR) and naïve prediction using traffic speed data collected at four sites located in Sweden. The data were aggregated into two periods, a short term period (5-min) and a long term period (1-hour). The results of this study showed that using RF is a promising method for predicting mean speed in the two proposed periods.. It is concluded that in terms of performance and computational complexity, a simplistic input features to the predicitive model gave a marked increase in the response time of the model whilse still delivering a low prediction error.

A data driven approach for automating vehicle activated signs

Vehicle activated signs (VAS) display a warning message when drivers exceed a particular threshold. VAS are often installed on local roads to display a warning message depending on the speed of the approaching vehicles. VAS are usually powered by electricity; however, battery and solar powered VAS are also commonplace. This thesis investigated devel-opment of an automatic trigger speed of vehicle activated signs in order to influence driver behaviour, the effect of which has been measured in terms of reduced mean speed and low standard deviation. A comprehen-sive understanding of the effectiveness of the trigger speed of the VAS on driver behaviour was established by systematically collecting data. Specif-ically, data on time of day, speed, length and direction of the vehicle have been collected for the purpose, using Doppler radar installed at the road. A data driven calibration method for the radar used in the experiment has also been developed and evaluated. Results indicate that trigger speed of the VAS had variable effect on driv-ers’ speed at different sites and at different times of the day. It is evident that the optimal trigger speed should be set near the 85th percentile speed, to be able to lower the standard deviation. In the case of battery and solar powered VAS, trigger speeds between the 50th and 85th per-centile offered the best compromise between safety and power consump-tion. Results also indicate that different classes of vehicles report differ-ences in mean speed and standard deviation; on a highway, the mean speed of cars differs slightly from the mean speed of trucks, whereas a significant difference was observed between the classes of vehicles on lo-cal roads. A differential trigger speed was therefore investigated for the sake of completion. A data driven approach using Random forest was found to be appropriate in predicting trigger speeds respective to types of vehicles and traffic conditions. The fact that the predicted trigger speed was found to be consistently around the 85th percentile speed justifies the choice of the automatic model.

Avaliação de compostos com actividade inibidora da glicoproteína-P

A resistência a múltiplos fármacos é um grande problema na terapia anti-cancerígena, sendo a glicoproteína-P (P-gp) uma das responsáveis por esta resistência. A realização deste trabalho incidiu principalmente no desenvolvimento de modelos matemáticos/estatísticos e “químicos”. Para os modelos matemáticos/estatísticos utilizamos métodos de Machine Learning como o Support Vector Machine (SVM) e o Random Forest, (RF) em relação aos modelos químicos utilizou-se farmacóforos. Os métodos acima mencionados foram aplicados a diversas proteínas P-gp, p53 e complexo p53-MDM2, utilizando duas famílias: as pifitrinas para a p53 e flavonóides para P-gp e, em menor medida, um grupo diversificado de moléculas de diversas famílias químicas. Nos modelos obtidos pelo SVM quando aplicados à P-gp e à família dos flavonóides, obtivemos bons valores através do kernel Radial Basis Function (RBF), com precisão de conjunto de treino de 94% e especificidade de 96%. Quanto ao conjunto de teste com previsão de 70% e especificidade de 67%, sendo que o número de falsos negativos foi o mais baixo comparativamente aos restantes kernels. Aplicando o RF à família dos flavonóides verificou-se que o conjunto de treino apresenta 86% de precisão e uma especificidade de 90%, quanto ao conjunto de teste obtivemos uma previsão de 70% e uma especificidade de 60%, existindo a particularidade de o número de falsos negativos ser o mais baixo. Repetindo o procedimento anterior (RF) e utilizando um total de 63 descritores, os resultados apresentaram valores inferiores obtendo-se para o conjunto de treino 79% de precisão e 82% de especificidade. Aplicando o modelo ao conjunto de teste obteve-se 70% de previsão e 60% de especificidade. Comparando os dois métodos, escolhemos o método SVM com o kernel RBF como modelo que nos garante os melhores resultados de classificação. Aplicamos o método SVM à P-gp e a um conjunto de moléculas não flavonóides que são transportados pela P-gp, obteve-se bons valores através do kernel RBF, com precisão de conjunto de treino de 95% e especificidade de 93%. Quanto ao conjunto de teste, obtivemos uma previsão de 70% e uma especificidade de 69%, existindo a particularidade de o número de falsos negativos ser o mais baixo. Aplicou-se o método do farmacóforo a três alvos, sendo estes, um conjunto de inibidores flavonóides e de substratos não flavonóides para a P-gp, um grupo de piftrinas para a p53 e um conjunto diversificado de estruturas para a ligação da p53-MDM2. Em cada um dos quatro modelos de farmacóforos obtidos identificou-se três características, sendo que as características referentes ao anel aromático e ao dador de ligações de hidrogénio estão presentes em todos os modelos obtidos. Realizando o rastreio em diversas bases de dados utilizando os modelos, obtivemos hits com uma grande diversidade estrutural.

A comparison of statistical methods for genomic selection in a mice population

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Determination of trace elements in bovine semen samples by inductively coupled plasma mass spectrometry and data mining techniques for identification of bovine class

The reproductive performance of cattle may be influenced by several factors, but mineral imbalances are crucial in terms of direct effects on reproduction. Several studies have shown that elements such as calcium, copper, iron, magnesium, selenium, and zinc are essential for reproduction and can prevent oxidative stress. However, toxic elements such as lead, nickel, and arsenic can have adverse effects on reproduction. In this paper, we applied a simple and fast method of multi-element analysis to bovine semen samples from Zebu and European classes used in reproduction programs and artificial insemination. Samples were analyzed by inductively coupled plasma spectrometry (ICP-MS) using aqueous medium calibration and the samples were diluted in a proportion of 1:50 in a solution containing 0.01% (vol/vol) Triton X-100 and 0.5% (vol/vol) nitric acid. Rhodium, iridium, and yttrium were used as the internal standards for ICP-MS analysis. To develop a reliable method of tracing the class of bovine semen, we used data mining techniques that make it possible to classify unknown samples after checking the differentiation of known-class samples. Based on the determination of 15 elements in 41 samples of bovine semen, 3 machine-learning tools for classification were applied to determine cattle class. Our results demonstrate the potential of support vector machine (SVM), multilayer perceptron (MLP), and random forest (RF) chemometric tools to identify cattle class. Moreover, the selection tools made it possible to reduce the number of chemical elements needed from 15 to just 8.

Multi-element determination in Brazilian honey samples by inductively coupled plasma mass spectrometry and estimation of geographic origin with data mining techniques

Multi-element analysis of honey samples was carried out with the aim of developing a reliable method of tracing the origin of honey. Forty-two chemical elements were determined (Al, Cu, Pb, Zn, Mn, Cd, Tl, Co, Ni, Rb, Ba, Be, Bi, U, V, Fe, Pt, Pd, Te, Hf, Mo, Sn, Sb, P, La, Mg, I, Sm, Tb, Dy, Sd, Th, Pr, Nd, Tm, Yb, Lu, Gd, Ho, Er, Ce, Cr) by inductively coupled plasma mass spectrometry (ICP-MS). Then, three machine learning tools for classification and two for attribute selection were applied in order to prove that it is possible to use data mining tools to find the region where honey originated. Our results clearly demonstrate the potential of Support Vector Machine (SVM), Multilayer Perceptron (MLP) and Random Forest (RF) chemometric tools for honey origin identification. Moreover, the selection tools allowed a reduction from 42 trace element concentrations to only 5. (C) 2012 Elsevier Ltd. All rights reserved.

Identification of protein expression signatures in gastric carcinomas using clustering analysis

Background and Aim: The identification of gastric carcinomas (GC) has traditionally been based on histomorphology. Recently, DNA microarrays have successfully been used to identify tumors through clustering of the expression profiles. Random forest clustering is widely used for tissue microarrays and other immunohistochemical data, because it handles highly-skewed tumor marker expressions well, and weighs the contribution of each marker according to its relatedness with other tumor markers. In the present study, we e identified biologically- and clinically-meaningful groups of GC by hierarchical clustering analysis of immunohistochemical protein expression. Methods: We selected 28 proteins (p16, p27, p21, cyclin D1, cyclin A, cyclin B1, pRb, p53, c-met, c-erbB-2, vascular endothelial growth factor, transforming growth factor [TGF]-beta I, TGF-beta II, MutS homolog-2, bcl-2, bax, bak, bcl-x, adenomatous polyposis coli, clathrin, E-cadherin, beta-catenin, mucin (MUC) 1, MUC2, MUC5AC, MUC6, matrix metalloproteinase [ MMP]-2, and MMP-9) to be investigated by immunohistochemistry in 482 GC. The analyses of the data were done using a random forest-clustering method. Results: Proteins related to cell cycle, growth factor, cell motility, cell adhesion, apoptosis, and matrix remodeling were highly expressed in GC. We identified protein expressions associated with poor survival in diffuse-type GC. Conclusions: Based on the expression analysis of 28 proteins, we identified two groups of GC that could not be explained by any clinicopathological variables, and a subgroup of long-surviving diffuse-type GC patients with a distinct molecular profile. These results provide not only a new molecular basis for understanding the biological properties of GC, but also better prediction of survival than the classic pathological grouping.

Evaluation of LBP and HOG descriptors for clothing attribute description

[EN]In this work an experimental study about the capability of the LBP, HOG descriptors and color for clothing attribute classification is presented. Two different variants of the LBP descriptor are considered, the original LBP and the uniform LBP. Two classifiers, Linear SVM and Random Forest, have been included in the comparison because they have been frequently used in clothing attributes classification. The experiments are carried out with a public available dataset, the clothing attribute dataset, that has 26 attributes in total. The obtained accuracies are over 75% in most cases, reaching 80% for the necktie or sleeve length attributes.

Un sensore multi-purpose: l'accelerometro

questa tesi propone una prospettiva completa sull'accelerometro, descrivendone le caratteristiche le tipologie, gli utilizzi che se ne possono fare e gli ambiti di applicazione. Sviluppa anche argomenti che riguardano interfacce e piattaforme per l'interazione con il sensore e algoritmi machine learning.

Metodologie di apprendimento automatico applicate alla generazione di dati 3D

Il framework in oggetto, è un ambiente ideato con lo scopo di applicare tecniche di Machine Learning (in particolare le Random Forest) alle funzionalità dell'algoritmo di stereo matching SGM (Semi Global Matching), al fine di incrementarne l'accuratezza in versione standard. Scopo della presente tesi è quello di modificare alcune impostazioni di tale framework rendendolo un ambiente che meglio si adatti alla direzionalità delle scanline (introducendo finestre di supporto rettangolari e ortogonali e il training di foreste separate in base alla singola scanline) e ampliarne le funzionalità tramite l'aggiunta di alcune nuove feature, quali la distanza dal più vicino edge direzionale e la distintività calcolate sulle immagini Left della stereo pair e gli edge direzionali sulle mappe di disparità. Il fine ultimo sarà quello di eseguire svariati test sui dataset Middlebury 2014 e KITTI e raccogliere dati che descrivano l'andamento in positivo o negativo delle modifiche effettuate.

Tumor classification of six common cancer types based on proteomic profiling by MALDI imaging

In clinical diagnostics, it is of outmost importance to correctly identify the source of a metastatic tumor, especially if no apparent primary tumor is present. Tissue-based proteomics might allow correct tumor classification. As a result, we performed MALDI imaging to generate proteomic signatures for different tumors. These signatures were used to classify common cancer types. At first, a cohort comprised of tissue samples from six adenocarcinoma entities located at different organ sites (esophagus, breast, colon, liver, stomach, thyroid gland, n = 171) was classified using two algorithms for a training and test set. For the test set, Support Vector Machine and Random Forest yielded overall accuracies of 82.74 and 81.18%, respectively. Then, colon cancer liver metastasis samples (n = 19) were introduced into the classification. The liver metastasis samples could be discriminated with high accuracy from primary tumors of colon cancer and hepatocellular carcinoma. Additionally, colon cancer liver metastasis samples could be successfully classified by using colon cancer primary tumor samples for the training of the classifier. These findings demonstrate that MALDI imaging-derived proteomic classifiers can discriminate between different tumor types at different organ sites and in the same site.

Application and comparison of classification algorithms for recognition of Alzheimer's disease in electrical brain activity (EEG)

The early detection of subjects with probable Alzheimer's disease (AD) is crucial for effective appliance of treatment strategies. Here we explored the ability of a multitude of linear and non-linear classification algorithms to discriminate between the electroencephalograms (EEGs) of patients with varying degree of AD and their age-matched control subjects. Absolute and relative spectral power, distribution of spectral power, and measures of spatial synchronization were calculated from recordings of resting eyes-closed continuous EEGs of 45 healthy controls, 116 patients with mild AD and 81 patients with moderate AD, recruited in two different centers (Stockholm, New York). The applied classification algorithms were: principal component linear discriminant analysis (PC LDA), partial least squares LDA (PLS LDA), principal component logistic regression (PC LR), partial least squares logistic regression (PLS LR), bagging, random forest, support vector machines (SVM) and feed-forward neural network. Based on 10-fold cross-validation runs it could be demonstrated that even tough modern computer-intensive classification algorithms such as random forests, SVM and neural networks show a slight superiority, more classical classification algorithms performed nearly equally well. Using random forests classification a considerable sensitivity of up to 85% and a specificity of 78%, respectively for the test of even only mild AD patients has been reached, whereas for the comparison of moderate AD vs. controls, using SVM and neural networks, values of 89% and 88% for sensitivity and specificity were achieved. Such a remarkable performance proves the value of these classification algorithms for clinical diagnostics.