977 resultados para Classification tree
Resumo:
Abstract
Resumo:
Abstract
Resumo:
This letter presents advanced classification methods for very high resolution images. Efficient multisource information, both spectral and spatial, is exploited through the use of composite kernels in support vector machines. Weighted summations of kernels accounting for separate sources of spectral and spatial information are analyzed and compared to classical approaches such as pure spectral classification or stacked approaches using all the features in a single vector. Model selection problems are addressed, as well as the importance of the different kernels in the weighted summation.
Resumo:
The glasses of the rosette forming the main window of the transept of the Gothic Cathedral of Tarragona have been characterised by means of SEM/EDS, XRD, FTIR and electronic microprobe. The multivariate statistical treatment of these data allow to establish a classification of the samples forming groups having an historical significance and reflecting ancient restorations. Furthermore, the decay patterns and mechanisms have been determined and the weathering by-products characterised. It has been demonstrated a clear influence of the bioactivity in the decay of these glasses, which activity is partially controlled by the chemical composition of the glasses.
Resumo:
The glasses of the rosette forming the main window of the transept of the Gothic Cathedral of Tarragona have been characterised by means of SEM/EDS, XRD, FTIR and electronic microprobe. The multivariate statistical treatment of these data allow to establish a classification of the samples forming groups having an historical significance and reflecting ancient restorations. Furthermore, the decay patterns and mechanisms have been determined and the weathering by-products characterised. It has been demonstrated a clear influence of the bioactivity in the decay of these glasses, which activity is partially controlled by the chemical composition of the glasses.
Resumo:
The objective of this work was to estimate the genetic variability and divergence among 22 superior rubber tree (Hevea sp.) genotypes of the IAC 400 series. Univariate and multivariate analyses were performed using eight quantitative traits (descriptors), including yield. In the univariate analyses, the estimated parameters were: genetic and environmental variances; genetic and environmental coefficients of variation; and the variation index. The Mahalanobis generalized distance, the Tocher agglomerative method and canonical variables were used for the multivariate analyses. In the univariate analyses, variability was verified among the genotypes for all the variables evaluated. The Tocher method grouped the genotypes into 11 clusters of dissimilarity. The first four canonical variables explained 87.93% of the cumulative variation. The highest genetic variability was found in rubber yield-related traits, which contributed the most to the genetic divergence. The most divergent pairs of genotypes are suggested for crossbreeding. The genotypes evaluated are suitable for breeding and may be used to continue the IAC rubber tree breeding program.
Resumo:
We answer the following question: given any n∈ℕ, which is the minimum number of endpoints en of a tree admitting a zero-entropy map f with a periodic orbit of period n? We prove that en=s1s2…sk−∑i=2ksisi+1…sk, where n=s1s2…sk is the decomposition of n into a product of primes such that si≤si+1 for 1≤i
Resumo:
University of Helsinki, Faculty of Agriculture and Forestry, Department of Forest Resource Management
Resumo:
The objective of this work was to study the influence of cyanogenesis on the onset of irreversible tapping panel dryness (TPD) and the physiological and histological aspects of secondary phloem in the trunk (tapping panel) of rubber trees (Hevea spp.). Two cyanogenic compounds, linamarin and KCN, were applied separately on the trunk bark of healthy mature trees belonging to two Brazilian clones (Fx 4098 and Fx 3899). Changes in histology, latex pressure potential (ΨP) and cyanogenic potential (HCNp) were followed in the trunk inner barks. In addition, the HCNp levels were determined in TPD-affected plants of both clones. The applications of linamarin or KCN in healthy plants decreased latex ΨP, and formed tylosoids associated with in situ coagulation of latex. The clone Fx 4098 had the higher HCNp and showed the quicker and stronger responses to the cyanogenic compounds. Plants with TPD syntoms had a higher HCNp than the untreated healthy ones. Since histological changes are also structural markers of early TPD, it can be inferred that excessive release of cyanide can induce it in sensitive rubber clones
Resumo:
This manual provides a set of procedural rules and regulations for use in functionally classifying all roads and streets in Iowa according to the character of service they are intended to provide. Functional classification is a requirement of House File 394 (Functional Highway Classification Bill) enacted by the 63rd General Assembly of the Iowa Legislature. Functional classification is defined in this Bill as: "The grouping of roads and streets into systems according to the character of service they will be expected to provide, and the assignment of jurisdiction over each class to the governmental unit having primary interest in each type of service."
Resumo:
This manual provides a set of procedural rules and regulations for use in functionally classifying all roads and streets in Iowa according to the character of service they are intended to provide. Functional classification is a requirement of the 1973 Code of Iowa (Chapter 306) as amended by Senate File 1062 enacted by the 2nd session of the 65th General Assembly of Iowa. Functional classification is defined as the grouping of roads and streets into systems according to the character of service they will be expected to provide, and the assignment of jurisdiction over each class to the governmental unit having primary interest in each type of service. Stated objectives of the legislation are: "Functional classification will serve the legislator by providing an equitable basis for determination of proper source of tax support and providing for the assignment of financial resources to the governmental unit having responsibility for each class of service. Functional classification promotes the ability of the administrator to effectively prepare and carry out long range programs which reflect the transportation needs of the public." All roads and streets in legal existence will be classified. Instructions are also included in this manual for a continuous reporting to the Highway Commission of changes in classification and/or jurisdiction resulting from new construction, corporation line changes, relocations, and deletions. This continuous updating of records is absolutely essential for modern day transportation planning as it is the only possible way to monitor the status of existing road systems, and consequently determine adequacy and needs with accuracy.
Resumo:
The objective of this work was to assess the performance of panel clones under crowns resistant to South American leaf blight (Microcyclus ulei). The experiment was carried out with 18 panel clones crown-budded with Hevea pauciflora x H. guianensis, in a Xanthic Ferralsol (Oxisol) in Manaus, AM, Brazil. The following parameters were evaluated: dry rubber yield, plant nutritional status, and anatomical and physiological characteristics of the latex vessels. In the first three years of evaluation, the panel clones IAN 2878, IAN 2903, CNS AM 7905, CNS AM 7905 P1, and PB 28/59 showed the highest dry rubber yield potential, while the clones IAN 6158, IAN 6590, and IAN 6515 should not be recommended for crown budding. Higher potassium and copper foliar content in panel clones were associated to an increase in dry rubber yield. The simultaneous evaluation of anatomical and physiological characteristics of latex is fundamental for the selection of panel clones in the Amazon region. Crown budding is an efficient technology for South American leaf blight management in endemic regions.
Resumo:
The objective of this work was to assess and characterize two clones, 169 and 685, of Cabernet Sauvignon grapes and to evaluate the wine produced from these grapes. The experiment was carried out in São Joaquim, SC, Brazil, during the 2009 harvest season. During grape ripening, the evolution of physical-chemical properties, phenolic compounds, organic acids, and anthocyanins was evaluated. During grape harvest, yield components were determined for each clone. Individual and total phenolics, individual and total anthocyanins, and antioxidant activity were evaluated for wine. The clones were also assessed regarding the duration of their phenological cycle. During ripening, the evolution of phenolic compounds and of physical-chemical parameters was similar for both clones; however, during harvest, significant differences were observed regarding yield, number of bunches per plant and berries per bunch, leaf area, and organic acid, polyphenol, and anthocyanin content. The wines produced from these clones showed significant differences regarding chemical composition. The clones showed similar phenological cycle and responses to bioclimatic parameters. Principal component analysis shows that clone 685 is strongly correlated with color characteristics, mainly monomeric anthocyanins, while clone 169 is correlated with individual phenolic compounds.
Resumo:
This paper presents a validation study on statistical nonsupervised brain tissue classification techniques in magnetic resonance (MR) images. Several image models assuming different hypotheses regarding the intensity distribution model, the spatial model and the number of classes are assessed. The methods are tested on simulated data for which the classification ground truth is known. Different noise and intensity nonuniformities are added to simulate real imaging conditions. No enhancement of the image quality is considered either before or during the classification process. This way, the accuracy of the methods and their robustness against image artifacts are tested. Classification is also performed on real data where a quantitative validation compares the methods' results with an estimated ground truth from manual segmentations by experts. Validity of the various classification methods in the labeling of the image as well as in the tissue volume is estimated with different local and global measures. Results demonstrate that methods relying on both intensity and spatial information are more robust to noise and field inhomogeneities. We also demonstrate that partial volume is not perfectly modeled, even though methods that account for mixture classes outperform methods that only consider pure Gaussian classes. Finally, we show that simulated data results can also be extended to real data.