Analysis of full-waveform LiDAR data for classification of an orange orchard scene

Full-waveform laser scanning data acquired with a Riegl LMS-Q560 instrument were used to classify an orange orchard into orange trees, grass and ground using waveform parameters alone. Gaussian decomposition was performed on this data capture from the National Airborne Field Experiment in November 2006 using a custom peak-detection procedure and a trust-region-reflective algorithm for fitting Gauss functions. Calibration was carried out using waveforms returned from a road surface, and the backscattering coefficient c was derived for every waveform peak. The processed data were then analysed according to the number of returns detected within each waveform and classified into three classes based on pulse width and c. For single-peak waveforms the scatterplot of c versus pulse width was used to distinguish between ground, grass and orange trees. In the case of multiple returns, the relationship between first (or first plus middle) and last return c values was used to separate ground from other targets. Refinement of this classification, and further sub-classification into grass and orange trees was performed using the c versus pulse width scatterplots of last returns. In all cases the separation was carried out using a decision tree with empirical relationships between the waveform parameters. Ground points were successfully separated from orange tree points. The most difficult class to separate and verify was grass, but those points in general corresponded well with the grass areas identified in the aerial photography. The overall accuracy reached 91%, using photography and relative elevation as ground truth. The overall accuracy for two classes, orange tree and combined class of grass and ground, yielded 95%. Finally, the backscattering coefficient c of single-peak waveforms was also used to derive reflectance values of the three classes. The reflectance of the orange tree class (0.31) and ground class (0.60) are consistent with published values at the wavelength of the Riegl scanner (1550 nm). The grass class reflectance (0.46) falls in between the other two classes as might be expected, as this class has a mixture of the contributions of both vegetation and ground reflectance properties.

Vedegenskaper i härskande och behärskade granar vid Remingstorp : Wood properties in dominant and sub-dominant trees of Norway spruce at Remingstorp

Variation in wood properties for Picea abies trees and logs of different dimensions has been studied at two sites in southern Sweden of different site quality class. Trees have been classified as dominant or sub-dominant, according to their height. Log and board grades were classified and strength grade of boards, basic density and annual ring width measured. A similar study made on four northern sites was used as reference material.Sub-dominant trees were of superior quality in comparison to dominant trees, when classified by log and board grades or strength grading. Differences were accentuated for the second log where the sub-dominant trees had superior strength and low amount of boards with coarse branches. The results correspond well to those from the northern region, Jämtland. The classifica¬tion of boards as well as bending strength indicated superior properties on timber from northern sites even though the basic density was similar.

Estudo longitudinal das alterações dentoesqueléticas à longo prazo da má oclusão de classe II, divisão 1ª tratada com o Bionator de Balters utilizando-se a cefalometria radiográfica e a morfometria geométrica

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Boulder with plaque beside elm tree planted by the class of 1869. In front of Angell Hall to the left of the steps. Gift of the Class of 1869.

On verso: M-10

Plaque on floor of Admission's office honoring tree planted by class of 1869

On verso: M-14

Multi-class sentiment analysis using a hierarchical logistic model tree approach

En este trabajo se propone un nuevo sistema híbrido para el análisis de sentimientos en clase múltiple basado en el uso del diccionario General Inquirer (GI) y un enfoque jerárquico del clasificador Logistic Model Tree (LMT). Este nuevo sistema se compone de tres capas, la capa bipolar (BL) que consta de un LMT (LMT-1) para la clasificación de la polaridad de sentimientos, mientras que la segunda capa es la capa de la Intensidad (IL) y comprende dos LMTs (LMT-2 y LMT3) para detectar por separado tres intensidades de sentimientos positivos y tres intensidades de sentimientos negativos. Sólo en la fase de construcción, la capa de Agrupación (GL) se utiliza para agrupar las instancias positivas y negativas mediante el empleo de 2 k-means, respectivamente. En la fase de Pre-procesamiento, los textos son segmentados por palabras que son etiquetadas, reducidas a sus raíces y sometidas finalmente al diccionario GI con el objetivo de contar y etiquetar sólo los verbos, los sustantivos, los adjetivos y los adverbios con 24 marcadores que se utilizan luego para calcular los vectores de características. En la fase de Clasificación de Sentimientos, los vectores de características se introducen primero al LMT-1, a continuación, se agrupan en GL según la etiqueta de clase, después se etiquetan estos grupos de forma manual, y finalmente las instancias positivas son introducidas a LMT-2 y las instancias negativas a LMT-3. Los tres árboles están entrenados y evaluados usando las bases de datos Movie Review y SenTube con validación cruzada estratificada de 10-pliegues. LMT-1 produce un árbol de 48 hojas y 95 de tamaño, con 90,88% de exactitud, mientras que tanto LMT-2 y LMT-3 proporcionan dos árboles de una hoja y uno de tamaño, con 99,28% y 99,37% de exactitud,respectivamente. Los experimentos muestran que la metodología de clasificación jerárquica propuesta da un mejor rendimiento en comparación con otros enfoques prevalecientes.

Designing mixed species tree plantations for the tropics : balancing ecological attributes of species with landholder preferences in the Philippines

A mixed species reforestation program known as the Rainforestation Farming system was undertaken in the Philippines to develop forms of farm forestry more suitable for smallholders than the simple monocultural plantations commonly used then. In this study, we describe the subsequent changes in stand structure and floristic composition of these plantations in order to learn from the experience and develop improved prescriptions for reforestation systems likely to be attractive to smallholders. We investigated stands aged from 6 to 11 years old on three successive occasions over a 6 year period. We found the number of species originally present in the plots as trees >5 cm dbh decreased from an initial total of 76 species to 65 species at the end of study period. But, at the same time, some new species reached the size class threshold and were recruited into the canopy layer. There was a substantial decline in tree density from an estimated stocking of about 5000 trees per ha at the time of planting to 1380 trees per ha at the time of the first measurement; the density declined by a further 4.9% per year. Changes in composition and stand structure were indicated by a marked shift in the Importance Value Index of species. Over six years, shade-intolerant species became less important and the native shade-tolerant species (often Dipterocarps) increased in importance. Based on how the Rainforestation Farming plantations developed in these early years, we suggest that mixed-species plantations elsewhere in the humid tropics should be around 1000 trees per ha or less, that the proportion of fast growing (and hence early maturing) trees should be about 30–40% of this initial density and that any fruit tree component should only be planted on the plantation margin where more light and space are available for crowns to develop.

Bayesian and Decision Tree Approaches for Pattern Recognition Including Feature Measurement Costs

The minimum cost classifier when general cost functionsare associated with the tasks of feature measurement and classification is formulated as a decision graph which does not reject class labels at intermediate stages. Noting its complexities, a heuristic procedure to simplify this scheme to a binary decision tree is presented. The optimizationof the binary tree in this context is carried out using ynamicprogramming. This technique is applied to the voiced-unvoiced-silence classification in speech processing.

Landscape structure influences tree density patterns in fragmented woodlands in semi-arid eastern Australia.

Landscape and local-scale influences are important drivers of plant community structure. However, their relative contribution and the degree to which they interact remain unclear. We quantified the extent to which landscape structure, within-patch habitat and their confounding effects determine post-clearing tree densities and composition in agricultural landscapes in eastern subtropical Australia. Landscape structure (incorporating habitat fragmentation and loss) and within-patch (site) features were quantified for 60 remnant patches of Eucalyptus populnea (Myrtaceae) woodland. Tree density and species for three ecological maturity classes (regeneration, early maturity, late maturity) and local site features were assessed in one 100 × 10 m plot per patch. All but one landscape characteristic was determined within a 1.3-km radius of plots; Euclidean nearest neighbour distance was measured inside a 5-km radius. Variation in tree density and composition for each maturity class was partitioned into independent landscape, independent site and joint effects of landscape and site features using redundancy analysis. Independent site effects explained more variation in regeneration density and composition than pure landscape effects; significant predictors were the proportion of early and late maturity trees at a site, rainfall and the associated interaction. Conversely, landscape structure explained greater variation in early and late maturity tree density and composition than site predictors. Area of remnant native vegetation within a landscape and patch characteristics (area, shape, edge contrast) were significant predictors of early maturity tree density. However, 31% of the explained variation in early mature tree differences represented confounding influences of landscape and local variables. We suggest that within-patch characteristics are important in influencing semi-arid woodland tree regeneration. However, independent and confounding effects of landscape structure resulting from previous vegetation clearing may have exerted a greater historical influence on older cohorts and should be accounted for when examining woodland dynamics across a broader range of environments.

Tree mortality and deadwood dynamics in late-successional boreal forests

Here I aimed at quantifying the main components of deadwood dynamics, i.e. tree mortality, deadwood pools, and their decomposition, in late-successional boreal forests. I focused on standing dead trees in three stand types dominated by Picea mariana and Abies balsamea in eastern Canada, and on standing and down dead trees in Picea abies-dominated stands in three areas in Northern Europe. Dead and living trees were measured on five sample plots of 1.6-ha size in each study area and stand type. Stem disks from dead trees were sampled to determine wood density and year of death, using dendrochronological methods. The results were applied to reconstruct past tree mortality and to model deadwood decay class dynamics. Site productivity, stand developmental stage, and the occurrence of episodic tree mortality influenced deadwood volume and quality. In all study areas tree mortality was continuous, leading to continuity in deadwood decay stage distribution. Episodic tree mortality due to either autogenic or allogenic causes influenced deadwood volume and quality in all but one study area. However, regardless of productivity and disturbance history deadwood was abundant, accounting for 20 53% of total wood volume in European study areas, and 15 27% of total standing volume in eastern Canada. Deadwood was a persistent structural component, since its expected residence time in early- and midstages of decay was 18 yr even in the area with the most rapid decomposition. The results indicated that in the absence of episodic tree mortality, stands may eventually develop to a steady state, in which deadwood volume fluctuates around an equilibrium state. However, in many forests deadwood is naturally variable, due to recurrent moderate-severity disturbances. This variability, the continuous tree mortality, and variation in rates of wood decomposition determine the dynamics and availability of deadwood as a habitat and carbon storage medium in boreal coniferous forest ecosystems.

Activity recognition using binary tree SVM

This paper presents an effective classification method based on Support Vector Machines (SVM) in the context of activity recognition. Local features that capture both spatial and temporal information in activity videos have made significant progress recently. Efficient and effective features, feature representation and classification plays a crucial role in activity recognition. For classification, SVMs are popularly used because of their simplicity and efficiency; however the common multi-class SVM approaches applied suffer from limitations including having easily confused classes and been computationally inefficient. We propose using a binary tree SVM to address the shortcomings of multi-class SVMs in activity recognition. We proposed constructing a binary tree using Gaussian Mixture Models (GMM), where activities are repeatedly allocated to subnodes until every new created node contains only one activity. Then, for each internal node a separate SVM is learned to classify activities, which significantly reduces the training time and increases the speed of testing compared to popular the `one-against-the-rest' multi-class SVM classifier. Experiments carried out on the challenging and complex Hollywood dataset demonstrates comparable performance over the baseline bag-of-features method.

Time and Energy Complexity of Distributed Computation of a Class of Functions in Wireless Sensor Networks

We consider a scenario in which a wireless sensor network is formed by randomly deploying n sensors to measure some spatial function over a field, with the objective of computing a function of the measurements and communicating it to an operator station. We restrict ourselves to the class of type-threshold functions (as defined in the work of Giridhar and Kumar, 2005), of which max, min, and indicator functions are important examples: our discussions are couched in terms of the max function. We view the problem as one of message-passing distributed computation over a geometric random graph. The network is assumed to be synchronous, and the sensors synchronously measure values and then collaborate to compute and deliver the function computed with these values to the operator station. Computation algorithms differ in (1) the communication topology assumed and (2) the messages that the nodes need to exchange in order to carry out the computation. The focus of our paper is to establish (in probability) scaling laws for the time and energy complexity of the distributed function computation over random wireless networks, under the assumption of centralized contention-free scheduling of packet transmissions. First, without any constraint on the computation algorithm, we establish scaling laws for the computation time and energy expenditure for one-time maximum computation. We show that for an optimal algorithm, the computation time and energy expenditure scale, respectively, as Theta(radicn/log n) and Theta(n) asymptotically as the number of sensors n rarr infin. Second, we analyze the performance of three specific computation algorithms that may be used in specific practical situations, namely, the tree algorithm, multihop transmission, and the Ripple algorithm (a type of gossip algorithm), and obtain scaling laws for the computation time and energy expenditure as n rarr infin. In particular, we show that the computation time for these algorithms scales as Theta(radicn/lo- g n), Theta(n), and Theta(radicn log n), respectively, whereas the energy expended scales as , Theta(n), Theta(radicn/log n), and Theta(radicn log n), respectively. Finally, simulation results are provided to show that our analysis indeed captures the correct scaling. The simulations also yield estimates of the constant multipliers in the scaling laws. Our analyses throughout assume a centralized optimal scheduler, and hence, our results can be viewed as providing bounds for the performance with practical distributed schedulers.

Tree mortality, fire scar formation and regeneration 8 years after low-intensity fire in managed Pinus sylvestris stands

Fire is an important driver of the boreal forest ecosystem, and a useful tool for the restoration of degraded forests. However, we lack knowledge on the ecological processes initiated by prescribed fires, and whether they bring about the desired restoration effects. The purpose of this study was to investigate the impacts of low-intensity experimental prescribed fires on four ecological processes in young commercial Scots pine (Pinus sylvestris) stands eight years after the burning. The processes of interest were tree mortality, dead wood creation, regeneration and fire scar formation. These were inventoried in twelve study plots, which were 30 m x 30 m in size. The plots belonged to two different stand age classes: 30-35 years or 45 years old at the time of burning. The study was partly a follow-up of study plots researched by Sidoroff et al. (2007) one year after burning in 2003. Tree mortality increased from 183 stems ha-1 in 2003 to 259 stems ha-1 in 2010, corresponding to 15 % and 21 % of stem number respectively. Most mortality was experienced in the stands of the younger age class, in smaller diameter classes and among species other than Scots pine. By 2010, the average mortality of Scots pine per plot was 18%, but varied greatly ranging from 0% to 63% of stem number. Delayed mortality, i.e. mortality that occurred between 2 and 8 years after fire, seemed to become more important with increasing diameter. The input of dead wood also varied greatly between plots, from none to 72 m3 ha-1, averaging at 12 m3 ha-1. The amount of fire scarred trees per plot ranged from none to 20 %. Four out of twelve plots (43 %) did not have any fire scars. Scars were on average small: 95% of scars were less than 4 cm in width, and 75% less than 40 cm in length. Owing to the light nature of the fire, the remaining overstorey and thick organic layer, regeneration was poor overall. The abundance of pine and other seedlings indicated a viable seed source existed, but the seedlings failed to establish under dense canopy. The number of saplings ranged from 0 to 12 333 stems ha-1. The results of this study indicate that a low intensity fire does not necessarily initiate the ecological processes of tree mortality, dead wood creation and regeneration in the desired scale. Fire scars, which form the basis of fire dating in fire history studies, did not form in all cases.