CENSUS HISTOGRAMS: A SIMPLE FEATURE EXTRACTION and MATCHING APPROACH FOR FACE RECOGNITION

Most face recognition approaches require a prior training where a given distribution of faces is assumed to further predict the identity of test faces. Such an approach may experience difficulty in identifying faces belonging to distributions different from the one provided during the training. A face recognition technique that performs well regardless of training is, therefore, interesting to consider as a basis of more sophisticated methods. In this work, the Census Transform is applied to describe the faces. Based on a scanning window which extracts local histograms of Census Features, we present a method that directly matches face samples. With this simple technique, 97.2% of the faces in the FERET fa/fb test were correctly recognized. Despite being an easy test set, we have found no other approaches in literature regarding straight comparisons of faces with such a performance. Also, a window for further improvement is presented. Among other techniques, we demonstrate how the use of SVMs over the Census Histogram representation can increase the recognition performance.

Aplicação de sistemas imunológicos artificiais para biometria facial: Reconhecimento de identidade baseado nas características de padrões binários

Pós-graduação em Engenharia Elétrica - FEIS

Texture analysis by multi-resolution fractal descriptors

This work proposes a novel texture descriptor based on fractal theory. The method is based on the Bouligand- Minkowski descriptors. We decompose the original image recursively into four equal parts. In each recursion step, we estimate the average and the deviation of the Bouligand-Minkowski descriptors computed over each part. Thus, we extract entropy features from both average and deviation. The proposed descriptors are provided by concatenating such measures. The method is tested in a classification experiment under well known datasets, that is, Brodatz and Vistex. The results demonstrate that the novel technique achieves better results than classical and state-of-the-art texture descriptors, such as Local Binary Patterns, Gabor-wavelets and co-occurrence matrix.

Combining fractal and deterministic walkers for texture analysis and classification

In this paper,we present a novel texture analysis method based on deterministic partially self-avoiding walks and fractal dimension theory. After finding the attractors of the image (set of pixels) using deterministic partially self-avoiding walks, they are dilated in direction to the whole image by adding pixels according to their relevance. The relevance of each pixel is calculated as the shortest path between the pixel and the pixels that belongs to the attractors. The proposed texture analysis method is demonstrated to outperform popular and state-of-the-art methods (e.g. Fourier descriptors, occurrence matrix, Gabor filter and local binary patterns) as well as deterministic tourist walk method and recent fractal methods using well-known texture image datasets.

An evolutive approach for smile recognition in video sequences

Facial expression recognition is one of the most challenging research areas in the image recognition ¯eld and has been actively studied since the 70's. For instance, smile recognition has been studied due to the fact that it is considered an important facial expression in human communication, it is therefore likely useful for human–machine interaction. Moreover, if a smile can be detected and also its intensity estimated, it will raise the possibility of new applications in the future

Pattern of nutrient availability and plant community assemblage in Everglades Tree Islands, Florida, USA

We address the relative importance of nutrient availability in relation to other physical and biological factors in determining plant community assemblages around Everglades Tree Islands (Everglades National Park, Florida, USA). We carried out a one-time survey of elevation, soil, water level and vegetation structure and composition at 138 plots located along transects in three tree islands in the Park’s major drainage basin. We used an RDA variance partitioning technique to assess the relative importance of nutrient availability (soil N and P) and other factors in explaining herb and tree assemblages of tree island tail and surrounded marshes. The upland areas of the tree islands accumulate P and show low N concentration, producing a strong island-wide gradient in soil N:P ratio. While soil N:P ratio plays a significant role in determining herb layer and tree layer community assemblage in tree island tails, nevertheless part of its variance is shared with hydrology. The total species variance explained by the predictors is very low. We define a strong gradient in nutrient availability (soil N:P ratio) closely related to hydrology. Hydrology and nutrient availability are both factors influencing community assemblages around tree islands, nevertheless both seem to be acting together and in a complex mechanism. Future research should be focused on segregating these two factors in order to determine whether nutrient leaching from tree islands is a factor determining community assemblages and local landscape pattern in the Everglades, and how this process might be affected by water management.

Toward robust image detection of crown-of-thorns starfish for autonomous population monitoring

Robust texture recognition in underwater image sequences for marine pest population control such as Crown-Of-Thorns Starfish (COTS) is a relatively unexplored area of research. Typically, humans count COTS by laboriously processing individual images taken during surveys. Being able to autonomously collect and process images of reef habitat and segment out the various marine biota holds the promise of allowing researchers to gain a greater understanding of the marine ecosystem and evaluate the impact of different environmental variables. This research applies and extends the use of Local Binary Patterns (LBP) as a method for texture-based identification of COTS from survey images. The performance and accuracy of the algorithms are evaluated on a image data set taken on the Great Barrier Reef.

Evaluation of spectral and texture features for object-based vegetation species classification using Support Vector Machines

The use of appropriate features to characterize an output class or object is critical for all classification problems. This paper evaluates the capability of several spectral and texture features for object-based vegetation classification at the species level using airborne high resolution multispectral imagery. Image-objects as the basic classification unit were generated through image segmentation. Statistical moments extracted from original spectral bands and vegetation index image are used as feature descriptors for image objects (i.e. tree crowns). Several state-of-art texture descriptors such as Gray-Level Co-Occurrence Matrix (GLCM), Local Binary Patterns (LBP) and its extensions are also extracted for comparison purpose. Support Vector Machine (SVM) is employed for classification in the object-feature space. The experimental results showed that incorporating spectral vegetation indices can improve the classification accuracy and obtained better results than in original spectral bands, and using moments of Ratio Vegetation Index obtained the highest average classification accuracy in our experiment. The experiments also indicate that the spectral moment features also outperform or can at least compare with the state-of-art texture descriptors in terms of classification accuracy.

Color and texture feature fusion using kernel PCA with application to object-based vegetation species classification

A good object representation or object descriptor is one of the key issues in object based image analysis. To effectively fuse color and texture as a unified descriptor at object level, this paper presents a novel method for feature fusion. Color histogram and the uniform local binary patterns are extracted from arbitrary-shaped image-objects, and kernel principal component analysis (kernel PCA) is employed to find nonlinear relationships of the extracted color and texture features. The maximum likelihood approach is used to estimate the intrinsic dimensionality, which is then used as a criterion for automatic selection of optimal feature set from the fused feature. The proposed method is evaluated using SVM as the benchmark classifier and is applied to object-based vegetation species classification using high spatial resolution aerial imagery. Experimental results demonstrate that great improvement can be achieved by using proposed feature fusion method.

Dynamic texture reconstruction from sparse codes for unusual event detection in crowded scenes

Unusual event detection in crowded scenes remains challenging because of the diversity of events and noise. In this paper, we present a novel approach for unusual event detection via sparse reconstruction of dynamic textures over an overcomplete basis set, with the dynamic texture described by local binary patterns from three orthogonal planes (LBPTOP). The overcomplete basis set is learnt from the training data where only the normal items observed. In the detection process, given a new observation, we compute the sparse coefficients using the Dantzig Selector algorithm which was proposed in the literature of compressed sensing. Then the reconstruction errors are computed, based on which we detect the abnormal items. Our application can be used to detect both local and global abnormal events. We evaluate our algorithm on UCSD Abnormality Datasets for local anomaly detection, which is shown to outperform current state-of-the-art approaches, and we also get promising results for rapid escape detection using the PETS2009 dataset.

Unusual Event Detection in Crowded Scenes Using Bag of LBPs in spatio-temporal patches

Modelling events in densely crowded environments remains challenging, due to the diversity of events and the noise in the scene. We propose a novel approach for anomalous event detection in crowded scenes using dynamic textures described by the Local Binary Patterns from Three Orthogonal Planes (LBP-TOP) descriptor. The scene is divided into spatio-temporal patches where LBP-TOP based dynamic textures are extracted. We apply hierarchical Bayesian models to detect the patches containing unusual events. Our method is an unsupervised approach, and it does not rely on object tracking or background subtraction. We show that our approach outperforms existing state of the art algorithms for anomalous event detection in UCSD dataset.

Human identification from video using advanced gait recognition techniques

The solutions proposed in this thesis contribute to improve gait recognition performance in practical scenarios that further enable the adoption of gait recognition into real world security and forensic applications that require identifying humans at a distance. Pioneering work has been conducted on frontal gait recognition using depth images to allow gait to be integrated with biometric walkthrough portals. The effects of gait challenging conditions including clothing, carrying goods, and viewpoint have been explored. Enhanced approaches are proposed on segmentation, feature extraction, feature optimisation and classification elements, and state-of-the-art recognition performance has been achieved. A frontal depth gait database has been developed and made available to the research community for further investigation. Solutions are explored in 2D and 3D domains using multiple images sources, and both domain-specific and independent modality gait features are proposed.

东营凹陷辛镇复杂断块油藏形成机制和剩余油分布规律研究

In this paper, the complex faulted-block oil reservoir of Xinzhen area in Dongying depression is systematically studied from basic conditions forming faulted-block oil and gas reservoir integrating geology, seismic, logging and reservoir engineering information and computer; guided by petroleum geology, geomechanics, structural geology and geophysics and other theories. Based on analysis of background condition such as regional strata, structure and petroleum geology, structural research on geometry, kinemaitcs and dynamics, oil-controlling fault research on the seal features, sealing mechanism and sealing pattern, and research on enrichment rules and controlling factors of complex faulted-block oil reservoir are carried out to give out the formation mechanics of oil reservoir of Xinzhen complex faulted-block oil reservoir. As a result, the reservoir formation pattern is established. At the same time, through dissecting the characteristics and hydrocarbon enrichment law of complex faulted-block oil reservoir, and studying its distribution law of remaining oil after entering extra high water-cut period, a set of technologies are formed to predict complex faulted-block oil reservoir and its remaining oil distribution and to enhance oil recovery (EOR). Based on the time relationship between migration of hydrocarbon and trap formation, accumulating period of Xinzhen oil reservoir is determined. The formation of Xinzhen anticlinal trap was prior to the primary migration. This is favorable to formation of Xinzhen anticlinal hydrocarbon reservoir. Meanwhile, because anticline top caving isn't at the sane time as that of moving or faulted-trap forming inner anticline, oil and gas migrated many times and Xinzhen complex faulted-block oil reservoir formed from ES_3~(upper) to EG. Accumulating law and controlling factors of complex faulted-block reservoir are analyzed from many aspects such as regional structure background controlling hydrocarbon accumulating, plastic arch-open structure controlling oil-bearing series and reservoir types, sealing-opening of fault controlling hydrocarbon distribution and structure pattern controlling enriched trap types. Also, we established the structure pattern in Xinzhen a'ea: the arch-open of underlying strata cause expanding fracture. The main block groups developed here are shovel-like normal fault block group in the north area of Xinzhen and its associated graben block group. Block groups dominate the formation and distribution of reservoirs. We studied qualitatively and quantitatively the sealing characteristics, sealing history and sealing mechanism of faults, too. And, the sealing characteristics are evaluated and the distribution pattern of hydrocarbon controlled by faults is researched. Due to movement intensity of big faults, deep falling of downthrown block, high degree of repture and development of fracture, shallow layers close to the downthrown block of secondary faults are unfavorable to hydrocarbon accumulation. This is confirmed by the exploration practice in Xinzhen anticline. In terms of the downthrown blocks of sencondary contemporaneous faults lied in the south and north area of Xinzhen, hydrocarbon is poor close to fracture belt, while it is relatively abundant in tertiary companion faults. Because of long-term movement of faults that control hydrocarbon, fi'om ES3 to EG, six set of oil-bearing series formed. And their opening causes the inhomogeneity in hydrocarbon abundance among each block--in two flanks of anticline reservoirs are abundant while in the axial area, oil and gas are sporadic. There the sealing characteristics control oil-bearing area of oil/gas accumulation and the height of oil reservoir. Longitudinally, oil and gas are enriched in dip-flat areas in mid-plane of faults. It is established that there are four types of accumulating patterns in complex faulted-block oil reservoirs in Xinzhen. The first is accumulating pattern of lithologic oil reservoirs in E~S_3~(mid-lowwer), that is, self-generating-self-reserving-self-covering lithologic trap pattern. The second is drag-anticline accumulating pattern in Xinzhen. The structure traps are drag anticlines formed by the contemporaneous faults of the second basement in the north of Xinzhen, and the multiple source rocks involve Ek_2, Es_4, Es_3 and Es_1 members. The reservoirs are fluvial-delta sandstones of the upper member of Shahejie formation and Guantao formation, covered by regional thick mudstone of the upper member of Guantao formation and MingHuazhen formation. The third is the accumulating pattern of reverse listric fault, the third-degree fault of Xinzhen anticline limb and the reservoirs form reservoir screened by reverse listric faults. The forth is accumulating pattern of crossing faults which form closing or semi-closing faulted-blocks that accumulate hydrocarbon. The technologies of predicting remaining oil in complex faulted-block reservoir during the mid and late development stage is formed. Remaining oil in simple large faulted-blocks enriches in structural high, structural middle, structural low of thick bottom water reservoirs, points near bent edge-fault oftertiary faults and part the fourth ones with big falling displacement, microstructure high place of oil-sandbodies and areas where local well pattern isn't perfect. While that in small complex faulted-blocks enriches near small nose, small high point, angle of small faults, small oil-bearing faulted-blocks without well and areas with non-perfect well pattern. The technologies of enhancing recovery factor in complex faulted-block reservoir during the mid and late development stage is formed as follows: fine reservoir description, drilling adjust wells, designing directional wells, sub-dividing layer series of development, improving flooding pattern, changing water-injection direction and enhancing swept volume, cyclic waterflooding and gas-injection, etc. Here, directional wells include directional deflecting wells, lateral-drilling wells, lateral-drilling horizontal wells and horizontal wells. The results of this paper have been used in exploration and development of Shengli oilfield, and have achieved great social and economic profit, especially in predicting distribution of complex faulted-block reservoir, remaining oil distribution during middle and late stage of development, and in EOR. Applying the achievement of fault-closure research, new hydrocarbon-bearing blocks are discovered in flanks of Dongying central uplift and in complex blocks with proved reserves 15 million tons. With the study of remaining oil distribution law in complex faulted-block reservoirs, recovery factors are increased greatly in Dongxin, Xianhe and Linpan complex faulted-block reservoirs and accumulated oil production increment is 3 million tons.

HDAC3 is crucial in shear- and VEGF-induced stem cell differentiation toward endothelial cells

Reendothelialization involves endothelial progenitor cell (EPC) homing, proliferation, and differentiation, which may be influenced by fluid shear stress and local flow pattern. This study aims to elucidate the role of laminar flow on embryonic stem (ES) cell differentiation and the underlying mechanism. We demonstrated that laminar flow enhanced ES cell-derived progenitor cell proliferation and differentiation into endothelial cells (ECs). Laminar flow stabilized and activated histone deacetylase 3 (HDAC3) through the Flk-1-PI3K-Akt pathway, which in turn deacetylated p53, leading to p21 activation. A similar signal pathway was detected in vascular endothelial growth factor-induced EC differentiation. HDAC3 and p21 were detected in blood vessels during embryogenesis. Local transfer of ES cell-derived EPC incorporated into injured femoral artery and reduced neointima formation in a mouse model. These data suggest that shear stress is a key regulator for stem cell differentiation into EC, especially in EPC differentiation, which can be used for vascular repair, and that the Flk-1-PI3K-Akt-HDAC3-p53-p21 pathway is crucial in such a process.

An evaluation of the pedestrian classification in a multi-domain multi-modality setup

The objective of this article is to study the problem of pedestrian classification across different light spectrum domains (visible and far-infrared (FIR)) and modalities (intensity, depth and motion). In recent years, there has been a number of approaches for classifying and detecting pedestrians in both FIR and visible images, but the methods are difficult to compare, because either the datasets are not publicly available or they do not offer a comparison between the two domains. Our two primary contributions are the following: (1) we propose a public dataset, named RIFIR , containing both FIR and visible images collected in an urban environment from a moving vehicle during daytime; and (2) we compare the state-of-the-art features in a multi-modality setup: intensity, depth and flow, in far-infrared over visible domains. The experiments show that features families, intensity self-similarity (ISS), local binary patterns (LBP), local gradient patterns (LGP) and histogram of oriented gradients (HOG), computed from FIR and visible domains are highly complementary, but their relative performance varies across different modalities. In our experiments, the FIR domain has proven superior to the visible one for the task of pedestrian classification, but the overall best results are obtained by a multi-domain multi-modality multi-feature fusion.