The development of effective content-based multimedia search systems is an important research issue due to the growing amount of digital audio-visual information. In the case of images and video, the growth of digital data has been observed since the introduction of 2D capture devices. A similar development is expected for 3D data as acquisition and dissemination technology of 3D models is constantly improving. 3D objects are becoming an important type of multimedia data with many promising application possibilities. Defining the aspects that constitute the similarity among 3D objects and designing algorithms that implement such similarity definitions is a difficult problem. Over the last few years, a strong interest in methods for 3D similarity search has arisen, and a growing number of competing algorithms for content-based retrieval of 3D objects have been proposed. We survey feature-based methods for 3D retrieval, and we propose a taxonomy for these methods. We also present experimental results, comparing the effectiveness of some of the surveyed methods.

/pdf/feature-based-similarity-search-in-3d-object-databases-2smlg2mk38.pdf

Feature-based similarity search in 3D object databases

. Primary biological aerosol particles (PBAP) are an important subset of air particulate matter with a substantial contribution to the organic aerosol fraction and potentially strong effects on public health and climate. Recent progress has been made in PBAP quantification by utilizing real-time bioaerosol detectors based on the principle that specific organic molecules of biological origin such as proteins, coenzymes, cell wall compounds and pigments exhibit intrinsic fluorescence. The properties of many fluorophores have been well documented, but it is unclear which are most relevant for detection of atmospheric PBAP. The present study provides a systematic synthesis of literature data on potentially relevant biological fluorophores. We analyze and discuss their relative importance for the detection of fluorescent biological aerosol particles (FBAP) by online instrumentation for atmospheric measurements such as the ultraviolet aerodynamic particle sizer (UV-APS) or the wide issue bioaerosol sensor (WIBS). In addition, we provide new laboratory measurement data for selected compounds using bench-top fluorescence spectroscopy. Relevant biological materials were chosen for comparison with existing literature data and to fill in gaps of understanding. The excitation-emission matrices (EEM) exhibit pronounced peaks at excitation wavelengths of ~280 nm and ~360 nm, confirming the suitability of light sources used for online detection of FBAP. They also show, however, that valuable information is missed by instruments that do not record full emission spectra at multiple wavelengths of excitation, and co-occurrence of multiple fluorophores within a detected sample will likely confound detailed molecular analysis. Selected non-biological materials were also analyzed to assess their possible influence on FBAP detection and generally exhibit only low levels of background-corrected fluorescent emission. This study strengthens the hypothesis that ambient supermicron particle fluorescence in wavelength ranges used for most FBAP instruments is likely to be dominated by biological material and that such instrumentation is able to discriminate between FBAP and non-biological material in many situations. More detailed follow-up studies on single particle fluorescence are still required to reduce these uncertainties further, however.

/pdf/autofluorescence-of-atmospheric-bioaerosols-fluorescent-15ndsaotpy.pdf

Autofluorescence of atmospheric bioaerosols – fluorescent biomolecules and potential interferences

In this paper, we describe a multiscale and multishape morphological method for pattern-based analysis and classification of gray-scale images using connected operators. Compared with existing methods, which use structuring elements, our method has three advantages. First, in our method, the time needed for computing pattern spectra does not depend on the number of scales or shapes used, i.e., the computation time is independent of the dimensions of the pattern spectrum. Second, size and strict shape attributes can be computed, which we use for the construction of joint 2D shape-size pattern spectra. Third, our method is significantly less sensitive to noise and is rotation-invariant. Although rotation invariance can also be approximated by methods using structuring elements at different angles, this tends to be computationally intensive. The classification performance of these methods is discussed using four image sets: Brodatz, COIL-20, COIL-100, and diatoms. The new method obtains better or equal classification performance to the best competitor with a 5 to 9-fold speed gain

/pdf/connected-shape-size-pattern-spectra-for-rotation-and-scale-mocktw0o1m.pdf

Connected Shape-Size Pattern Spectra for Rotation and Scale-Invariant Classification of Gray-Scale Images

The histogram of oriented gradients (HOG) is widely used for image description and proves to be very effective. In many vision problems, rotation-invariant analysis is necessary or preferred. Popular solutions are mainly based on pose normalization or learning, neglecting some intrinsic properties of rotations. This paper presents a method to build rotation-invariant HOG descriptors using Fourier analysis in polar/spherical coordinates, which are closely related to the irreducible representation of the 2D/3D rotation groups. This is achieved by considering a gradient histogram as a continuous angular signal which can be well represented by the Fourier basis (2D) or spherical harmonics (3D). As rotation-invariance is established in an analytical way, we can avoid discretization artifacts and create a continuous mapping from the image to the feature space. In the experiments, we first show that our method outperforms the state-of-the-art in a public dataset for a car detection task in aerial images. We further use the Princeton Shape Benchmark and the SHREC 2009 Generic Shape Benchmark to demonstrate the high performance of our method for similarity measures of 3D shapes. Finally, we show an application on microscopic volumetric data.

https://www.tpbin.com/Uploads/Subjects/f26c55f4-9fbe-449a-8165-d9a9523e8d71.pdf

Rotation-Invariant HOG Descriptors Using Fourier Analysis in Polar and Spherical Coordinates

New parameterisations were developed to calculate the emission and resuspension flux of pollen grains. These new parameterisations were included in the comprehensive mesoscale model system KAMM/DRAIS. Two types of simulations were performed. In the first case, horizontally homogeneous meteorology was assumed. In this case, pollen concentration rapidly decreased with the distance from the source. In the second case, where the fully three-dimensional model was applied, atmospheric lifetime of the pollen grains increased remarkably. This was mainly caused by the vertical wind speeds induced by topographic effects. Consequently, the pollen grains could travel much larger distances until they were deposited and finally removed from the atmosphere. This is an important finding with respect to the problem of cross pollination. Due to the lack of measurements, a number of parameters had to be assumed. However, the parameterisation proposed may well serve as a starting point of a daily pollen forecast with numerical models.

Numerical modelling of pollen dispersion on the regional scale

Identifying and counting of pollen grains in ambient air samples is still a demanding and time-consuming task even for an experienced microscopist. This article describes a technique which may be employed to establish a fully automated system for this task. Based on a 3D volume fluorescence image of a pollen grain taken with a confocal laser scanning microscope, the described system is able to recognize the pollen taxa. The system autonomously extracts all required information for the recognition from a data base with reference objects (self-learning system) and only needs to calculate very general purpose features of the volumetric data sets (so-called gray scale invariants). This allows for easy adaptation of the system to other conditions (e.g., pollen of a special area) or even other objects than pollen (e.g., spores, bacteria etc.) just by exchanging the reference data base. When using a reference data base with the 26 most important German pollen taxa, the recognition rate is 92%. With a special database for allergic purposes recognizing only Corylus, Alnus, Betula, Poaceae, Secale, Artemisia and ``allergically non-relevant'' the recognition rate is 97.4%.

Automated pollen recognition using 3D volume images from fluorescence microscopy

A technique is described which may be employed to establish a fully automated system for the recognition of airborne pollens. As different pollen taxa have only marginal differences, a full 3D volume data set of the pollen grain was recorded with a confocal laser scanning microscope (LSM) at a voxel size of about (0.2 /spl mu/m)/sup 3/. This represents an intrinsic and complete data set. 14 invariant gray-scale features based on an integration over the 3D Euclidian transformation group with nonlinear kernels were extracted from these volume data sets. The classification was done with support vector machines. The use of these general gray scale features allows one to easily adapt the system to other objectives (e.g., pollen of a special area) or even other objects than pollen (e.g., spores, bacteria, etc.) just by exchanging the reference database. When using a reference database with the 26 most important German pollen taxa (385 samples), the recognition rate is 92%. With a special database for allergological purposes recognizing only Corylus, Alnus, Betula, Poaceae, Secale, Artemisia and "allergological nonrelevant", the recognition rate is 97.4%.

/pdf/general-purpose-object-recognition-in-3d-volume-data-sets-3w6chig35y.pdf

General-purpose object recognition in 3D volume data sets using gray-scale invariants - classification of airborne pollen-grains recorded with a confocal laser scanning microscope

Automated pollen recognition using gray scale invariants on 3D volume image data

Since the middle of the twentieth century there has been growing concern about air quality. Since the mid-1990th, new European framework directives relating to air quality gave rise to intense discussions about limit values for ambient exposure and their practical application. With respect to the impact on human health, particulate matter has to be differentiated by its aerodynamic diameter [1]. Particles with an aerodynamic diameter less than 10 μm, denoted as PM10, can enter the lower respiratory tract, i.e., the bronchial tubes in the lung. However, the pulmonary alveoli can only be reached by the finer particle fraction with an aerodynamic diameter less than 2.5 μm, denoted as PM2.5. A study by the World Health Organization (WHO) notes a correlation of ambient exposure of PM2.5 with mortality and cardiovascular diseases [2]. An increased risk of respiratory diseases correlated with ambient exposure to PM10 was also noted [3]. Beyond the determination of the total mass concentration of particulate matter, it is reasonable to focus on aerosol components owing to their broad spectrum of health impacts. While soot particles are alleged to be carcinogenic, the beneficial effect of airborne sea salt on respiratory diseases is clearly evident. Microscopy is a powerful tool for the differentiation of coarse particulate matter and allows a detailed analysis of air quality.

/pdf/online-monitoring-of-airborne-allergenic-particles-omnibuss-3qrjvjoi6f.pdf

Online Monitoring of Airborne Allergenic Particles (OMNIBUSS)

Automatic pollen recognition has been developed based on socalled gray-scale invariants, which characterise pollen grains independently from their position and orientation on the microscopic sample. Thus, pollen features can be extracted from the gray-scale images of transmitted light and fluorescence microscopy. In a first step, this approach is demonstrated with Ambrosia pollen of samples from a Burkard sampler, where pollen are collected from ambient air on a sticky tape mounted on a slowly rotating drum. Self-learning Support Vector Machines create a classification model from the gray-scale invariants of the particles on three Burkard samples from Mezzana (Ticino), Switzerland. Automatic pattern recognition is tested with 13 other samples from the period between July, 20th and September, 9th 2004. A recall of 77.3 % has been found for the automatic recognition of Ambrosia pollen, together with a precision of 84.0 % for this classification. Falsely negative classified objects can partly be ascribed to agglomerated pollen, the number of falsely positive classified objects can be reduced by a more specific classification mode. Automatic pollen recognition provides the basis for the development of a fully automated system that combines sampling, particle deposition onto a surface suitable for optical analysis, automatic preparation, microscopic imaging techniques, pattern recognition and the hourly output of number concentration of airborne pollen. Keywords: automatic pollen recognition, Ambrosia pollen, primary fluorescence, gray-scale invariants, Support Vector Machines, pattern recognition, pollen monitor, Burkard sampler, online- measurement Automatische Pollenerkennung – Entwicklungen und Perspektiven Die automatische Erkennung von Pollen wurde auf der Basis von so genannten Grauwert-Invarianten entwickelt, die Pollen unabhangig von ihrer Position und Orientierung auf der mikroskopischen Probe charakterisieren. Somit konnen Polleneigenschaften aus den mikroskopischen Graustufen-Bildern (Durchlicht und Fluoreszenz) abgeleitet werden. Zunachst wird dieser Ansatz am Beispiel von Ambrosia -Pollen auf Proben aus Burkardfallen demonstriert, in denen Pollen aus der Ausenluft auf einem Klebeband gesammelt werden, das auf einer langsam rotierenden Trommel befestigt ist. Selbstlernende Support-Vector-Machines erzeugen ein Klassifikationsmodell aus den Grauwert-Invarianten der Partikel von drei Burkardproben aus Mezzana (Tessin), Schweiz. Die automatische Mustererkennung wird an 13 weiteren Proben aus der Zeit zwischen dem 20. Juli und dem 9. September 2004 getestet. Fur die automatische Erkennung von Ambrosia -Pollen ergibt sich eine Erkennungsrate von 77,3 % bei einer Bestimmungsgenauigkeit von 84,0 % fur diese Klassifizierung. Falsch negativ klassifizierte Objekte konnen teilweise auf agglomerierte Pollen zuruckgefuhrt werden, die Zahl falsch positiv klassifizierter Objekte kann durch eine spezifischere Klassifikation reduziert werden. Die automatische Pollenerkennung bildet die Basis der Entwicklung eines voll automatisierten Systems, das Probenahme, Partikelabscheidung auf einer mikroskopierfahigen Oberflache, automatische Praparation, mikroskopische Abbildungstechniken, Mustererkennung und die stundliche Angabe der Anzahlkonzentration luftgetragener Pollen vereint. Stichworter: automatische Pollenerkenung, Ambrosia -Pollen, Eigenfluoreszenz, Grauwert-Invarianten, Support-Vector-Machines, Mustererkennung, Pollenmonitor, Burkardfalle, Online- Messung

Eckart Schultz

Papers

Automated pollen recognition using 3D volume images from fluorescence microscopy

General-purpose object recognition in 3D volume data sets using gray-scale invariants - classification of airborne pollen-grains recorded with a confocal laser scanning microscope

Automated pollen recognition using gray scale invariants on 3D volume image data

Online Monitoring of Airborne Allergenic Particles (OMNIBUSS)

Automatic pollen recognition : developments and perspectives