Fusion of Stochastic Properties by Fuzzy Integrals and Applications on Detection within Images
نویسنده
چکیده
This paper is concerned with the detection and separation of textures or stochastic information within an image. The texture is described mathematical by their stochastic. Because different kinds of stochastic processes generate the textures, they are represented by a system of coupled non-linear stochastic differential equations. If parameters a priori are not known, then their rank and derivations of this detect the stochastic. The distance of the interaction approximates the non-linearity. The entire knowledge does not exist and so the probability theory cannot be used, because the normalization implicit the knowledge of all possible properties. The usage of fuzzy measures compensates the loss of additivity. By fuzzy measures special properties without conditioned relationships to other properties can be described. The fuzzy measure allows the representation of different kinds of properties. The fusion of different kinds of properties related to their importance is realized by the fuzzy integral. The fuzzy integral is used as a new fuzzy measure and by iteration selected properties are isolated from the image successively. This method is applied for different stochastic or textural features within images such as clouds, contrails, error structures of surfaces of tissues, metals or medical images with stochastic structures. KURZFASSUNG Der Vortrag beschäftigt sich mit dem Auffinden und der Separation von texturellen oder stochastischen Informationen in einem Bild. Die Textur wird mathematisch durch ihre Stochastik beschrieben. Da verschiedene Arten von stochastischen Prozessen die Textur erzeugen, werden diese durch ein System von gekoppelten stochastischen nichtlinearen Differentialgleichungen beschrieben. Wenn a priori keine Kenntnis über stochastische Parameter existieren, wird über die Bildung des Ranges die relevante stochastische Information aus dem Bild herausgezogen. Die Entfernung der Wechselwirkung wird über die Art der Nichtlinearität repräsentiert. Da eine vollständige Kenntnis über das System nicht existiert, kann mit Wahrscheinlichkeiten nicht gearbeitet werden, da die Normierung eine vollständige Kenntnis von allen Eigenschaften erfordert. Die Benutzung des Fuzzy-Maßes kann diesen Verlust der Additivität kompensieren. Mit dem Fuzzy Maß können verschiedenartige Eigenschaften beschrieben werden. Die Fusion dieser verschiedenen Eigenschaften erfolgt mit dem Fuzzy Integral. Das Fuzzy Integral wird als neue Fuzzy Maß benutzt und damit kann iterativ eine stochastische Eigenschaft aus dem Bild isoliert werden. Die Ergebnisse werden angewandt auf Erkennung von Texturmuster wie Wolken, Kondensstreifen, Störungen in Oberflächen von Stoffen, Metallen oderr in medizinischen Bilden mit stochastischen Strukturen.
منابع مشابه
Separation of Different Textures in Images using Fuzzy Measures and Fuzzy Functions and their Fusion by Fuzzy Integrals
This paper shows the detection of textures by analysing the image by stochastic properties. Different parts of the composed stochastic properties are fused by a decision directed method. The stochastic is mapped on a fuzzy measure and a fuzzy function. The fusion of the so represented stochastic information is achieved by the fuzzy integral. The result of the fuzzy integral is used for a new fu...
متن کاملSoft Data Fusion for the Industrial Application of Computer Vision
The relevance of information fusion methodologies increases due to the complementary development of computer and sensory technologies. Newly hardware and software facilities allow the inclusion of different information sources in a computer system. Information fusion basically attains the transformation of the information delivered by multiple sources into one representational form. The fused d...
متن کاملDynamic Systems and Applications 19 (2010) 473-494 FUZZY STOCHASTIC INTEGRAL EQUATIONS
In this paper we propose a new approach to fuzzy stochastic integrals of Itô and Aumann type. Then a fuzzy equation with fuzzy stochastic integrals is investigated. The existence and uniqueness of solution is proven. Some typical properties of the solution are also obtained. Similar results to set-valued stochastic integral equations are stated.
متن کاملAn efficient method for cloud detection based on the feature-level fusion of Landsat-8 OLI spectral bands in deep convolutional neural network
Cloud segmentation is a critical pre-processing step for any multi-spectral satellite image application. In particular, disaster-related applications e.g., flood monitoring or rapid damage mapping, which are highly time and data-critical, require methods that produce accurate cloud masks in a short time while being able to adapt to large variations in the target domain (induced by atmospheric c...
متن کاملThe using of Haar wavelets for the expansion of fractional stochastic integrals
Abstract: In this paper, an efficient method based on Haar wavelets is proposed for solving fractional stochastic integrals with Hurst parameter. Properties of Haar wavelets are described. Also, the error analysis of the proposed method is investigated. Some numerical examples are provided to illustrate the computational efficiency and accuracy of the method.
متن کامل