Mathematical Morphology And Its Application Research In Remote Sensing Image Processing

Mathematical morphology is considered to be a fresh technique for image analysis with strong mathematical flavor, in particular for those applications where image geometric shape and structure characteristics quantitative analysis description aspects are relevant. It is a powerful nonlinear tool for extending by linear approach. One of the most important features of morphological transformation is that it can complete the mass complex image processing operations by using basic shift and logic operation. That is why the morphological method brings much great interest, attracts so many people to research its theory and arithmetic, and at the same time, stands as an established image analysis technique with applications in almost all science and engineering disciplines dealing with digital spatial data, which also let morphological theory and techniques go deep into development and innovation. Accordingly, It is important and significant to study how to use this new technique to geoscience and remote sensing, and make the remote sensing technique to be more powerful.The earth observation technique of remote sensing (RS) is critical to national security and sustainable development. Since the first Earth-viewing spacebome synthetic aperture radar was put into orbit, numerous radar sensors have been launched. The images provided by these satellites are now very useful in many remote sensing applications. The mathematical morphology approach for processing spatial data can be well summarized as recognition of an object simply means that all the rest has been eliminated from the scene. Image analysis tasks that can be tackled by morphological operations include image filtering, image segmentation and image measurements, which are relevant to most applications dealing with earth observation imagery. It is therefore not surprising that there are so many successful applications in geoscience and remote sensing. It is also reveal that the theory of mathematical morphology still remains largely unexplored. This has motivated us to research the mathematical morphology analysis approach in remote sensing image processing. From the point of view of remote sensing application, in this paper, we extend and establish some new theories and applied methods relevant to mathematical morphology and remote sensing image processing. The main work of this paper is concluded as following several aspects.1) Propose the theory system and structural form of morphological transformationsIn this section, the approach of structure function is used to extend the basic concept of morphological operators, which based on the shift transformation, and consolidate transformation pattern is set up by studying the algebraic and topology frame of the model space of morphological operators. The research work concerning the relation of Boolean function and morphological transformation shows that all the binary morphological transformation can be represented by rank transformation or, more generally, weighed morphological convolution transformation. By selecting diverse Boolean function, morphological transformation can be extended both in form and in performance. This work also establishs a bridge between gray and binary morphology, and propose a new idea for the study of morphological analysis arithmetic. Based on the structural form of binary morphological operators, the configurable characteristics of threshold morphological transformation are also discussed. A way to farther research the theories and applied methods of mathematical morphology is designated by studying the connected morphological transformation, reconstruction open and close morphological transformation.2) Propose the elements thinking of morphological wavelet and theory scheme of linear and non-linear morphological wavelet transformationWavelet analysis is a popular tool in the field of image processing. In this section, an applied scheme to build multiresolution linear morphological wavelets and non-linear morphological adjunction pyramids transformations is proposed, which based on the theory of wavelet and mathematical morphology. An effective way to gain morphological wavelet transformation isprovided by the scheme, and the corresponding theory has been proved. The theory of adjunction morphological operator is successfully introduced to the technique of multiresolution analysis, with which, a series of non-linear morphological adjunction pyramids transformations and corresponding arithmetics will be acquired based on different morphological adjunction transformations.3) Theorize the generalized morphological transformation systems and applied scheme Mathematical morphology is a powerful tool for shape analysis, which is grounded on theintuitive notion of probing an image with a shape of a given size and modifying the image value depending on whether or not the shape fits within the image components. The generalized morphological transformations established in this section are extended by ordinary morphological transformation based on the concept of set cardinality, which have the ability to exhibit the intensity of the given shape fill within the image. This research shows that the ordinary morphology operators are the special case of generalized morphological transformation, and the generalized morphological transformation can also be regarded as the morphological filter form produced by structure function or threshold operator.4) Propose application scheme of generalized morphological pattern spectrum in image recognition and classificationThe shape quantity analysis and description are the important content in computer vision and image processing. In multitudinous image processing technique, mathematical morphology provides a effective way to solve this question due to its basic method of image probing. Morphological pattern spectrum, which based on the morphological open and close, is a successful paradigm by decomposing the object in different scalar sequence similar to the structure element to get the shape information scale distribution. This method can be applied to image recognition and classification due to image representation with no redundancy in multiresolution form.5) Propose morphological sampling and interpolation suit to remote sensing imageThe simplest sampling method is point sampling. In our case, we want to develop tools that will enable us to preserve some details that are considered important even if they are theoretically too small, while trying to avoid aliasing. Morphological sampling based on the image content can realize this aim by selecting suitable reference image, which is built by means of morphological tools. The main interest of reference sampling is that it allows the construction of content dependent sampling methods. Using the theory of generalized morphological transformation, we also propose a sampling scheme based on it, which can reduce image information redundancy and rebuild the input image effectively.In the last section, the theory and morphological transformation methods of multivalued image for remote sensing image processing are put forward to extend the gray morphology to multivalued morphology. Ordering relations are fundamental to the theory of mathematical morphology. The key point to extend morphological operators to the remote sensing image is to establish a vectorial ordering relation that induces a lattice structure on the multivalued image space. We discuss the possible extensions found in the literature, study an ordering scheme based on the reduced ordering, and propose the theory of multivalued morphology based on that ordering relation. We also provide some properties of multivalued morphology transformation, which indicate that the multivalued morphological filters and morphological operators based on different data can be built by the ordering relation.