The Efficiency of Pyramidal Algorithms for Image Representation using Morphological Filters
نویسنده
چکیده
Pyramidal multiresolution image representation is an ordered sequence of decreasing spatial resolution images in which each image is filtered version of its predecessor. Pyramidal representation of an image is applied in various fields of image processing: progressive image transmission used in interactive image communication over low-bandwidth channels, for image compression used in transmission and storage of visual information, in computer vision for image filtering and analysis at multiple scales. The pyramid approach is attractive due to low computational complexity and simple parallel implementation. A morphological hierarchical representation reflects the natural decomposition of shapes in the image. Mathematical morphology operates directly with shapes without affecting the remaining image structure and therefore provides a better basis than linear techniques for constructing pyramids. The morphological pyramidal coder has several attractive features: 8 bit integer operations only, a perfect reconstruction mode, progressive transmission and progressive computation property. Pyramidal image representation and compression techniques using morphological filters are studied through experimental analysis based on a collection of five images with a broad spectrum of structural detail. Comparisons of results for three types of morphological pyramidal representation techniques, basic and modified expansive and non-expansive, are made based on different structuring elements, different morphological antialiasing and interpolation filters. The focus in the research is put on morphological filters. Pyramidal algorithms for image representation are implemented with different antialiasing morphological filters: dilate, erode, close, open, open(close), close(open), with two types of structuring elements: square and cross-type. Morphological interpolation filters dilate and close with square type structuring element is implemented. As a result of such analysis, modified expansive pyramidal algorithm for image representation is more efficient than basic expansive algorithm. The superiority in terms of lossless compression performance and computational simplicity is obtained using non-expansive algorithm with morphological filters. Key-Words: pyramidal image representation and compression, mathematical morphology CSCC'99 Proc.pp.1351-1356
منابع مشابه
Image Restoration with Two-Dimensional Adaptive Filter Algorithms
Two-dimensional (TD) adaptive filtering is a technique that can be applied to many image, and signal processing applications. This paper extends the one-dimensional adaptive filter algorithms to TD structures and the novel TD adaptive filters are established. Based on this extension, the TD variable step-size normalized least mean squares (TD-VSS-NLMS), the TD-VSS affine projection algorithms (...
متن کاملA comprehensive benchmark between two filter-based multiple-point simulation algorithms
Computer graphics offer various gadgets to enhance the reconstruction of high-order statistics that are not correctly addressed by the two-point statistics approaches. Almost all the newly developed multiple-point geostatistics (MPS) algorithms, to some extent, adapt these techniques to increase the simulation accuracy and efficiency. In this work, a scrutiny comparison between our recently dev...
متن کاملPalarimetric Synthetic Aperture Radar Image Classification using Bag of Visual Words Algorithm
Land cover is defined as the physical material of the surface of the earth, including different vegetation covers, bare soil, water surface, various urban areas, etc. Land cover and its changes are very important and influential on the Earth and life of living organisms, especially human beings. Land cover change monitoring is important for protecting the ecosystem, forests, farmland, open spac...
متن کاملOn the use of Textural Features and Neural Networks for Leaf Recognition
for recognizing various types of plants, so automatic image recognition algorithms can extract to classify plant species and apply these features. Fast and accurate recognition of plants can have a significant impact on biodiversity management and increasing the effectiveness of the studies in this regard. These automatic methods have involved the development of recognition techniques and digi...
متن کاملA morphological pyramidal image decomposition
A multiresolution image representation is presented in which iterative morphological filters of many scales hut identical shape serve as basis functions. Structural pattern decomposition is achieved by subtracting successive layers in the multiresolution representation. The representation differs from established techniques in that the code elements have a well defined location and size. The re...
متن کامل