Research On The Image Smoothing Technology Based On PDE And It's Application

Of all the medias for human being getting information the most important is image. Because of noises existing in everywhere so there are always some noises in image. How to remove and degrade it has been becoming a constant research topic in the fields of image processing. Many smoothing image methods have been searched and used to remove the noise from image. In our developing State "15" ResearchProject-he System of Forensic Image Processing, searching and using newsmoothing image methods is one of our main tasks. This paper have deeply researched the newly formed image smoothing technology based on partial differential equation (PDE) after investigating lots of image smoothing methods. We get some new findings on removing noises from image and smoothing image. And we realized its application by using this technology in our developing Forensic Image Processing System.Smoothing image should balance the removing noises and preserving features. Image smoothing technology based on PDE is an adaptive smoothing method. During the course of smoothing, image features and their directions are extracted. Less smoothes in the locations with strong image features, and more smoothes in the locations with weak image feature; minimal smoothes in the directions across the image features, and maximal smoothes in the directions along the image features. The smoothing can make noises removed while preserve edge well. And it is a quite well smoothing method.Perona and Malik's work on anisotropic diffusion has been one of the most influential papers in the area. They proposed to replace Gaussian smoothing by a directional diffusion that preserves edges. Their work opened a number of theoretical and practical questions in the PDE image processing fields. After consulting many references we make the following research:1) Analysing Perona-Malik Equation by taking it into the framework of robust3estimation. Getting their equal relationship. By investigating the property of diffusion function we proposed an anisotropic diffusion function, the smoothing effects produces sharper boundaries.2) Processing image by using Perona-Malik Equation evolves image to partly planar image. This is the main reason causing blocky effects. We have exploited the smoothing method based on four order PDE. Smoothing image by four order PDE we can get more natural image.3) Evolves Perona-Malik Equation to the form of weighted smoothing in two orthogonal directions. Brings Hessian method, which is an approach to extract the image feature direction, into the technology of smoothing image. Perona-Malik Equation extracts image feature direction based on gradient method, which can be very erroneous. However the damages of image feature -edge, line and wave caused by the Hessian method is significantly less.4) We have been developing an image processing system in VC++6.0 IDE. We developed an image class. Every image processing technology mentioned in this paper is realized as one of the image class member functions.