Study On Methods Of Pseudo-color Enhancement And Visualization For Polarimetric SAR Image

As an advanced remote sensing instrument, polarimetric synthetic aperture radar (SAR) cannot only work at all-day and all-weather condition, but also obtain polarimetric images frommulti-bands. Aiming at improving the SAR image visual effects and enriching detailsinformation which make target recognition and images interpretation of polarimetric SAR morefeasible, study emphasis is put on the key techniques for polarimetric SAR image enhancement,such as the pseudo-color enhancement algorithm for single polarimetric SAR images,pseudo-color fusion algorithm and three-dimensional terrain visualization technologies formulti-polarimetric SAR images, etc.The main work and innovations of this thesis are listed as follows:1. Explanation were given on the descriptions and characters of RGB, HIS and lαβ colorspaces which was involved in pseudo-color enhancement for single polarimetric SAR images,and the transformation methods between the RGB color space and HIS, lαβ color spaces werepresented too. The classic methods for assessing image quality were summarized, and a newmethod based on fractal dimension was introduced.2. A pseudo-color enhancement algorithm based on non-linear coding and lαβ color spacetransformation for single polarimetric SAR images was proposed. And the quality of thepseudo-color enhanced images was assessed. The validity of the proposed algorithm was provedby the result of assessment.3. The methods for colorizing the single polarimetric SAR using color transfer algorithmswere studied. The main factors which affect the treatment effect were analyzed. The colortransfer algorithms based on histogram matching and neighbor areas searching method weredesigned for colorizing treatment of single polarimetric SAR images.4. According the different kind of polarization states, the pseudo-color fusion algorithmsbased on the pixel operation, Pauli decomposition, H-α decomposition and on the best bandchoosing model and HIS color space transformation were designed separately. The experimentswere finished by using fpolarimetric SAR images. The validities of the designed algorithms wereproved by the result of assessment.5. The method for creating and displaying three-dimensional visualization scence by usingOSG was designed. And the method for three-dimensional visualization of terrain was realizedbased on pseudo-color enhancing polarimetric SAR images.