The Research On Remote Sensing Image Compression Based On JPEG2000

Nowadays, the latest flying development of remote sensing technique has made it a realization to get a multi-platform, multi-sensor and all-day observation of the earth. Consequently, it makes the amount of the obtained image really gigantic and it urgently need efficient image compression technique. In 2000, JPEG2000 was approved as a new still image compression standard by a collaborative effort between the international Organization for Standardization (ISO) and International Electrotechnical Commission (IEC). JPEG2000 forms a whole suit of image compression standard by effectively integrating the two excellent fruits, wavelet transform theory and efficient data structure method: embedded block coding with optimized truncation (EBOCT), which were developed from these latest ten years digital signal process domain. JPEG2000 not only dramatically improve the image compression rate and the compression coding/decoding efficiency, but also greatly optimize image data structure and management system by offering some specially function. Furthermore, because of the standard's exoteric character, JPEG2000 will replace original image compression technical in the near future.In the remote sensing image compression research domain, there are two primary research interest, one is to develop some efficiently image compression algorithms which is specially for aviation or remote sensing image characteristic, the other is to appraise the impression of the image compression method for the photogrammetric and remote sensing application by research the compressed images' quality attenuation degree. The thesis works on remote sensing image compression based on JPEG2000 standard. It realizes a JPEG2000 image compression experiment system and carries through a series of remote sensing image compression performance analyzed experiments based on this experiment system. Furthermore, by these series of experiments analyses, a comprehensive objective JPEG2000 remote sensing image compression potential can be acquired and it then can make preparations to the development of remote sensing database system based on JPEG2000 and promotes the JPEG2000 standard application in remote sensing domain.The thesis discussed common technical methods and image compression quality appraise method etc firstly, then it analyzed the JPEG2000's compression principles, method routine and new characteristics, based on these analyses, it finished two mainly work.Firstly, the thesis realized a JPEG2000 remote sensing image compression experiment system by taking Jasper, a JPEG2000 software library commended by JPEG2000 standard organize, as the system's core compression coder/decoder in the VC++ software development environment. The compression system has four mainly function modules: basically JPEG2000 compression coding control module, JPEG2000 decoding control module, JPEG2000 new characters such as ROI compression module, remote sensing image compression appraisement module. The system has a preparatory practicability for the system's function.Secondly, the thesis carried through some JPEG2000 compression coding control experiments and some JPEG2000 remote sensing image compression potential experiments based on our JPEG2000 remote sensing experiment system. To the JPEG2000 compression coding control problem, the thesis put interests in two wavelet transform factors: one is wavelet type, the other is wavelet reduce levels. By experiment, it is concluded that the reversible 5/3 wavelet has strongpoint than the unreversible 9/7 wavelet and wavelet reduce level is suitable at about 3 to 5. To the JPEG2000 remote sensing image compression performance problem, the thesis introduces the widely applied base JPEG standard as comparison. Based on our developed system's remote sensing image compression appraisement module, the thesis put up a great number compression experiments to kinds of remote sensing images. It was concluded that JPEG2000 has better subjective image quality and better whole and part image radiate quality than base JPEG, the quality difference was especially remarkably when the compression ratio is comparatively high, on the other hand, for the image geometry quality which was cared by remote sensing application, it was concluded that when the compression ratio is within about 10, JPEG2000 and JPEG both can have good image geometry quality. When the ratio is within about 20 , JPEG2000 and JPEG has about similar geometry quality. JPEG2000 has much better image geometry quality than JPEG when the compression rate is above about 20.