基于城市POI的遥感影像渐进压缩技术

传统栅格影像均采用简单的、由低分辨率到高分辨率的像素级渐进压缩模式,较少考虑到用户需求特征和知识,因此,文中利用POI数据特点及图像感兴趣区编码特性,提出一种基于城市POI的遥感影像渐进压缩思想:首先根据大量的POI数据分析挖掘用户关注的热度信息,建立兴趣场,并以此确定遥感影像的感兴趣区域,然后综合SPIHT算法与Maxshift算法对遥感影像进行渐进压缩编码。实验结果表明,该方法在低码率下仍可以高质量保留图像所含重要信息,能够很好地满足用户的需求,实现了知识层级的遥感影像渐进压缩,有效提高图像压缩编码的实用性和优越性。

Remote sensing image progressive compression based on city POI

The traditional raster image is a simple pixel level compression mode from low resolution to high resolution without considering much about users'requirements.With the characteristics of POI (points of interest)data reflecting user needs and ROI coding,a method for remote sensing image progressive compression based on city POI is proposed.First,the application and importance of a great deal of POI data are analyzed from rich attribute information,position characteristics and user interest.A concept of space interest field is put forward.Then SPIHT algorithm and Maxshift algorithm are integiated to encode with compression for the remote sensing images.The experimental results show that the important and users'interested information of remote sensing image can be retained with high quality when the bit rate is low,and can well meet the needs of users. The method realizes remote sensing image progressive compression by knowledge hierarchy, and also improves the practicability and superiority of image compression.