| 基于感兴趣区域的无人机侦察图像压缩 |
| |
| 作者姓名： | 邰华凤 黄大庆 |
| |
| 作者单位： | 南京航空航天大学信息科学与技术学院;南京航空航天大学无人机研究院; |
| |
| 摘 要： | 根据无人机侦察图像压缩的特殊性,综合JPEG2000中MaxShift和GSBM的特点提出一种改进的感兴趣区域图像压缩算法。该方法允许用户在一幅图像中指定任意形状的区域建立掩模,不需要将感兴趣区域的形状信息传递到解码端,能灵活调节感兴趣区域和背景区域的重构图像的相对质量。实验表明,该算法在低解码率(小于0.5bpp)时,与MaxShift相比,具有更高的编码质量(峰值信噪比最高可高出10dB左右)。
|
| 关 键 词： | JPEG2000 感兴趣区域 图像编码 峰值信噪比 |
UAV Reconnaissance Image Compression Based on Region of Interest |
| |
| Authors: | Tai Huafeng Huang Daqing |
| |
| Abstract: | According to the particularity of UAV reconnaissance image compression,this paper puts forward an improved method of ROI(Region of Interest) image compression coding which integrates the characteristics of MaxShift and GSBM in JPEG2000. This method allows users to establish the mask of the area of designated arbitrary shape in a picture and doesn't need to transmit the shape information of ROI to the decoding.It can adjust the contrast reconstruction image quality between the region of interest and the regi... |
| |
| Keywords: | JPEG2000 Region of interest Image coding PSNR |
| 本文献已被 CNKI 等数据库收录！ |
