给出了一种新的图像信息量的表述方法，提出了五种基于信息熵的科学分割遥感图像的改进方法。用MATLAB对这些算法进行了实验仿真，并进行了算法性能优化。得出了各种不同算法的最佳阈值和耗时，并对它们进行了比较和分析。实验数据表明，新算法较以往的算法在分割效果和速度上都有明显的改善和提高。

A new definition of image information quantum is proposed. Five improved entropy information based on scientific segmentation methods for remote sensing images is brought forward. These algorithms are simulated with MATLAB,and some optimizations of algorithms' performances are made. The best thresholds and consumed time of different algorithms are compared and analyzed. These experiment data show that these new algorithms make qualitative improvements in precision and speed compared with other algorithms.