IRS-P6 LISS-3和Landsat-5 TM两种多光谱传感器影像数据的对比分析

Study on Relationship between IRS-P6 LISS-3 and Landsat-5 TM Remote Sensing Images

随着载有各种新型传感器的卫星相继发射升空,不同传感器之间的相互比较成了一个研究热点。首先从传感器的轨道特征、光谱范围等方面对IRS-P6 LISS-3和Landsat-5 TM进行了机理方面的对比分析,选取3对同日过空的遥感影像,研究了IRS-P6 LISS-3和Landsat-5 TM遥感数据在可见光—近红外、短红外各对应光谱波段之间的关系,建立TM和LISS-3各波段之间的相互转换公式,与Chander等提出的转换公式进行对比分析。结果表明:实际TM和模拟TM多光谱数据之间具有较强的相关关系,决定系数R2均大于0.97;模拟TM与实际TM的水体指数(MND-WI)和归一化植被指数(NDVI)空间散点分布和实际LISS-3与实际TM的空间散点分布相比,具有更强的相关关系,其决定系数R2有一定提高,散点分布趋于对称。因此,所求的关系转换方程具有较高的精度和有效性,效果优于Chander等提出的转换公式。

The successful launches of many new satellite sensors have brought a new hot topic of comparison in different sensors. Firstly,the study conducted the sensors characteristics by comparison between IRS-P6 LISS-3 and Landsat-5 TM, including orbit characteristic and spectrum. Secondly, this three synchronous sensor imagery of two sensors were obtained as material, and the comparison analysis was carried between corresponding V-NIR and SWIR bands of two sensors. And then,a conversion equation was developed be- tween TM and LISS-3. Finally,a comparative analysis about this formula was made with the Chander＇s validate the conversion equation. The results showed that the determinant coefficient comparison between actual TM and simulated TM were remarkably improved before conversion, Coefficients of determination were all greater than 0.97. Some comparisons were based on Modified Normalized Difference Water Index （MNDWI） and Normalized Difference Vegetation Index （NDVI） respectively,were carried out using actual TM and simulated TM data. These determinant coefficient comparisons between actual TM and simulated TM were a certain improved before conversion,and these scatter spatial distributions of MNDWI and ND- VI were more reasonable and symmetric uniform. Therefore,it is feasible and effective to provide relation e- quations. This effect is batter than conversion equations from Chander.