Crop growth monitoring study by using multi-temporal index image cube analysis

Crop physiology analysis and growth monitoring are of great importance for precision agriculture. Remote sensing technology supplies us with more selections and available spaces in this dynamic change study by producing images of different spatial, different spectral and different temporal resolutions. Especially, the remote sensing data of high spectral and high temporal resolution will play a key role in land cover studies at national, regional and global scales. In this paper, a new effective data structure conception, temporal index image cube (MIIC) is proposed in aid of the parameterization of the multi-dimension spectral curve. MIIC is proved very useful for the dynamic analysis of vegetation. Based on multi-temporal meteorological satellite data and multi-temporal ground spectral measurements, the temporal characteristics of different vegetation physiological parameters are contrasted and analyzed from the temporal index image cube. It turned out, each crop has its differentiated temporal index curve. Based on this multi-temporal index image cube, the classification and identification on various crops is very practicable. Some spectral indices perform fairly well on monitoring the growth situation of crops. But more efforts still should be made to develop new effective and general spectral indices, which will be more sensible to land cover types and the growth situation, moreover, it will screen the effects of atmosphere and background to a large extent.