Study On Soil Salinization Information Extraction In Arid Oasis Area Based On GF-1 Image

Soil salinization, one of the forms of land degradation, usually appear in the arid and semi- arid regions where the climate is drought, soil evaporation is very intensity, and the water table is high and contains soluble salts. Soil salinization is one of obstacles to development of oasis agriculture in Xinjiang due to its severe effects on agriculture productivity and sustainable development. In order to understand the threat level of soil salinization in oasis agriculture, to ensure the sustainable development of oasis agriculture in the arid and semi-arid regions, it is necessary to study the method of soil salinization monitoring. At present, remote sensing technology has been widely used to monitor soil salinization and it is very useful. However, monitoring soil salinization is mainly based on the low-resolution satellite images, which is not enough for monitoring soil salinization details.With GF-1 satellite successfully launched, it makes possible to use high-resolution images to monitor soil salinity. In this paper, aiming GF-1 image, according to its imaging features, salinization information extract ion methods were studied for GF-1 WFV images and GF-1 PMS respectively.First of all, we extract spectral indices from GF-1 WFV multispectral images, then analyze the correlation between spectral index and the measured soil salt. According to correlation coefficient, we assess which spectral index is suitable for monitoring soil salinity based on GF-1 WFV images in arid oases areas. Then we compare spectral indices extracted from Landsat8 O LI image, getting the following conclusions:1) In the selection of 12 kinds of salt and three planting index, SI, SI1, SI3, S3 index is close to the salt and soil salt content, high correlation, positive correlation, the S3 salt index correlation is highest. Vegetation index NDVI is in good correlation with soil salinity, negative correlation. These spectral indices are more suitable for evaluating soil salinization in arid oasis areas based on GF-1WFV images.2) Due to differences in resolution, the correlation between the measured soil salt context and spectral index extracted from the GF-1 WFV image is far higher than Landsat8 O LI image. Due to low resolution of remote sensing image, the mixed pixels contain many kinds of feature information, resulting a decline in correlation between image salt index and the measured salinity. High resolution remote sensing image can more truly reflect the features of information, more suitable for the evaluation of soil salinization.3) Salt index and vegetation index extracted from the GF-1 WFV images, not only can reflect the distrib ution of soil salinization on a large regional scale in the entire study area, but also on the field scale in the farmland area, it can be a good indicator of soil salinization.In addition, we use the domestic GF-1 PMS image for extraction of salinization information based on the advanced object-oriented method. At first, we use fractal net evolution approach to segment image and build classification rules for salinization information extraction. In addition, we use the maximum likelihood method extract soil salinization from the domestic GF-1 PMS image and Landsat OLI image in the same region at the same time. We compare the results of two different methods and different latest sensors for salinization information extraction. The results show:1) The overall accuracy of object-oriented method for salinization information extraction based on GF-1 PMS image is 92.94% and the kappa coefficient is 0.91. The overall accuracy of maximum likelihood method for salinization information extraction based on GF-1 PMS image is 87.78% and the kappa coefficient is 0.77. Compared with the maximum likelihood method, object-oriented method is better for soil salinization extraction based on GF-1 PMS image, and the overall accuracy is improved by 5 percentage points. This illustrates that object-oriented method is more suitable for GF-1 PMS image when monitoring soil salinization. In addition, object-oriented method can make full use of the relationship between pixels through scale segmentation technology, and can more fully utilize the information contained in the image to improve the accuracy of salinization extraction from high-resolution remote sensing image information.2) The overall accuracy of Landsat8 OLI image for soil salinization extraction is only 63.47%. Compared with Landsat O LI image, the overall accuracy of GF-1 PMS image is improved by 30 percentage points. The ability of GF-1image for soil salinization extraction is stronger than Landsat image. We can extract the vegetation that is affected by soil salinizatio n, which is meaningful for study of agricultural field scale salinization. This illustrates that GF-1 PMS images has great potential for salinization monitoring on agricultural field scale. In this study, we use domestic GF-1 PMS image to extract soil salinization information based on the advanced object-oriented method for the first time. The result is positive.Above all, this study shows that domestic GF-1 image can be a good data source to monitor soil salinization in the arid and semi-arid regions.