No abstract

As yet various methods have been used for determining the salinity rate of seas and oceans water. The current method of determining salinity rate of seas water has been field examination of various points of sea and determining its salinity rate. In the last decade, remote sensing satellite images have had high capability in determining sea waters salinity rate. Regarding that the present methods in remote sensing depend on the studied regions, therefore, the necessity of customization of these methods is felt. Fresh water springs due to impact on water salinity and temperature and also the environment physics and density like sound velocity are very significant and since coasts and islands of Persian Gulf are considered among arid and semi-arid regions and lack drinking water, access to fresh water springs has more significance. After studies performed, preparation of salinity rate observations and catching two series of proper images for felid data for complete coverage of the region, preprocessing and calibration was performed. For this purpose in turning the acquired radiance to reflection, ENVI software was used. The histogram of calibrated shades of gray rates in images was specified, so that reflection of each sample can be extracted from images. In this paper, the rate of least method efficiency in determining salinity rate of Persian Gulf waters was examined and finally identifying fresh water pits using remote sensing technique was done. The obtained results in the least squares methods after combining various bands of image with each other specified that combining 4 bands of 2, 3, 5 and 7 has the least standard deviation rate with training data and test, which is equal to 0.385 and 0.991978.

Soil and water conservation is one of the key measures to improve the ecological environment of small watersheds and maintain the healthy life of the region. With the current method used to monitor the evolution of topographic and hydrological characteristics of a small watershed, the edge keeping index of the image and the signal-to-noise ratio of the image are low, the monitoring results are inaccurate, and there are some problems such as low edge keeping ability, poor denoising effect, and low monitoring accuracy. A monitoring method based on remote sensing and GIS for the evolution of topographic and hydrological characteristics of small watersheds is proposed. The hyperspectral data was transformed by the Principal Component Analysis (PCA). Group principal component images, use sparse representation method based on adaptive dictionary and dual-tree complex wavelet transform method to denoise principal component images with a small amount of information were the objectives of the paper; also the use the multi-scale wavelet transform to detect image edge; build a binary model of a pixel, extract vegetation index, and terrain factor based on a binary model of the pixel, and realize terrain and hydrological characteristics of the small watershed. The experimental results show that the proposed method has the high edge-preserving ability, good denoising effect, and high monitoring accuracy.