Evaluation of super-resolution algorithm for detection and recognition of features from MODIS and OLI images at sub-pixel scale using Hopfield Neural Network

Abstract

Fuzzy classification techniques have been developed recently to estimate the class<br />composition of image pixels, but their output provides no indication of how these<br />classes are distributed spatially within the instantaneous field of view represented by<br />the pixel. Super-resolution land-cover mapping is a promising technology for<br />prediction of the spatial distribution of each land-cover class at the sub-pixel scale.<br />This distribution is often determined based on the principle of spatial dependence and<br />from land-cover fraction images derived with soft classification technology. As such,<br />while the accuracy of land cover target identification has been improved using fuzzy<br />classification, it remains for robust techniques that provide better spatial representation<br />of land cover to be developed. An approach was adopted that used the output from a<br />fuzzy classification to constrain a Hopfield neural network formulated as an energy<br />minimization tool. The network converges to a minimum of an energy function. This<br />energy minimum represents a “best guess" map of the spatial distribution of class<br />components in each pixel. The technique was applied to remote sensing imagery<br />(MODIS & OLI images), and the resultant maps provided an accurate and improved<br />representation of the land covers. Low RMSE, high accuracy. By using a Hopfield<br />neural network, more accurate measures of land cover targets can be obtained, The Hopfield neural network used in this way represents a simple, robust, and efficient<br />technique, and results suggest that it is a useful tool for identifying land cover targets<br />from remotely sensed imagery at the sub-pixel scale. The present research purpose was<br />evaluation of HNN algorithm efficiency for different land covers (Land, Water,<br />Agriculture land and Vegetation) through Area Error Proportion, RMSE and<br />Correlation coefficient parameters on MODIS & OLI images and related ranking,<br />results of present super resolution algorithm has shown that according to precedence,<br />most improvement in feature's recognition happened for Water, Land, Agriculture<br />land and ad last Vegetation with RMSEs 0.044, 0.072, 0.1 and 0.108.