Parameter investigation of Artificial Neural Network and Support Vector Machine for image classification

Comprehensive review on land use/land cover change classification in remote sensing

“…SVM Classification 4,20,40,41,44,47,51,56,62,63,67 This method helps to find the hyper-plane that separates two or more LU/LC classes in the satellite image.…”

“…ANN Classification [4][5][6][7]15,17,21,35,43,53,56,72,73 The pixels in the satellite images are trained and separated into LU/LC classes through the learning process.…”

“…Researchers use image enhancement techniques to reduce the dimensionality of the satellite data: PCA, MNF, Independent Component Analysis (ICA) and wavelet dimensionality reduction. 9,19,25,29,30,34,46,47,51,53,54,56,[61][62][63]67,80,82 Frequently used atmospheric correction methods are Dark Object Subtraction (DOS), Quick Atmospheric Correction (QUAC), FLAASH, Apparent Reflectance Model (ARM) and the F mask method. [7][8][9]12,15,[18][19][20]23,28,29,33,35,37,40,41,45,49,[61][62][63]66,67,70,74,75,79 Geometric corrections include Orthorectification, Geo-referencing, Image Registration, ASCII Coordinate Conversion and Resampling.…”

“…Image Enhancement 9,19,25,29,30,34,46,47,51,53,54,56,[61][62][63]67,80,82 This method helps in reducing the dimensionality and enhancing the contrast of the satellite image.…”

“…The overall accuracy and the kappa coefficient are computed to validate the LU/LC classified result. 1,3,4,[7][8][9][10][12][13][14][15][16][17]19,20,22,23,[25][26][27][28][53][54][55][56][57]62,63,[66][67][68]70,72,73,80…”