Land cover/land use mapping using different wavelet packet transforms for LISS IV imagery

Augmenting Weighted Average with Confusion Matrix to Enhance Classification Accuracy

“…are being calculated with the help of different data and they are being compared [2,3,4,5,6,7,8,9]. In [2], Mittal et al devised to compare producer and user accuracies on land cover images with the help of expectation-maximization algorithm applying on data provided by JAXA, Japan.…”

“…In [4], Samiappan et al present a Non-Uniform Random Feature Selection (NU-RFS) within a Multi-Classifier System (MCS) framework and experimental results demonstrate the superiority of the proposed approach compared to SVM and RFS. In [5], Experimental results show that a multi-band and multi-level wavelet packet approach can be used to drastically increase the classification accuracy. In [6], a new method is proposed using a data structure called Peano Count Tree (Ptree) for decision tree classification and the accuracy is possessed using the parameters overall accuracy, User's accuracy and Producer's accuracy for image classification methods of object oriented classification, Knowledge Base Classification, Post classification and P-tree Classifier.…”

The Vegetation Extraction and Hierarchical Classification using an IRS-1C LISS III Image

“…Pixel with value i occurs horizontally adjacent to a pixel with the value j. Each element (i, j) in Co-occurrence matrix specifies the number of times that the pixel with value i occurs horizontally adjacent to a pixel with the value j [8]. Using total number of pixels in an image and pixels present in each clustered image, percentage of each cluster in an image has been computed.…”

“…NDVI for a given pixel always results in a number that ranges from −1 to +1. Generally, non-vegetated areas gives values close to zero and vegetated areas gives values close to one indicating the high possible density of green leaves [8]. The NDVI images were examined, mean and standard deviation values were calculated and a thresholding technique was applied to separate vegetation from other land cover.…”

“…As satellite sensor resolution is not high enough, the remote sensing image here is expressed by mixed pixels [7] with presence of two or more feature in single pixel. If the window size is too small, insufficient ground level information is extracted to characterize a specific land cover and if the window size is too large, it can overlap two types of ground cover or mixed pixels will occur and thus introduce erroneous spatial information [8]. So frames of satellite image were divided into an average area which was expedient for analysis.…”