Vegetos

Abstract

A field study has been carried out in Pantnagar; district U.S. Nagar of Uttarakhand State for wheat acreage estimation, during Rabi season of the year 2010–2011. Two types of data viz. satellite data of Haldwani region dated 22^nd March 2010 and ground truth data have been used for the study. ENVI-4.8 (Environment for Visualizing Images-4.8) software was used for image pre processing and analysis of the data. Wheat crop at the maximum vegetative stage was used to capture the crop spectral characteristics. Supervised classification was used to identify the wheat crop in Pantnagar. Ground truth data combined with visual interpretation obtained from LANDSAT image dated 22^nd March 2010 were used as training sets for supervised classification. The training pixels were selected from different fields. The total numbers of five classes were identified for which separate ROIs (Region of Interest) were generated in order to train the image processing software. The accuracy of the classification was determined by comparing the test set with the classification results to generate producers, user s and the overall accuracies. The overall accuracy obtained was 100%. Post classification statistics was generated to find out pixels belonging to different categories. The acreage of all the classes was computed by multiplying the total number of pixels in a category with the area of single pixel. The size of each pixel in LANDSAT image is 30m and hence the area covered by each pixel is 30mx30m (900m²). The total number of pixels obtained was 16,495. Therefore, total area under study was found to be 1484.55 ha.