ON THE USE OF WORLDVIEW -2 IMAGE AND FEATURE EXTRACTION TECHNIQUE FOR RICE FIELDS MONITORING IN EGYPT

Monitoring the rice cultivation in Egypt has become a challenging issue for thecoming years to efficiently manage Egypt s complex agricultural production systemand vast irrigation and drainage networks. Herein, a good approach is adapted to mapflooding rice fields, by automating the process of rice polygon extraction fromWorldView- 2 satellite image, using imagine objective module . A sequence ofoperators and processes where applied on true color; 0.5 mlpixel resolutionWorldView 2 image dated 30111 April 2011 for Mit -Yazid area. A classification of bothrice pixels in transplanting stage and non-rice pixels was carried out by sampling andtraining these pixels. A mathematical morphology on the objects in the output rasterdomain was performed, followed by the conversion of the objects from the rasterdomain to the vector domain in order to change the shape of the vector objects throughcleaning the extracted polygons of rice fields. Also other GIS processing works werecarried out to assist the cultivated truth rice areas .Finally this approach willdramatically reduce the human work ,particularly that the values of the areas obtainedfrom digitizing work was almost similar to those from the feature extraction module,thus being 17.14 and 16.12 km² respectively for the same study area of 64 km².Thisbesides that the digitizing work is time consuming as compared to the featureextraction module which has been done in no time i.e. almost one hour. Nevertheless,this approach is reliable, cost-effective and helpful for decision makers to plan actionsneeded in water managements.