predicting morphological changes in rivers using image processing (case study: qizil ouzan river)

Morphological studies are among the most important topics in river engineering, dealing with the geometric shape, bed form, and longitudinal profile of the waterway, cross-sections, deformations, and lateral migration of rivers over time. researchers worldwide have noted the capability of satellite imagery and its integration with geographic information systems (gis) to provide comprehensive information about river conditions and monitor spatial changes at different time intervals. in this paper, a new method is proposed to detect the current morphology of different parts of the qizil ouzan river and its main branches using a cnn neural network with support vector machine (svm) and multilayer perceptron (mlp). the results of the studies show that the current meander length of the river is more than anything else the result of non-river factors such as stream piracy and mass movements like landslides and lateral collapses of waterways. moreover, using gis software, the meander and its changes over different time periods were obtained and compared with each other. also, the present research aims to predict river morphological changes using image processing in matlab. the results demonstrate the efficacy of combining these two methods for predicting the morphological changes of the qizil ouzan river.