ANALYSIS OF CHANGE DETECTION ALGORITHMS FOR STEREO AND AREA SURVEY DATA PROCESSING

The paper examines change detection methods in satellite imagery acquired through stereo and area imaging. It reviews traditional algorithms (image differencing, regression analysis, PCA, PCC) alongside neural network approaches (CNN, RNN, hybrid models, autoencoders). Mathematical models, training principles, and evaluation metrics (Accuracy, Precision, Recall, F-score) are presented, the advantages and limitations of the methods are discussed.