Optical remote sensing image change detection method, storage medium and computing device

Abstract

The invention discloses an optical remote sensing image change detection method, a storage medium and computing equipment, and the method comprises the steps: enabling an optical remote sensing imagedata set to generate a training set and a test set, and expanding the data set; constructing two encoder networks with the same network structure and shared parameters as a twin neural network for extracting multistage features; calculating multi-scale difference characteristics and establishing a decoder network; building a sub-network; training a network by using the expanded data; inputting thetest image into a network, obtaining a network output prediction result at one end of a decoder, detecting according to the prediction result, setting thresholds for all elements of the prediction result, comparing each element in the prediction result with the threshold, if the element value is greater than the threshold, classifying the elements into a change class, and if the element value isless than the threshold, classifying the elements into a non-change class. According to the method, data set features are learned under the condition of few samples, and the method has good predictioncapability for changed and unchanged areas in optical remote sensing images of different time phases in the same area.

Description

technical field [0001] The invention belongs to the technical field of image processing, and specifically relates to a method for detecting changes in optical remote sensing images based on the TernausNet twin neural network, a storage medium and a computing device, which can detect changes in optical remote sensing images with multiple temporal phases and multiple resolutions, and are used in urban planning and natural areas such as disaster assessment are of great interest. Background technique [0002] Detecting changes in the Earth's surface is becoming increasingly important for monitoring the environment and resources. With the development of remote sensing technology, surface information can be observed through remote sensing images. Therefore, changes in the Earth's surface can be identified by using image change detection techniques. Change detection is defined as the process of identifying changes in an object or phenomenon by observing it at different times. It...

AI Technical Summary

Problems solved by technology

However, supervised learning requires a large number of labeled samples for model training. It is difficult to achieve excellent performance in the case of poor label quality and insufficient quantity. In addition, the model is also susceptible to noise.

Images