The severe soil erosion on the Loess Plateau is the root of the Yellow River problem, and the mathematical model of soil erosion is the core of intelligent soil and water conservation. How to extract soil and water conservation measures accurately, quickly, and quantitatively is of great significance for developing the parameterization method of soil and water conservation measures in the soil erosion model and quantitatively revealing the influence of soil and water conservation measures allocation on runoff and sediment yield process. It is the direct demand to support the research and development of distributed soil erosion models, soil erosion prediction, and digital twin watershed construction. Based on analyzing the current high-resolution remote sensing image data and deeply learning the development frontier of artificial intelligence, this paper studies and explores the technology and method of intelligent extraction of terrace measures underpinned by an artificial neural network algorithm model under deep learning. The main research results are as follows: (1) The current deep learning theory development and model structure are deeply studied. (2) The intelligent extraction algorithm of high-resolution remote sensing images based on depth learning is developed. (3) Accuracy analysis of the model results are carried out. The research results of the project will provide basic data for regional soil erosion prevention and control, and at the same time, provide technical support for high-precision parameter acquisition and calibration of distributed soil erosion model, which has important theoretical practice and application value.
VHR imagery change detection is one of research hotspots and difficulties in the field of remote sensing. However, the traditional remote sensing image change detection method is a waste of time and energy and low efficiency. In recent years, deep learning approaches in remote sensing image change detection verified feasible and save time to improve efficiency. A UNet change detection method based on aggregation residuals and attention mechanism is proposed, using prior knowledge of deep learning. The UNet model is used as the basic model, and the aggregation residual module is introduced in the up-down sampling stage, which can fully extract the feature information of the image. The weight of each component in the feature graph can be adjusted by adding attention module in the jump connection layer. In the process of experiment based on the model parameters are reasonable and effective set of data sets to Longnan remote sensing image change detection, and the experimental results showing that compared with the traditional deep learning semantic segmentation method, this article methods F1 value of 0.873, the generated change detection figure closer to label figure, higher accuracy, shorter.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.