Polarization feature selection and classification method based on object-oriented random forest

Abstract

The invention discloses a polarization feature selection and classification method based on an object-oriented random forest, which solves the problem of feature selection and image classification when a plurality of multi-polarization features participate in classification. The method adopts the object-oriented method to segment a feature set in multi-scale, performs random forest modeling on thesegmented sample objects, calculates the importance of each feature, and optimizes the feature set by using the forward selection algorithm. The invention adopts the object-oriented random forest method to improve the model training efficiency and the classification accuracy. The sequence forward selection algorithm and the highest precision iterative termination condition are used to construct the optimal feature subset to avoid falling into the local optimal solution. The algorithm can improve the classification accuracy, and provide quantitative reference for the reasonable optimization offeature set.

Description

technical field [0001] The invention belongs to the field of image processing, and mainly relates to feature extraction and classification of polarimetric SAR images, in particular to an object-oriented random forest-based polarization feature selection and classification method. Background technique [0002] In recent years, polarimetric SAR images have been increasingly used in the extraction of surface information. The method based on target decomposition is an important means to analyze and extract information from polarimetric SAR images. For a more complex landform environment, it is difficult to effectively distinguish all surface types based on a certain polarization decomposition method or a few characteristic parameters. Therefore, integrating multiple polarization decomposition algorithms and combining multiple polarization characteristic parameters has become an effective way to solve this problem. However, in practical applications, there may be interdependenc...

AI Technical Summary

Problems solved by technology

However, in practical applications, there may be interdependence between too many feature parameters, which can easily lead to a series of problems, such as: analyzing features and training models takes too long, causing "dimension disaster", complex models, and increased calculations. Large, the classification accuracy decreases, etc.

Images