Hyperspectral small sample classification method based on lightweight network and semi-supervised clustering

Abstract

The invention relates to a hyperspectral small sample classification method based on a lightweight network and semi-supervised clustering. A lightweight network model is constructed by using a Point-wise convolution kernel, a Depth-wise convolution kernel and double loss. The Point-wise convolution kernel and the Depth-wise convolution kernel can greatly reduce the number of parameters, and reducethe demand for training samples in the network training process. The depth feature space can be more separable through the double-loss strategy, and classification and clustering in the depth featurespace are better facilitated. In addition, the semi-supervised approximate order clustering algorithm can select more self-confident pseudo tags, and more favorable conditions are provided for improving the network training effect. According to the method, autonomous extraction and high-precision classification of hyperspectral image depth features and label data are realized under the conditionof small samples.

Description

technical field [0001] The invention relates to a small-sample hyperspectral classification method, which is a collaborative learning based on lightweight network and semi-supervised clustering, and belongs to the field of image processing. Background technique [0002] Hyperspectral remote sensing images have high spectral resolution, multiple imaging bands, and large amounts of information, and are widely used in remote sensing applications. Hyperspectral image classification technology is a very important content in hyperspectral image processing technology. It mainly includes two parts: feature extraction and classification. Among them, features are extracted from the original hyperspectral image. This step has a great impact on the classification accuracy of hyperspectral images: Classification The strong separability of the features can greatly improve the classification accuracy; on the contrary, the classification features with poor separability will significantly re...

AI Technical Summary

Problems solved by technology

[0004] However, the existing methods for extracting spatial spectral features of hyperspectral images using deep models require a large number of training samples to train the network, and it is very difficult to mark the collected hyperspectral images in practice. Field surveys require a lot of manpower and material resources, and require consumes a lot of time

Images