Hyperspectral image target detection method based on meta learning and conjoined network

Abstract

The invention discloses a hyperspectral image target detection method based on meta learning and a conjoined network. The method comprises the following steps: firstly, designing a one-dimensional deep residual convolutional network to construct a three-channel deep residual convolutional conjoined network; secondly, designing a spectrum triplet loss function in order to enable the intra-class distance of the spectrums of the same type of pixels to be small and the inter-class distance of the spectrums of different types of pixels to be large; metatraining is carried out on a designed three-channel deep residual convolutional conjoined network by using a task constructed on a known labeled source domain hyperspectral data set, and the similarity and diversity between spectrums are learned in an Euclidean feature space. And updating the learned meta-knowledge by using a priori target pixel spectrum through a designed two-channel deep residual convolutional conjoined network so as to quickly adapt to a new detection task. Wherein the structure and the parameter of the one-dimensional deep residual convolutional network of each channel in the conjoined network are the same. And finally, processing the detection graph of the two-channel depth residual convolutional conjoined network by using guide graph filtering and morphological closed operation in combination with spatial information to obtain a final detection result graph.

Description

technical field [0001] The invention relates to the field of hyperspectral image target detection, in particular to a hyperspectral image target detection method based on meta-learning and Siamese network. Background technique [0002] Hyperspectral imagery is a three-dimensional cube image data containing rich spectral and spatial information captured by imaging spectrometers in hundreds of narrow and continuous bands to capture surface reflectance. Fine spectral features can effectively reflect the subtle characteristics of different substances. . Thanks to the rich spectral information, hyperspectral images have been applied and played an important role in civilian search and rescue, agricultural production, medical diagnosis, urban planning and other fields. Object detection has received more and more attention in these fields, and it has become an urgent need to accurately detect objects. [0003] Traditional hyperspectral image target detection algorithms, such as ta...

AI Technical Summary

Problems solved by technology

In order to expand the training samples, the subtraction of the same kind of pixel spectra in the known labeled hyperspectral data sets, and the subtraction of different types of pixel spectra, thereby expanding the number of training samples, but the subtraction of pixel spectra will The fine spectral features of the loss pixel will have a great impact on the accuracy of target detection

Images