A Double-Strategy-Check Active Learning Algorithm for Hyperspectral Image Classification

Abstract

Applying limited labeled samples to improve classification results is a challenge in hyperspectral images. Active Learning (AL) and Semisupervised Learning (SSL) are two promising techniques to achieve this challenge. Combining AL with SSL is an excellent idea for hyperspectral image classification. The traditional method, such as the Collaborative Active and Semisupervised Learning algorithm (CASSL), may introduce many incorrect pseudolabels and shows premature convergence. To overcome these drawbacks, a novel framework named Double-Strategy-Check Collaborative Active and Semisupervised Learning (DSC-CASSL) is proposed in this paper. This framework combines two different AL algorithms and SSL in a collaborative mode. The double-strategy verification can gradually improve the pseudolabeling accuracy and facilitate SSL. We evaluate the performance of DSC-CASSL on four hyperspectral data sets and compare it with that of four hyperspectral image classification methods. Our results suggest that DSC-CASSL leads to consistent improvement for hyperspectral image classification.

A New Framework for Hyperspectral Image Classification Using Multiple Semisupervised Collaborative Classification Algorithm

“…As an indispensable part in MSS-MVSL, we present an effective framework named combination of unsupervised secondary screening algorithm (CUSS) for joint selecting samples. The idea comes from weighted combination of ensemble learning classifier [65], [66]. The combination of the model is extended to the combination of the strategy so as to achieve the fusion of multiple strategies in a single framework and achieves higher stability.…”

Deep Learning-Based Intelligent Image Recognition and Its Applications in Financial Technology Services

“…The financial technology service industry involves a large number of image and text information processing tasks, such as customer identity verification, loan approval, and credit evaluation. By automatically processing images and text information, financial institutions can greatly reduce labor costs, improve overall operational efficiency, and help financial institutions identify and predict risks more accurately, thereby improving risk management capabilities [12][13][14][15][16][17][18][19][20][21][22][23]. Therefore, the application of deep learning technology will bring tremendous changes to the financial technology service industry.…”