English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Tools annual

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Optical aerial images change detection is an important task in earthobservation and has been extensively investigated in the past few decades.Generally, the supervised change detection methods with superior performancerequire a large amount of labeled training data which is obtained by manualannotation with high cost. In this paper, we present a novel unsupervisedself-training algorithm (USTA) for optical aerial images change detection. Thetraditional method such as change vector analysis is used to generate thepseudo labels. We use these pseudo labels to train a well designedconvolutional neural network. The network is used as a teacher to classify theoriginal multitemporal images to generate another set of pseudo labels. Thentwo set of pseudo labels are used to jointly train a student network with thesame structure as the teacher. The final change detection result can beobtained by the trained student network. Besides, we design an image filter tocontrol the usage of change information in the pseudo labels in the trainingprocess of the network. The whole process of the algorithm is an unsupervisedprocess without manually marked labels. Experimental results on the realdatasets demonstrate competitive performance of our proposed method.

Unsupervised Self-training Algorithm Based on Deep Learning for Optical Aerial Images Change Detection