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

Synthetic aperture radar automatic target recognition (SAR ATR) methods fallshort with limited training data. In this letter, we propose a causalinterventional ATR method (CIATR) to formulate the problem of limited SAR datawhich helps us uncover the ever-elusive causalities among the key factors inATR, and thus pursue the desired causal effect without changing the imagingconditions. A structural causal model (SCM) is comprised using causal inferenceto help understand how imaging conditions acts as a confounder introducingspurious correlation when SAR data is limited. This spurious correlation amongSAR images and the predicted classes can be fundamentally tackled with theconventional backdoor adjustments. An effective implement of backdooradjustments is proposed by firstly using data augmentation withspatial-frequency domain hybrid transformation to estimate the potential effectof varying imaging conditions on SAR images. Then, a feature discriminationapproach with hybrid similarity measurement is introduced to measure andmitigate the structural and vector angle impacts of varying imaging conditionson the extracted features from SAR images. Thus, our CIATR can pursue the truecausality between SAR images and the corresponding classes even with limitedSAR data. Experiments and comparisons conducted on the moving and stationarytarget acquisition and recognition (MSTAR) and OpenSARship datasets have shownthe effectiveness of our method with limited SAR data.

Unveiling Causalities in SAR ATR: A Causal Interventional Approach for Limited Data