In recent years, followed by the development of synthetic aperture radar imagery, target feature extraction and detection becomes a research hotspot. To facilitate SAR target recognition, a simple but powerful method, named by principal component analysis net (PCANet), is introduced for an accurate target feature extraction. Compared with conventional methods, shorter training time and fewer training samples are required in the proposed method. Experiment conducted on the canonical moving and stationary target acquisition and recognition (MSTAR) database is executed to examine the proposed method after the relevant features are extracted by the PCANet. Finally, support vector machine is adopted for the target recognition. It can be noted from the experimental results, the highest accuracy of the recognition can achieve 99.8% when using half of the training samples, which can validate the effectiveness of our proposed method.
CITATION STYLE
Li, P., Li, J., Yin, W., & Yang, L. (2020). Target Feature Extraction and Recognition of SAR Images Based on PCANet. In Lecture Notes in Electrical Engineering (Vol. 582, pp. 393–400). Springer. https://doi.org/10.1007/978-981-15-0474-7_37
Mendeley helps you to discover research relevant for your work.