Insect Biological Parameter Estimation Based on the Invariant Target Parameters of the Scattering Matrix

Cheng Hu*, Weidong Li, Rui Wang, Teng Long, Changjiang Liu, V. Alistair Drake

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

34 Citations (Scopus)

Abstract

For radar observations, invariant target parameters extracted from the scattering matrix (SM) provide information about the target's geometry and composition. By studying the invariant target parameters of a small sample of insects, it is shown that the two eigenvalues and the determinant of the Graves power matrix are strongly correlated with insect body length and mass. Therefore, two methods are proposed to estimate the body length and mass from the eigenvalues or the determinant. The two eigenvalues identify the maximum of the insect polarization pattern and the perpendicular to the maximum direction, while the determinant is the product of these two values. A sample of 207 insect specimens measured at X-band in the laboratory rigs is used to determine the relationships between SM parameters and body lengths and masses. The results show that the length and mass can be estimated with a good performance and without an initial classification stage. In addition, the use of the determinant of the Graves power matrix is shown to provide an improved, SM-based, method of determining insect orientation and to solve the 90° orientation-extraction error problem that arises at X-band with very large insects.

Original languageEnglish
Article number8680761
Pages (from-to)6212-6225
Number of pages14
JournalIEEE Transactions on Geoscience and Remote Sensing
Volume57
Issue number8
DOIs
Publication statusPublished - Aug 2019

Keywords

  • Body length
  • entomological radar
  • insect
  • invariant target parameter
  • mass
  • orientation

Fingerprint

Dive into the research topics of 'Insect Biological Parameter Estimation Based on the Invariant Target Parameters of the Scattering Matrix'. Together they form a unique fingerprint.

Cite this