Efficient hyperspectral image segmentation for biosecurity scanning using knowledge distillation from multi-head teacher

Phan, Minh Hieu; Phung, Son Lam; Luu, Khoa; Bouzerdoum, Abdesselam

doi:10.1016/j.neucom.2022.06.095

Cited by 9 publications

(1 citation statement)

References 43 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Since their inception, they have played a vital role in the success of various information mining and image interpretation tasks since their inception [1]. Contemporary examples include, but are not limited to, mapping (or classification) and monitoring applications of land use/land cover change [2], health care, weather, and climate forecasts [3], and Planetary exploration missions [4]. Imaging systems use sensors, which are sensitive to one or more wavelengths of an electromagnetic spectrum, to capture images for the visual representation of an area.…”

Section: Introductionmentioning

confidence: 99%

HSIS-Net: Hyperspectral Image Segmentation Using Multi-view Active Learning Based FCSN

2023

IJIES

View full text Add to dashboard Cite

Hyperspectral image (HSI) segmentation and classification is trending research in military and civil applications area. However, HSI classification is facing various challenges in analyzing spectral and spatial regions. In order to improve the performance of HSI classification models, segmentation is essential step. Therefore, this article is focused on implementation of unified HSI segmentation network (HSIS-Net) using active learning. Initially, HSI preprocessing operation is performed to normalize the spectral-spatial regions. Then, joint spatial-spectral boundary extraction operation is performed using spatial information divergence (SID) and spectral correlation mapper (SCM). Finally, segmentation of boundary estimated HSI bands is performed using multi-view active learning network based fully convolutional segmentation network (MAL-FCSN). The simulations revealed that the proposed HSIS-Net resulted in superior segmentation performance with segmentation accuracy (SA) of 0.999, and segmentation F1-score of 0.999 as compared to the existing HSI classification approaches for four publicly available HSI datasets.

show abstract

Section: Introductionmentioning

confidence: 99%