1
Return

Identifying diverse iron ores based on feature enhancement and deep learning classification with hyperspectral images

delete2026-08-10
delete0
PRE
AI
Z
Zhifang Xu
X
Xiaojun Qiao
J
Jinbao Jiang *
K
Kangning Li
Z
Zezheng Li
X
Xiaoyu Zu
DOI：10.1016/j.measurement.2026.122716delete
deleteOriginal
deleteOriginal request for help
deleteShare
deleteSave
Abstract

Abstract

En 中文
• Hybrid framework merging feature enhancement and deep learning for iron-ore grading. • Convert 1D spectra into 2D spatial dimension to enhance separability. • Multi-scale kernel weighted CNN designed to exploit the enhanced features. • More than 84.65% identification accuracy obtained across four mining areas.
Keywords:
Band selection
Deep learning
GADF transformation
Hyperspectral
Iron-ore identification
Feature enhancement

Journal

Measurement cover
Measurement
IF:
5.6
Papers:
1.9W
Citations:
5.4W

Organization

No organization information available
Cited Papers

Cited Papers

Citing Papers

Citing Papers