1 HYPERSPECTRAL REMOTE SENSING Contact mirza waqarseecs edu

1 HYPERSPECTRAL REMOTE SENSING Contact: mirza. waqar@seecs. edu. pk Mirza Muhammad Waqar

Contents 2 Introduction What is Hyperspectral Sensing Hyperspectral vs Multispectral Data Cube – A way to visualize the data The Ultimate Benefit Imaging Spectrometry Concept Spectrometry Imaging Spectrometry n-Dimensional Data

Overview 3 In previous lessons, you were introduced to extraction of thematic information from the spectral content of digital multiband imagery. You learned to manipulate various � Band combinations for display and analysis � Performed simple classification and change detection using classified and unclassified imagery. � Performed advance classification techniques

Cont… 4 The same concepts can hyperspectral imagery, but � Hyperspectral be applied to image analysis is much more complex than multispectral

Objectives 5 1. 2. 3. 4. Explain the fundamental principles of hyperspectral remote sensing Explain the georeferencing and radiometric calibration of hyperspectral data Identify standard hyperspectral mapping products and applications Explain the basic principles of hyperspectral image analysis

Introduction 6 Remote sensing involves examination of features observed in several regions of electromagnetic spectrum. � Multispectral remote sensing is based upon use of several broadly defined spectral regions � Hyperspectral remote sensing is based upon examination of many narrowly defined spectral channels.

Cont… 7 Hyperspectral remote sensing combines imaging and spectroscopy � Generally 100 to 200 or more narrow spectral bands (5 – 10 nanometers wide) � Compared to multispectral sensors with typically 4 to 12 spectral bands (70 – 100 nm wide) Generates large data sets Requires new / different analysis methods

What is Hyperspectral Sensing? 8

9

Multispectral vs Hyperspectral 10

11

Data Cube – A way to visualize the data 12

The Ultimate Benefit… 13 Diagnostic / identifying characteristics are lost in wide bands

Imaging Spectrometry Concept 14

Analysis Approach 15 Direct identification using diagnostic absorption and reflection features Comparison to laboratory and field measured spectra

Spectroscopy 16 Spectroscopy pertains to the dispersion of an object's light into its component colors (i. e. energies) � Spectroscopy can be used to detect individual absorption features due to specific chemical bonds in a solid, liquid, or gas � By performing this dissection and analysis of an object's light, one can infer the physical properties of that object (such as temperature, mass, luminosity and composition, etc. ) Continuous spectra Discrete spectra � � Emission line spectra Absorption line spectra

Continuous vs Discrete Spectra 17

Image Spectroscopy 18 Imaging spectroscopy is a new tool that can be used to map specific materials by detecting specific chemical bonds It is an excellent tool for environmental assessments, mineral mapping and exploration, vegetation species and health studies, general land management studies, and others

Cont… 19 Hyperspectral imaging is the simultaneous acquisition of images in many narrow, contiguous, spectral bands Each pixel in the remotely acquired scene has an associated spectrum similar to the spectra of the material / mineral obtained in the laboratory

20

Properties of a Spectrometer 21 Parameters that describe the capability of a spectrometer � Spectral range � Spectral bandwidth � Spectral sampling � Signal-to noise ratio FWHM = Full Width at Half Max

Spectral resolution – Sampling Interval 22 Spectral resolution = narrowest spectral feature � that can be resolved by a spectrometer (full width at half maximum FWHM) Spectral sampling interval = interval, in wavelength units, � between data points in the measured spectrum. Spectral bandwidth � used for spectral sampling interval)

23

Properties of a Spectrometer 24 Parameters that describe the capability of a spectrometer � Spectral range � Spectral bandwidth � Spectral sampling � Signal-to noise ratio

Goniometer 25

ASD Fieldspec Pro 26

ASD Fieldspec Pro 27

ASD Fieldspec Pro 28

GER 3700 29

PIMA II 30 PIMA (Portable Infrared Mineral Analyser) � Spectral range 1300 – 2500 nm � Spectral resolution 7 nm, interval 2 -4 nm Measurement is made in contact mode � Measurement time 20 -60 seconds Produces reflectance spectra � Spectra measured on a sample area of about 10 mm by 2 mm � Internal illumination (does not need solar illumination) � Internal reflectance standard, wavelength calibration uses internal target

PIMA II (cont’d) 31

Spectral Libraries 32 SLI’s are collections of spectra different surface materials. Often grouped by surface type (vegetation vs. soils vs. man-made materials etc. ) and sometimes by grain size fraction (influence on spectra).

Spectral Libraries (cont’d) 33 Usually measured under laboratory conditions with excellent spectrometers. Availability � Publicly available SLI’s are included in ENVI � ASTER speclib on the internet � Create own from field/lab measurements � Create own from image (point measurements or average of ROI)

n-Dimensional Data 34 Hyperspectral data (or spectra) can be thought of as points in an n-dimensional scatterplot. The data for a given pixel corresponds to a spectral reflectance for that given pixel. The distribution of the hyperspectral data in n-space can be used to estimate � the number of spectral endmembers and their pure spectral signatures and to help understand the spectral characteristics of the materials which make up that signature.

Vegetation Spectral Reflectance extracted from AVIRIS data 35

Questions & Discussion