Onorbit MTF assessment of satellite cameras

1
On-orbit MTF assessment of satellite cameras

• Dominique Léger (ONERA)
• Françoise Viallefont (ONERA)
• Philippe Déliot (ONERA)
• Christophe Valorge (CNES)

2
Introduction

• Objective
• assessment of SPOT camera MTF
• to verify cameras requirements
• to compare in-flight and ground measurements
• to obtain accurate values to adjust deconvolution
  filters (SPOT5 THR)
• Need to focus camera before MTF assessment
• due to possible slight defocus
• vibrations during launch
• transition from air to vacuum

3
SPOT family Overview

• SPOT1,2,3
• HRV cameras
• Pa (10m) B1, B2, B3 (20m)
• SPOT4
• HRVIR cameras
• M (10m) B1, B2, B3, B4 (20m)
• Vegetation camera
• B0, B2, B3, B4(1km)
• SPOT5
• HRG cameras
• HM (5m) B1, B2, B3 (10m), B4 (20m)
• THR (2,5m)
• HRS cameras (10 m)
• Vegetation camera
• B0, B2, B3, B4 (1km)

SPOT5
SPOT4
SPOT2
4
Refocusing SPOT cameras

• Method
• Both cameras image the same landscape
• One is used as a reference
• Focusing mechanism of the other is moved
• Calculation of the ratio of image spectra
• integration in band 0.25 fs - 0.35 fs
• calculations in row and column directions
• result is a function of position p of mechanism
• The curve looks like a parabola
• a defocus model is fitted on measurements
• the vertex gives the best focus
• Calculations vs field area
• center and edges (SPOT5)

5
Refocusing SPOT cameras

• Refocusing operation sequence (SPOT5 HRG)
• Before launch, the cameras are set on best vacuum
  mean focus p0
• First stage slight defocusing around p0
• p0-8, p08, p0 (10 mm)
• mechanism validation
• first focus estimation p1
• Second stage sufficient defocusing to overpass
  p1
• Final estimation of best focus
• row-wise and columnwise ? astigmatism
• field center and field edges
• Setting the focus to best mean position

6
Refocusing SPOT cameras

• Results of HRG1 refocusing operations (First
  stage)
• Vertex outside measurement points
• Second stage needed

7
Refocusing SPOT cameras

• Results of HRG1 refocusing operations (second
  stage)
• Best focus (field center) p0-13
• Astigmatism -7
• (one focusing step 1.2 mm)

8
Refocusing SPOT cameras

• Best focus and astigmatism vs field area
• (with respect to p0)
• Final focusing
• HRG1 p0-12
• HRG2 p0-7

9
Relative MTF measurement method

• Both cameras image the same landscape (with and
  without shift)
• Landscapes with a large frequency content (e.g.
  big cities)
• Three kind of imaging
• 1 HRG1
• HRG2
• 2 HRG1
• HRG2
• 3 HRG1
• HRG2
• 1 ? Frequency content comparison between
  homologous areas
• Field centers, field edges
• 1 2 (3) ? Frequency content comparison in the
  field of one instrument
• e.g. 12 ? HRG1 left edge versus HRG1 center

L C R
10
Absolute MTF measurement methods

• Overview of methods from SPOT1 to SPOT5
• Visual assessment
• HRV cameras SPOT1, SPOT2, SPOT3
• Point source method
• SPOT3, SPOT4, SPOT5
• Step edge method
• Natural target SPOT4 HRVIR SPOT5 HRS
• Artificial target SPOT5 HRG
• Bi-resolution
• SPOT4 HRVIR (vs airborne) SPOT4 VGT (vs HRVIR)
• Periodic target
• SPOT5 HRG

11
MTF measurement methods Visual assessment

• SPOT1, SPOT2, SPOT3 HRV cameras
• Only panchromatic band
• Aerial imagery of urban sites
• 20 sites chosen in the south of France
• Simulation of the corresponding satellite imagery
• For each site, images with decreasing MTF are
  simulated
• The whole set of images is called MTF catalog
• In-flight, visual comparison of actual and
  simulated images
• MTF of the catalog image nearest to the actual
  image gives a rough assessment of the in-flight
  MTF

12
MTF measurement methods Point source

• SPOT3 HRV, SPOT4 HRVIR, SPOT5 HRG
• Pa and XS bands
• Image of a spotlight aimed at the satellite
• In SPOT5 THR mode, the PSF is sufficiently
  sampled
• MTF is obtained by Fourier transform of the PSF
• In other modes, two ways to overcome PSF
  undersampling
• To use a MTF model
• To combine several images to rebuild sufficiently
  sampled image
• or to use several spotlights

13
MTF measurement methods Point source

• Unique point source method
• Integrating point image (row-wise or columnwise)
• 1D problem
• Reference LSF FT(parametric 1D MTF model)
• Two parameters MTF and phase (versus sampling
  grid)
• Matching LSF samples with reference
• ? Value of the MTF parameter
• Corresponding MTF 1D in-flight MTF
• ? Value of the phase parameter
• Stability of MTF
• Possibility to mix the various sets of LSF
  samples
• If different phase parameters

14
MTF measurement methods Point source

• Two point source method
• Simplified version of point source array
• Integrating point image (row-wise or columnwise)
• 1D problem
• Hypothesis MTF is negligible beyond frequency
  sampling
• ? Two points are sufficient
• Experiment with two spotlights (SPOT5)

15
MTF measurement methods Point source
Spotlights on a grassy uniform area
Xe lamp 3kW
Xe lamp 1kW
16
MTF measurement methods Point source
17
MTF measurement methods step edge

• Step edge method
• Image of a target (artificial or natural) with a
  sharp transition between dark and bright area
• With a slight edge inclination, we can interleave
  successive rows (or columns) to rebuild a
  sufficiently sampled response to Heaviside
  function
• Again, this is not necessary with THR mode
• Modulus of ratio of FT (edge response) to FT
  (edge) in-flight MTF
• Two kinds of edge
• Natural edge agricultural fields
• Difficulty to find a good one and to validate it
• Artificial edge
• A checkerboard target has been laid out
  (Salon-de-Provence in south of France)
• 60 x 60 m

18
MTF measurement methods Natural step edge

• Fields near Phoenix (SPOT5 HRS2 10/06/02)
• Example of an almost horizontal edge
• along the track measurement

19
MTF measurement methods Natural step edge

• Example of result with HRS
• Method improvement MTF model is fitted on MTF
  curve

20
MTF measurement methods Artificial edge target

• Salon-de-Provence target (SPOT5 HRG1 26/07/02)

21
MTF measurement methods Bi-resolution

• Principle
• Same landscape acquired with two spatial
  resolutions (same spectral band)
• High resolution image reference
• Low resolution image sensor under assessment
• In-flight MTF Modulus of ratio of FT (LR image)
  to FT (HR image)
• Two situations
• Satellite image versus aerial image
• Attempt with SPOT4 HRVIR
• Both sensors on the same satellite
• Attempt with SPOT4 VGT1 versus HRVIR

22
MTF measurement methods Periodic target

• Opportunity to acquire Stennis Space Center
  radial target with SPOT5

HM (5m)
THR (2.5m)
23
MTF measurement methods Comparison

• Comparison of SPOT5 HRG1 MTF measurements
• Direction Rows Columns Diagonal
• Spotlight 0.35 0.32 0.15
• Step edge 0.33 0.30
• Radial target 0.38 0.18
• Ground 0.31 0.36
• Specification 0.25 0.23
• Close results for different methods
• In-flight and ground measurements similar and
  better than specification

24
MTF measurement Comments on best methods

• Artificial step edge
• Well suited to high-resolution satellites (GSD lt
  5 m Salon-de-Provence target)
• Target building and maintenance expensive
• Only two measurement directions
• Spotlight
• Suitable to GSD up to 30m
• No orientation constraint
• Needs a team on ground
• Bi-resolution
• Attractive with different GSD cameras aboard the
  same satellite
• Radial target
• Interest of visual assessment in addition to MTF
  measurements
• No orientation constraint
• Target building and maintenance expensive