On December 16, 2011, Pleiades-1A was launched successfully in orbit via a Russian Soyuz ST rocket out of French Guiana. Pleiades-1A is the first satellite of the Pleiades constellation. The Pleiades constellation will be composed of two very high-resolution optical Earth-imaging satellites on a Sun-synchronous orbit at 694km. With Pleiades 1A in orbit and Pleiades 1B scheduled for launch in 2013, the Pleiades constellation will provide very high- resolution optical products in record time, offering daily revisit to any point on the globe. The satellite consists of a panchromatic band with resolution of 0.7m in nadir viewing and four spectral bands (blue, green, red and near infrared) with a resolution of 2.8m in nadir viewing. Image correction accuracy within 1m and 10m can be obtained with and without ground control points (GCPs), respectively.
Pleiades image data is distributed in either JPEG or TIFF format. There are two levels of processing products available: primary and ortho. The primary product is closest to the image acquired by the sensor and restores perfect collection conditions. The primary product must be used for orthorectification and DEM extraction in OrthoEngine. The ortho product is georeferenced imagery in Earth geometry, corrected from off-nadir acquisition and terrain effects. The user can order bundled (panchromatic and multispectral) product or PMS (pan- sharpened) product.
The following is a brief tutorial showing a step by step procedure for orthorectifying and pansharpening Pleiades 1A imagery using Geomatica OrthoEngine 2013. A sample Pleiades primary data set consisting of panchromatic, multispectral, PMS and tri-stereoscopy images of Melbourne, Australia, was provided by Astrium Geo Services.
1. Open the Geomatica 2013 OrthoEngine application
2. In OrthoEngine click File > New
3. Give your project a Filename, Name and Description
a) Select Optical Satellite Modeling as the Math Modeling Method
b) Select Rational Function (Extract from image) under Options
c) Click OK
4. Input the appropriate Output projection and GCP projection information for your project
Similar to most high resolution satellites, Pleiades panchromatic and multispectral data provide the opportunity to create 0.7m multispectral pan-sharpened images. Although the user can purchase Pleiades pan-sharpened (PMS) product directly, in some cases the user may want to apply their own pan-sharpening.
It is always preferable to perform the pan-sharpening process before geometric correction if a pan-sharpened orthorectified image is desired. This method works for most areas with gentle terrain.
Performing pan-sharpening after geometric correction often results in small misalignments between the ortho data due to the accuracy of GCPs and DEMs used in the orthorectification process.
5. Click on Utilities > Merge/Pansharp Multispectral Image…
6. Select the Multispectral images in the dataset. Select the DIM_PHR1A_MS_.XML file
a) Select the Panchromatic image. Select the DIM_PHR1A_P_.XML file
b) Select the output filename and location for the output PIX file
c) Click Pansharp
d) When the pansharp process is complete click Yes when asked to add the scene into the project
Fig 1a: Pleiades multispectral image of Melbourne, Australia
Fig 1b: Pleiades panchromatic image of Melbourne, Australia
About the output
PCI’s Geomatica award winning pansharpening algorithm (Co-developed with UNB) is capable of taking a multispectral image and its coincident/co-registered panchromatic pair and produce outstanding pansharpened results.
The output is a beautifully sharp high resolution multispectral image with the resolution of the panchromatic image.
The pansharpened process takes a high resolution panchromatic image and a lower resolution multispectral image and produces a high resolution multispectral image (same resolution of panchromatic image) with very sharp edges.
In order to leverage the Pleiades images for applications such as GIS, it is necessary to orthorectify the images. A geometric model, ground control points (GCPs) and a digital elevation model (DEM) is required. The Rational Function Method (RFM) has been the most popular geometric correction method in orthorectifying high resolution images. This method uses the RPCs provided with the satellite data to perform orthorectification. Since the Pleiades primary product is provided with RPCs, RFM can be used to orthorectify the data.
7. Select Data Input as the Processing step
a) Click Open a new or existing image
b) Click 'New Image'
c) Navigate to the location of the data. Select the DIM_PHR1A_.xml file if you wish to use the raw imagery. If you pansharpened the imagery as shown in the steps above, the pansharpened PIX file will already be added to the project. Skip to step 8.
d) Select Yes when asked if you want to import the data file to a PIX file for optimized processing
e) Select a file name and location for your output PIX file
f) Select Yes when asked if you want to create overviews now
8. To view the output image select the image and click Open
9. In the OrthoEngine toolbar select GCP/TP Collection as the Processing step
a) Select Collect GCP’s Manually
Note: GCP’s can also be collected automatically using collect GCP’s automatically. When using a 0 RPC adjustment order a minimum of 1 GCP should be collected per image. When using a 1st RPC adjustment order a minimum of 3 GCP’s should be collected per image. These 3 GCP’s should be collected near the boundary of the image as the results are only accurate within the area bounded by the GCP’s.
If you are manually collecting GCP’s using Toutin’s Model, a minimum of 6 GCP’s are required.
b) Collect GCP’s for the project using manual entry, from geocoded images, vectors, chip databases, digitizing tablet or a text file. You can also collect tie points, if you need to pull multiple scenes together.
10. In the OrthoEngine toolbar select Model Calculations as the Processing step Click Perform block adjustment
11. In the OrthoEngine toolbar select Ortho Generation as the Processing step
a) Click Schedule ortho generation
b) Move the Available images to the Images to process window
c) Select an output file name and location for the ortho image
d) Select a DEM file
e) Select a Sampling interval
f) Change the Resampling method to Cubic
g) Click Generate Orthos
12. To view the completed ortho select File > Image View on the OrthoEngine Toolbar