Word of the Month – Focal Plane

In this edition of WOM, we explore the internal mechanics of high resolution satellites, specifically the focal plane. The focal plane is a crucial component of all satellites - as well as aerial and handheld digital cameras for that matter - as it contains the array of sensors which are measuring the reflectance of light (i.e. photons) from the surface of our planet. We conclude this edition of WOM with an interesting graphical illustration of the focal plane and its impact on high resolution satellite imagery.

To begin this discussion, it is important to note that the high resolution satellites eMap works with are pushbroom satellites. This means that their optical sensors are aligned in a near straight-line configuration so that high resolution imagery is obtained by ‘turning' the sensors on as the satellite passes over the Earth's surface from hundreds of miles up in space. The focal plane is the generic term for the stable platform which contains these linear arrays of sensors.

Now let's explore the focal plane of one of the original high resolution sensors, QuickBird. Launched in 2001, QuickBird features a single panchromatic band that is four times the resolution of its four multispectral bands (i.e. blue, green, red and near-infrared). Its focal plane consists of six panchromatic and six multispectral sensors arrays - referred to as staggered detector chip assemblies (DCAs) – which are slightly offset from each other as can be seen in Figure 1 below.

While Figure 1 is not drawn to scale, what it does show is that the DCAs for each band are slightly offset so that each band gets a view of the same piece of ground at a slightly different time (i.e. fractions of milliseconds). Given that each DCA images the same piece of ground at slightly different times, radiometric calibrations are applied to QuickBird imagery to produce a visually appealing and spectrally sound product. This DCA offset has other implications that we will explore later. As a final note on the QuickBird focal plane – and to give you sense of its complexity – there are 27,568 individual pan sensors and 6,892 multispectral sensors (both figures ignore overlapping sensors) per band or 27,568 total MS sensors; for a grand total of more than 55,100 tiny sensors packed into 12 DCAs.

Figure 1 – A graphical representation of the QuickBird focal plane, not drawn to scale. (Source: DigitalGlobe)



Launched in 2009, WorldView-2 is the most complex high resolution satellite ever constructed as it features eight multispectral plus one panchromatic band. Given the extra spectral bands, the focal plane of WorldView-2 is much more complex than that of QuickBird with additional DCAs and thus individual sensors. Its focal plane, as illustrated in Figure 2, consists of 50 staggered panchromatic DCAs and then two sets of 10 staggered multispectral DCAs on either side so that each DCA images the same piece of ground at slightly different times as with QuickBird.

One key difference between the focal plane of QuickBird and that of WorldView-2 is its ability to support bi-directional scanning so that imagery can be collected in either a forward or reverse ‘flicking’ motion of the satellite. The focal plane of WorldView-2 consist of more than 106,400 individual sensors so you can imagine the enormity of the radiometric calibration of this data versus that collected by QuickBird!

Figure 2 – A graphical representation of the WorldView-2 focal plane, not drawn to scale. The MS1 DCAs contain the ‘traditional’ blue, green, red and near-infrared 1 bands; while the MS2 DCAs contain the new coastal, yellow, red edge and near-infrared 2 bands. (Source: DigitalGlobe)



And finally we turn our attention to a visual analysis of the impact of staggered DCAs on a satellite’s focal plane. As stated above, each DCA on the focal plane images the same piece of ground at a slightly different time. While this time difference is much less than a millisecond it can have a dramatic affect when rapidly moving objects are imaged by a high resolution satellite.

Take for example an airplane which can travel at speeds over 500 miles per hour or ~730 feet per second. In Figure 3 below, you can see the visual impacts of an airplane traveling at these speeds as staggered DCAs image it from the focal plane of WorldView-2.

Figure 3 – notice the black and white appearing airplane on the western edge of this data collected by WorldView-2 on August 13, 2010 over Renton, WA. In front of the black-and-white airplane is a rainbow-colored airplane. (Image courtesy: DigitalGlobe)



Now let’s focus on four bands of WorldView-2 to get a better understanding of what is causing this airplane to appear in two locations at once. In Figure 4 below, you will see a short movie of four bands stacked upon each other so that the upper left corner is the exact same location in each image. What you will see is that the plane flies from north to south through the data as the each individual band (i.e. pan, blue, green and red) images the same location on the planet at a slightly different time!

Figure 4 – a short movie showing the airplane from Figure 3 moving across the focal plane of WorldView-2. (Image courtesy: DigitalGlobe)

• Radiometric Use of QuickBird – Technical Note
• Radiometric Use of WorldView-2 – Technical Note

Follow Us:

       

Copyright (C) 2011 eMap International. All rights reserved.

Email design by Ashley Lojko - Online Marketing For Small Business