How HRSC works – nine light-sensitive sensor lines are installed in the HRSC, aligned transversely to the flight path. The camera moves across the planet, on board the Mars Express spacecraft. The surface is scanned line by line. To obtain stereo images, it is necessary to map the surface from a variety of viewing angles, which is why HRSC has more than one sensor line. In addition to what is referred to as the ‘nadir channel’, which scans the surface perpendicular to the spacecraft’s flight path, there are four lines directed forward along the flight path and four directed to the rear. Four of these eight lines are fitted with colour filters to produce colour images. This enables imaging of each point on the surface from nine different viewing angles. Computers are used to convert this data into elevation information.
HRSC is a one-of-a-kind camera system in planetary research. It is the first digital stereo camera capable of delivering multispectral information, as well as being equipped with an extremely high-resolution imaging channel that is essentially equivalent to a 'magnifying glass'. The camera acquires unique images of the Martian surface that form the basis for numerous scientific investigations.
How does a digital camera work?
Traditional film material has become uncommon in this digital age. Instead, electronic sensors, known as CCDs (charge-coupled devices) are used. This kind of sensor is also fitted to commercial digital cameras. Digital CCD cameras are now the norm in satellite imaging systems.
A digital image consists of extremely small picture elements referred to as pixels (from 'picture element'), with a brightness value assigned to each pixel. This brightness data is stored as information 'bits'. The number of bits used for each pixel determines how many brightness levels can be distinguished in the image. In an image with a gradation or radiometric resolution of 8 bits, 28 (= 256) different gradations can be distinguished. The number of pixels, multiplied by the radiometric resolution, produces the image size, which is commonly measured in MB (megabytes. one million bytes, in which one byte is equivalent to eight bits). A digital image can be processed without loss of information using a computer.
How a CCD camera works
A CCD sensor uses the photoelectric effect described by Einstein in 1905, where photons striking the surface of metals will cause electrons to be emitted, hence generating an electric charge. In a CCD sensor, the ‘shutter’ (equivalent to its counterpart in a standard single-lens reflex camera) is opened for a period defined by the user to allow light to fall onto the CCD. During this period, a voltage is generated that is equivalent to the number of photons striking the surface. CCD sensors have a displaced colour sensitivity compared to that of the human eye; they are less sensitive in the shorter-wavelength blue range, but they are able to detect longer-wavelength light – even into the near-infrared range to which the human eye is not sensitive.
After exposure, the induced voltages in the individual pixels are transmitted to a signal processor using a complex read-out procedure, which converts the voltages into brightness values. Put simply, the charges that build up in the individual picture elements due to the photoelectric effect are moved to the edge of the sensor chip, pixel by pixel. After signal amplification, they are passed to an analogue to digital converter (ADC) for translation into a computer-readable format. As soon as a line has been converted in this way, the charges are displaced one picture element further on to enable read out of the next line. The precision of the conversion in the ADC is expressed in bits. (Source: Daniel Henke (http://www.patania.de/galaxies/index.htm))
Advantages of a CCD camera
Technically, it is possible to use photographic film on a spacecraft by exposing the film to light, developing and scanning the images in space and then transmitting these scanned images back to Earth – a process that was actually used during the Apollo era. CCD images, however, offer a number of advantages – in addition to their digital format – compared to traditional analogue cameras that expose a photographic emulsion:
high sensitivity – CCD cameras have substantially higher sensitivity, even compared to specially treated photographic plates. In comparison, CCD cameras can achieve sensitivities of up to 40,000 ISO with moderate grain, while even hyper-sensitised photographic films reach the limits of their usefulness at around 1600 ISO, when the images become visibly grainy,
quantum precision – quantitative observations using photographic emulsions entail extremely laborious procedures. In contrast, it is in the nature of CCD sensors to assign a precise (only noise limited) brightness value to each pixel, hence enabling photometric measurements
no reciprocity effect – unlike photochemical film material, CCD sensors exhibit linear sensitivity up to saturation (full well capacity). This means that images remain richly contrasted even in extremely bright areas, for instance at the centre of galaxies
wide spectral range – CCDs are sensitive not only in the visible spectrum, but also in the near infrared, enabling sharper and more contrasted images in the near-IR range.
Surface sensors
Most digital cameras use two-dimensional surface sensors (Figure 1, left). When an image is acquired, a surface comprising x lines and y columns is exposed simultaneously. The number of pixels is the product of x and y. A standard value for a normal digital camera might be 3–5 mega pixels (for example, 2000 columns × 1200 lines = 2.4 million pixels or 2.4 ‘mega’ pixels). All the lines and columns of the sensor are exposed simultaneously.
Line sensors
Electronic sensors that do not consist of a two-dimensional arrangement of lines and columns, but of a one-dimensional array of sensor elements, i.e. pixels, are called line sensors. They scan the object line by line, which means that a single sensor line is exposed to obtain an image. A large number of individual lines are combined using a computer to create a two-dimensional image, with the number of lines being theoretically unlimited. In order to obtain a complete image, it is necessary for each line that is acquired to be displaced by precisely one line width compared to the previous image. Either this involves moving the object (as is the case in the fax machine, in which a sheet of paper is transported through the device and hence displaced relative to the imaging sensor line), or causing the sensor to move relative to a stationary object. In the case of a camera on a spacecraft that is tasked with scanning the surface of a planet, the camera will, of course, be moved with the direction of travel of the spacecraft itself (Figure 1, right).
Line sensors
Figure 1: (left) A two-dimensional CCD surface sensor consists of numerous picture elements (pixels), which are arranged in a regular array of lines and columns; (right) a CCD line sensor consists of numerous picture elements arranged as a single imaging line. Observers can only see interpretable images once many image lines have been combined. The number of image lines (in this case ‘n’) is theoretically unlimited.
The HRSC camera is equipped with CCD line sensors (Figure 1, right), rather than surface sensors (Figure 1, left). In this case, the individual pixels are not arranged as lines and columns, but as a single line (comprising a great number of pixels – the HRSC lines have 5184 pixels). All the pixels in this one sensor line are exposed simultaneously when an image is acquired. This means, in order to obtain a two-dimensional image, it is necessary to scan a large number of lines in succession and then combine them using a computer to create a composite.
An important factor in this respect is precise coordination between the travelling speed of the camera (that is, the spacecraft) over the surface, the clock rate at which the individual lines are scanned, and the exposure time. The separate lines need to fit together as precisely as possible, and the pixels contained in the image should ideally be square. To achieve this, the clock rate must be adjusted continuously to accommodate changes in speed due to the elliptical orbit (Kepler’s second law).
The HRSC can scan any number of lines in succession, theoretically at least, so the images could also be infinitely long. In reality, though, a number of factors limit the size, of which the data volume that the orbiter is able to transmit to Earth is the most important. Therefore, HRSC acquires images that typically comprise between 30,000 and 60,000 lines.
The SRC 'magnifying glass' (Super Resolution Channel)
Upgrading the HRSC for use on the Mars Express mission also included the installation of a second lens, designed to further enhance the camera’s resolution. This telescopic lens, with a focal length of approximately one metre, has the ability to produce images with a resolution of just two to three metres per pixel. This allows the identification of objects the size of a house (naturally, an object must appear in several pixels in order to be recognisable – one alone would be insufficient). The SRC sensor is a surface sensor with 1024×1024 pixels.
The imaging principle
How HRSC works – nine light-sensitive sensor lines are installed in the HRSC, aligned transversely to the flight path. The camera moves across the planet, on board the Mars Express spacecraft. The surface is scanned line by line. To obtain stereo images, it is necessary to map the surface from a variety of viewing angles, which is why HRSC has more than one sensor line. In addition to what is referred to as the nadir channel, which scans the surface perpendicular to the spacecraft’s flight path, there are four lines directed forward along the flight path and four directed to the rear. Four of these eight lines are fitted with colour filters to produce colour images. This enables imaging of each point on the surface from nine different viewing angles. Computers are used to convert this data into elevation information.
Figure 2a.
The camera on board the orbiter moves across the planet, scanning the surface line by line. A very similar principle is used in every fax machine; rollers transport a sheet of paper through the device, scanning each line. The only difference with the HRSC operating principle is that the camera (which represents the fax machine in this analogy) itself moves, and not the surface (the sheet of paper) (Figure 2a).
The SRC ‘magnifying glass’ can be activated to produce extremely high-resolution, detailed images within the line sensor’s field of view. This includes precise information on which pixel obtained by the HRSC matches the corresponding pixel in the SRC data. The HRSC therefore provides a spatial context for the SRC images.
HRSC also has the ability to acquire colour images, as it is equipped with four additional CCD lines with front-mounted colour filters (blue, green, red, near infrared). This means that the surface of Mars is scanned using nine different sensor lines.
