In 1985, the " Boston Globe " announced that, ten years ago, the Viking mission discovered of a representation of a human face on planet Mars ("A face on Mars!"). This mission consisted of two probes, the instrumentation and power supplies of which were designed to trace and identify the possible metabolic reactions related to the existence of a putative primitive life form. The results of these experiments generated long-lasting debates amongst the scientific community. Evidence for metabolic process (gas exchange, oxidation of organic compounds, carbon dioxide traces...) were obtained. However, attempts to detect organic molecules or microbial life forms in the Martian " soil " repeatedly yielded negative results.
Whereas the Mariner probes, in the '60s, only allowed to map a limited part of the Martian surface, the two Viking probes took ca. 60,000 pictures of the planet and its 2 satellites: Phobos and Deimos. The Boston Globe paper claimed that one of these pictures revealed the existence of a strange structure in the Cydonia-Mensae region (40.9N, 9.45W), resembling a humanoid face contemplating the outer space. The official reaction from the JPL ( Jet Propulsion Laboratory ) came shortly after: for the Pasadena Research Center, this " structure " merely resulted from a delusive association of light and shadow. A few years later, two NASA scientists (V. DiPietro & G. Molenaar) discovered a second picture (frame 70A13) of what was now called " The Face ". This picture had been taken 35 days after that published by the " Boston Globe " (frame 35A72). Both were high resolution pictures (ca. 50m/pixel) of " The Face ", illuminated by a low incidence light of a Martian afternoon (light from NNW; incidence angle = 20 to 30°). Today, four pictures of this structure are known. One of these is a low resolution document (250 m/pixel) showing " The Face " in the morning Martian light (from NE).
All these pictures rapidly triggered growing speculation on the theme " Vestiges and monuments of an extinguished E.T. civilization ". Since the publication of these documents, numerous studies on the " Martian abnormalities " have been carried out (R. Hoagland, V. DiPietro & G. Molenaar, R. Pozos...), and an interesting analytical approach presented by Pr. Mark Carlotto in a 1988 paper ("Digital Imagery Analysis of Unusual Martian Surface Features") published in "Applied Optics".
To analyze pictures of "The Face", Mark Carlotto used a technique known as Shape-From-Shading (SFS), the purpose of which being to obtain a 3D- reconstruction of the structure underlying " The Face " from the available, high-resolution pictures (35A72 and 70A13). To make a long story short, the overall procedure aims at defining a grid system over the location of interest, and to calculate the topography gradient field (the slopes), i.e. to orientate each and every facets of the grid in such a way that:
i) taking in account the direction of the light, the facets
are illuminated in a fashion consistent with the intensity of the
pixels of the original picture (hypothesis of a Lambertian
ii) the orientation of one facet is compatible with those of its neighboring counterparts (constraints of integrability or gradient field consistency).
A preliminary processing of these relatively " high background " pictures was performed using Wiener's filter...
The UFOCOM (note: Jeff, you're too shy!) modified the iterative algorithms presented by M. Carlotto to process simultaneously both pictures 35A72 and 70A13, the main idea being to get rid of the constraints presented above (under ii). The volumic reconstitution of the topography is carried out from the gradient field, using the least-square method (PCG algorithm: preconditionned conjugate gradients) because, although the constraint n°2, this one is quite inconsistent (if we integrate a slope over two different paths from a point to another, we get a slightly different altitude!). At the end those two steps, we finally have the relief of "The Face".
To validate the whole process, it is possible to illuminate the relief produced by the algorithms and check that this generates the same images than those seen on the original pictures taken by Viking. Indeed, this is exactly what is obtained, as shown in figure 1, using the illumination condition of picture 70A13. The photographed structure is ca. 415 meter-high, and cover a 2.0 x 2.5 km area. The modified color map allow the visualization of irregularities at the right side of the structure (these are partly in the shade on picture 35A72 since the light incidence was different).
From the topography, it is possible to simulate 3D-images, " viewed " from various locations, as shown below:
The overall " Face " aspect indeed persists using a wide range of observation positions, although, as stated by M. Carlotto, the SFS algorithm is not suitable to analyze the right side of the relief, partly hidden by the shade of higher structures (frame 35A72) or presenting limited contrast (frame 70A13).
Additional studies are in progress, especially to yield computerized pictures using other illumination conditions... We are also tempting to examine the intrinsic characteristics of the relief (without texture and light) and to evaluate its symetry around what could be considered as the nose bone. A geomorphological analysis of " The Face ", performed " in the light " of other neighboring objects and structures of this region located at the boundaries of the Northern plains (Acidalia Planitia), might provide us with useful complementary data.
Can we, however, evaluate the symmetry of this structure? The computerized image presented below shows what is expected from the illumination of the relief rendered symmetrical (symmetrization) with respect to the " nose bone ".
It shows that if the topography were more or less symmetric, large shaded sectors should indeed be visible and fully illuminated (shaded areas are circled). The relief is likely strongly unsymmetrical. On the original picture of "The Face" (frame 70A13), we see a tiny conical structure above the left "eye". This structure is contiguous to another one resembling an arm, which bridges the eye socket. This latter structure remains in the shade of the "eyebrow arch" on picture 35A72, is barely seen on picture 70A13, and more visible on the right part of the illuminated symmetrization.
December 10, 1997
(translation by Yves and Fabrice)
- Dr Mark Carlotto's site - Frédéric Pache's site (in French)
Back to "Sciences"
Back to the main menu