The inclusion of a built-in flash in point-and-shoot cameras and mobile devices is, in my opinion, one of the worst things that happened to photography. The flash leads to harsh and unattractive lighting and shadows, and usually does more harm than good. In a forensic setting, however, these built-in flashes can be quite informative.
To begin, it is interesting to note that different camera manufacturers place the flash in different positions relative to the camera lens. Shown to the right are three such examples. For the Sony Cyber-shot the flash is to the left of the camera lens, for the Nikon Coolpix it is above, and for the Canon IXUS it is to the right. In addition, the position of the flash on different models from the same manufacturer can vary in terms of their distance relative to the lens, both horizontally and vertically.
From a consumer’s perspective, the location of the flash on a point-and-shoot or mobile device is of relatively little importance. From a forensic analyst’s perspective, however, the flash’s location can be informative.
Shown below are two images of a computer generated person standing in front of a wall. The image on the left was rendered with a camera in which the flash was positioned similar to that of the Canon IXUS. The image on the right was rendered with the flash positioned similar to the Sony Power-shot. You can see that the flash’s position affects the location of the shadows cast onto the wall. When the camera is pointing at the subject and the flash is positioned to the right (as viewed facing the camera), the shadows are cast rightward (as viewed facing the subject), and when the flash is positioned to the left, the shadows are cast leftward. Similarly, if the flash were positioned above or below the lens, then the cast shadows would be shifted downward or upward, respectively.
The reason for this shift is a matter of simple geometry. Shown below is a top view of the above rendered scene. Also shown is a camera with the flash positioned on either side of the lens. I’ve drawn yellow lines that connect either side of the head to the corresponding points on the cast shadow. As you can see, the shift in the location of the flash causes a corresponding shift in the cast shadow.
Of course, if the camera is not pointing at the subject, then the differences in the shift of the shadows may not be as apparent. As we will see next week, however, this analysis can be made more quantitative — we will be able to determine if the location of the flash on a specific camera make and model is consistent with the shadows and apparent scene geometry. I’ll also describe how the very nature of how a small localized flash emits light, which often gives rise to such unattractive photos, can be useful to forensic analysts.
This basic analysis is obviously not highly distinctive since many cameras will have a similarly positioned flash. It has been my experience, however, that even narrowly defined forensic tools such as this one are still useful.