"I'm building (USS) Barney (DD-149)...(looking) at the pics for (her) over at navsource.org...the gun tubs on the forward edge of the midships deck house...are pretty clear, but what's in them sure as heck ain't...(I'm) still trying to figure (that) out..."
Now Mike implies that it goes without saying - as indeed it does - that reference pics are the first place to search for answers. However, whenever pics are in digital form (i.e., scanned onto a computer) their examination can be powerfully boosted - easily, using image enhancement software that resides on almost every PC - vastly beyond what many may realize:
Depicting USS Barney in March of 1945, this is the fit I assume Mike wants to depict (or, at least the one he should, as it's got the coolest camo pattern! ) Seriously though, the first instinct of most (myself included) will be to enlarge a picture - just like taking a magnifying glass to a hardcopy print - however, in the digital realm, before magnifying it is actually best, IMHO, to first OPTIMIZE THE CONTRAST, to best show the particular feature(s) of interest. This in fact is the digital equivalent of brightly illuminating the photo - a "real world" first step that likewise goes without saying. In different software, it is variously named "contrast stretch", "histogram stretch", or simply "contrast and brightness" (or "lightness", etc.) adjustment, and is easily practiced until you can quickly get near-miraculous improvement in most pics. Not actual miracles, however - note in the enhanced version (bottom), some areas (particularly towards the stern) became completely washed-out ("oversaturated") - so, save your "gun tubs stretched" version as a separate copy, thus preserving the original for further use. Note, If a black-and-white original is presented (scanned) in color, as this one (top), then stretching the contrast of all three ("RGB") colors (and combining the result into "grayscale", if that's not too complex, in your software) will improve your result (its "precision", or "signal-to-noise" ratio - theoretically, by up to 3x). Once the best ("optimal", or "optimized") contrast has been achieved, apply an EDGE ENHANCEMENT - also variously called "sharpening", "unsharp masking" or (more technically) "convolution filtering", to take particular advantage of the new stretch - to sharpen features, as the names imply. Again, practice will reveal the maximum useful amount to apply - beyond which the image will become too grainy (and actually more difficult to interpret).
In the above, despite applying both steps (bottom), note the tub areas still do not reveal more detail - simply too dark (under-exposed) in the original - but this will almost never stay the case for all your pics:
Here seen on the same date from the port side, now the same digital enhancement (bottom) clearly reveals each tub to contain a large, dark gray object - the right size for a gun mount, possibly under a tarp - and Mike had (one way or another) arrived at this same interpretation, writing "...what's in there looks just a bit too big to be Mother Deuce" - whatever the hell that's supposed to mean! (Again, I kid.)
However, we can seriously discover what was in these tubs - and more - from a third reference pic, and demonstrating yet a third enhancement step:
Click on Image to Enlarge
Don't trust your (or my) own eyes? A further edge-enhancement (bottom-right) - variously called "edge detection", "edge tracing" or (more artistically) "pencil-" or "charcoal drawing" and "-outlines" - all simply names for another convolution filter, with "jacked up" settings - can actually draw (some) outlines automatically. The darker (sometimes lighter) the line, the more "sure" the computer is that it's really seeing a hard edge, the great advantage of this being that the result is truly objective - regardless of the operator's eyes, or prejudice(s) - especially useful whenever a feature is a "difficult call". In this case (rectangular inset, bottom-right), note the computer is quite sure of the crewman, as well as the gun (including its breach end), while the medium-dark gray of the squared-off gun shield indicates the machine is pretty "certain" about that, as well.
At this point, we can be pretty close to sure that these tubs contained either shielded single-20mms or shielded MGs - certainly nothing larger - which former, depicted in Mike's tiny scale of 1/600, would be about identical, anyway. Meantime, as we discovered he can also depict the "tubs" credibly as canvass-clad (i.e., filled-PE) railings, if desired.
This is just one of many possible examples - for this subject, in these 3 reference pics, alone - in which digital image enhancement can really help nail down a detail question - or at lest get "close enough" to the answer that you need. And, particularly when it delivers a useful result so quickly, in Mikey's own words: ...the research is half the fun...".
Cheers,
-Matty
Responses