Regarding the use of cameras and different kinds of lenses to use photos alone as a survey tool: that's even less reliable than using a tool like Spike. The Spike tool takes into account the combination of distance from the sign along with the kind of lens being used to take the photo. The main problem I have with Spike is it's only as good as the camera and screen being used. Spike on an iPhone is kind of lousy; I think it works better on an iPad. If you're just using a photo on its own with no numerical measurements of distance and perspective correction you're going to be spit-balling to some degree. Even using a high quality, perfectly distortion-free 50mm prime lens on a good DSLR won't provide blueprint accurate results. The effects of view point, perspective and fore-shortening all come into play to throw off the accuracy of measurements. When analyzing an image the farther you get away from the center of the image the more the any would-be measurements would be thrown off by the effects of perspective. We don't view the real world isometrically. There's tricks to shooting pictures of things like buildings in near isometric fashion, using tilt-shift lenses, but that actually distorts the image.
For some projects there's no substitute for measuring by hand. We're doing a lot of store front window wraps these days. For the imagery to appear seamless across a bunch of different window panes and for installation to go off without a hitch LOTS of field measurements are needed up front. We not only measure all the window panes, but measure all the divider bars and overall dimensions that way all of the window hardware can be accurately sized and coordinately positioned across the layout of the store front. When all of that is done correctly up front and laid out correctly in the computer the installation of the graphics will often be a breeze.