Historical Graham Cracker
If someone asked you what a histogram was, would you be able to give them a straight answer? Me either, my first thought was, “histogram? that’s like some old high school textbook smelling old people cereal that gets the brown flowing right?” Maybe, and if doesn’t exist yet I already call dibs on it. Histogram! Now with more fiber and less artificial preservatives, available at your nearest CVS pharmacy, must be over 18 to purchase. Anyways, a histogram in the digital photography world is a visual representation of the lights and darks in a photo. But what kind of visual? Remember when getting a cd player for your car was cool and you didn’t have to bother with re-winding cassette tapes anymore? But you know you were one bad SOB if your head unit had that equalizer bar graph thing that would dance around when your music was jamming out for the entire block to hear and old people would shake their fist in air but little did you know you thought they too liked your music and so you would turn it up louder but your cheap paper back car speakers would blow out then you’d have to resort to getting a boom box for your car’s new sound system. Well, a histogram looks like that.
Well what about the lights and darks of my photo? Besides I still have no clue what your babbling on about. Not to get super nerd on yall, because I cant anyways but, I will do my best. Some colors are brighter than others and the histogram will show the true values of the brights and darks in a photo via a fun EQ looking graph. The higher up the graph is on one end the more of that value the photo has. Take for example if there is a spike on the left but the graph is flat towards the right side, odds are your photo has tons of dark values but next to nothing in light values. The same thing could happen if your spike is on the right but none on the left, this would indicate a photo with a lot of light values but hardly any dark value. Also worth noting, the more bell curved the graph is (think of a gentle mountain), the greater range of value the photo has. More specifically in the mid-range but pretty much it will be balanced.
Lastly you could have the opposite effect, where they’re peaks at both ends but the center is flatlining and this would indicate your pic has an extreme amount of darks and extreme amounts of brights but, not much mid-tones.
Now that you know about as much as I, lets answer questions that nobody asked. How can you tell just by looking at a histogram if a photo is correctly exposed? The way you can tell if a photo is correctly exposed is that there isn’t/shouldn’t be any radical spikes or weird dips in the histogram, a smooth consistent curve is most ideal for a correctly exposed image. Also, the wider the curve then the more range of tones your photo will include.
If after all my crazy talk you haven't absorbed anything then, congratulations! You’ve took but a step in my world and you've earned the privilege to see my photos of over, under and correct exposure, aren't you special.
Without trying to single handedly take down the internet with numerous pictures, I composed 9 images of poor versus good exposure into one photograph. The top row are under exposed shots that I took in broad daylight, by reducing the exposure compensator setting on my camera and I believe tweaking shutter speed I was able to capture very dark valued photos'. The 3 photos in the center are of a correctly exposed photo, the exposure compensator was near the zero setting (give or take)a neutral position for my camera, and again I had to bump down the shutter speed because my camera is picky apparently. Then we get to the last row of 3, where the pics are bright, very bright, and that's because I had the exposure set to max which gave me a washed out image.
Well that's about it for exposure compensation I’m sure I've made a few heads out their ache and with that my job is done.
P.S below are links to where I gathered some research, before I forget.
https://www.masterclass.com/articles/how-to-use-bracketing-to-create-perfect-photos#quiz-0
