On the Berlin / IAAF site, I like to look at the photofinish shots that accompany the race results.
Why are they so freakishly distorted? One person's leg might look like a tree-trunk & the other leg like an oddly curved twig. Why don't these just look like a normal photo?
i might be wrong but i think it's a composite of a couple of different angles/cameras
Vernacularitis wrote:
i might be wrong but i think it's a composite of a couple of different angles/cameras
It's just one camera positioned dead on the line.
You on the right kind of path there, its a composite of a number of frames each of one pixel in width. Parts of the body that cross the line more slowly are captured in more frames, so appear elongated to a normal photo. In a normal photo your x and y axis measure distance. In a photo finish image, the x axis measures time.
The finish camera records each part of an individual's body as it crosses the finish line. If someone's foot happens to land directly on the finish line, there will be a "tree-trunk" effect because the leg was on the line for a significant length of time. OTOH if a bodypart (e.g. a freely swinging leg) moves very rapidly past the line, it will register as very thin/shortened.
It its a bunch of thin imagse stuck together over time (see the time scale at the bottom of the image) If something crosses slowly it will appear longer or streched and vice versa.
wisco123 wrote:
On the Berlin / IAAF site, I like to look at the photofinish shots that accompany the race results.
Why are they so freakishly distorted? One person's leg might look like a tree-trunk & the other leg like an oddly curved twig. Why don't these just look like a normal photo?
In the finish photo each part of the body is at the same place on the line. The different parts of the body cross the line at different speeds. The slower a part is the more "tree-trunk" like it will appear. The faster a part goes through the more narrow it will be. The appearance of curved limbs is also a factor of how that limb crosses the finish.
wisco123 wrote:
Why don't these just look like a normal photo?
Because they are not "normal" photos.
High speed camera takes a narrow window (about the width of the finish line) at about 10,000 frames/second. It basically is taking a slice of time at a specific spot. As people move through that narrow window, they get 'sliced' (and if they happen to plant their foot on the line such that it is stationary for multiple slices, it will get elongated in the time axis).
The horizontal axis is actually time, not distance as you are used to in a 'normal' photo.
Have a look at their simple animation:
http://finishlynx.com/products/finishlynx/overview/body.htmThese photos are essentially taken through a slit at the finish line recording whatever happened to pass through the plane of the finish as time advances. The x-axis is scaled such that something moving past the finish line plane at the average speed of the runner is undistorted. Portions of the runner's body that are moving faster (say an arm or leg swinging forward) are compressed and appear thinner. Portions of the body on the backswing spend more time in the finish line plane and appear thicker.
If you understand the calculus process "integration" then you understand the concept of modern photofinish. I would guess the Seiko has their camera's taking 800-1000 pictures per second, that is a normal speed for photofinish at the world class level. Then the overall picture is "built" by placing these pictures (each picture is only a "sliver" width) side by side. As a person crosses not all the body parts are moving at the same speed; fast moving parts appear thin (because it moves through the lens quickly, so fewer slivers) while slow moving parts appear fat (more slivers.)
Most track and field guys don't understand the concept of photofinish; I've had so many coaches argue with me that his guy actually won the race because I took the "still" photo prior to the actual finish .... I entertain myself by just laughing at them.
runamile wrote:
wisco123 wrote:Why don't these just look like a normal photo?
Because they are not "normal" photos.
High speed camera takes a narrow window (about the width of the finish line)
The "photo" is not even close to the width of the finish line ... that would defeat the purpose. Each "photo" might the width of a human hair; a well trained photo operator places the thin lens directly on the front edge of the finish line (this is not easy to learn, to do ... but is the MOST important job of the photosystem operator.) Realistically, the photo operator probably can get that lens on the front cm of the finish line, but it's still only going to take pictures of the slimmest width.
Thanks for all of the responses. I now understand what I'm looking at...