1

We like Kinovea and use it already as a perfect viewer of our videos.
If I would be a programmer myself, I believe it would be easy to create program how we would need it.

A little background information.
There are many amateur greyhound clubs in Europe, I am personally part of the greyhound-racing club in Tirol Austria 
Most of this clubs using old analog video finishing solution, as professional solutions start from 10’000 Euro up, what is a lot of money for an amateur club.
I understand very well that professional system for video finishing works a little bit different and has the possibility to measure 1/1000 second or even more.
What we would need
- If the gate of the start boxes open, a timer should be started.
- We are recording a video on the finish line and add on runtime the timer information.

(At the moment we would already be happy with a webcam using 75 Fps (Sony Egle Eye), but better would be 100 or even more frames per second.)
- When all greyhounds have passed the finish line we analyze the video with Kinovea and read the time and put that to an existing Excel program.

More details see my short overview:
https://onedrive.live.com/redir?resid=4 … file%2cpdf

Maybe someone tells me, that is all not problem or have an idea how it could be implemented.

Kind regards
Manfred

2 (edited by Chas Tennis 2016-03-31 21:47:55)

Rolling Shutter and Jello Effect Distortion

Most video cameras now sold are CMOS based.  They use an electronic readout called a "rolling shutter' to read the light signals from the sensor array.  They read one sensor line of elemental detectors at a time.  This results in a time delay from the first sensor line read to the later lines read.  The resulting image distortion is called "Jello Effect', it occurs for video frames and stills.

This issue should be evaluated as dogs imaged in one part of the frame may be seen behind those recorded in another part of the frame. For example, if you look along the finish line the dogs at the top of the frame may be imaged at an earlier time than those at the bottom of the frame. 

CCD based cameras with Global Shutters do not usually have this delay.

I measured some of these Jello Effect distortions for three CMOS video cameras, a Casio FH100 (240 fps), an Aiptek Action HD camera, 60 fps and a Canon Powershot 110 HS (30 fps).  See descriptions on Vimeo.

Casio Ex FH100 (240 fps), Jello Effect.  White line is straight.
https://vimeo.com/25242212

Aiptek Action HD (60 fps), Jello Effect. The black line is straight and parallel to the sides of the cylinder.
https://vimeo.com/30281692

Canon Powershot 110 HS (30 fps), Jello Effect. Black line is straight.
https://vimeo.com/72531589

Still picture with Jello Effect distortion. All blades of this fan are straight but two appear bent. The top blade was read out first - at an earlier time. It and the bottom blade are falsely bent.  For the horizontal blades one is widened and one has been made narrower. The one widened is moving in the direction of the rolling shutter scan and the one made narrower is moving against the scan direction. 
http://g1.img-dpreview.com/306AD1587A954231A81D344CCDC06012.jpg

3

Thank you for information about  "rolling shutter' effect.
I'm aware of it and in my case I can reduce it, I rotate the camera 90 degrees, as the dogs go in one direction.

My main problem at the moment is to put my own time information into the video.
Most of the programs offers the possibility to add time information out of the box, but there is always a delay at the start for initialize the camera.
I would be happy to take this possibility, if the vendor would a least take the exact time from the start command.
(I don't need the fist 20 seconds of the video, but I should have the exact time since the box has been opened)

And keep in mind, we are all amateur clubs and looking for an affordable solution.