My fellow techies: Help!
This is a post I made in SA, but I'm also reposting here. The project is for the Emerson College Performing Arts audition/interview process (I didn't want to associate with Emerson with this post, just in case).
Objective:
Take photos using a web cam, save photo onto a web server and associate data. All of this occurring within a web browser.
Given Circumstances:
- Must work on Windows AND OS X operating systems.
- Stay away from proprietary camera systems (Camera will only work with their own software).
- Minimal or no additional software installs.
- Must give visual conformation of image before officially saving image.
- Streamlined save to web along with data associated with image (No manual copy-file functions!).
Details:
I am in planning stages on a system for an electronic registration and adjudication audition system. There are a lot of parts to this project, but this post is focusing on taking photos of the auditionees. It is desired to make a standard photo for all of applicants rather than ask for head shots (so someone who has the money for professional headshots isn't dwarfed by someone who can't afford it).
In order for this to work, I need to be able to take a digital photo of someone, have visual conformation of the image ("Oh! I blinked!"), then sent to a web server for storage and retrieval. Since everything else in the system is browser-based, I want this picture functionality to be browser-based as well. Bonus points if a solution will work on both Windows and Mac OS X with little or no additional software installation. We can standardize on browser type (Mozilla Firefox) and photo-takers will be our own trained personal.
Possible Solution #1:
Mozilla Firefox + Flash Player 8 + (GD or ImageMagick)
It's not used too often, but Flash 8 Player allows for microphone and webcam use within the player. So why not take a still? However, there are some issues to go around.
1) Initializing and capturing still:
As long as the camera itself operates and the sure approves the player to read the camera, we're all set.
2) Exporting to a storable file:
Through the method above, you get a raw string of hex color information for this image. I can't find anything client-side that will narrow down this information, so it has to be done server-side. I found these instructions that will take the raw image information, send it via POST to a PHP script and using GD generates an image. As that site states, this is a lot of information to process.
I like how easy the system would be for the operator and I can build in all other functionality along with it, but is this too much of a server hit for three camera stations running at once? I'd retool this to use Imagemagick for better processing and I'm sure I'd shave some time, but there's a larger issue. My employer is stuck using ColdFusion, so I have no idea how that would work out. Suggestions are appreciated.