hi, i agree that the easier way would probably to extract the images when the thing lands, just afraid it might go off course..but how should i control the camera? i need to send commands to the arduino board. or do i use timers?
I'm not sure how they do it in Malaysia, but generally you can't just release a large weather balloon to fly off willy-nilly without notifying some agency or such (ie - here in the USA it would be the FAA - in Malaysia, probably this: http://www.dca.gov.my/
). Unless you want to take the chance of an aircraft hitting your balloon, possibly killing people...
The fact that you are asking such basic questions tells me you have a ton to learn about this subject; you're worried about positioning the camera, etc - but you haven't even thought about permits, sourcing of the weather balloon and helium (won't be cheap!), insulation, heating, and power, the need for various radio comms licenses (ie - amateur/HAM radio), etc.
You will have no control over the balloon - assume it will go "off course" - the best you can do is understand the weather and air currents at the altitudes you plan to fly at, and pick a day with a good forecast. You may or may not be able to control the camera (if you do, you will likely need to use some kind of amateur radio frequency or satellite comms for such control). You likely won't be able to get live video or such from the balloon (again - not without spending a boatload of money on the equipment, which you may or may not be able to recover).
In short - you need to do a lot more research on this project before you build anything - and also gain a handle on what kind of a budget you are likely to need to do this properly, as well as all the other logistics. This isn't something you can generally do as a single person; you will likely want a team of people to help plan, launch and (hopefully) recover. You will need vehicles and people to run them if you do plan on recovering anything. At best, expect to spend a minimum of $1000.00 USD (and that is probably very optimistic) and expect to lose it all on the first launch, with nothing to show for it. If you aren't willing to do this, this isn't the project for you.