Solution for Multiple Object Detection and Tracking

edit

Try to be more specific in your description of the problem. Is the tray clearly different from the objects? Is it smooth or textured? Is it moving? What about the lighting conditions? Are they constant or can very? And objects? Always the same or can differ? Post an example image.

@Witek , Tray is nothing but its just a plain moving sheet , it is black in color no texture(i missed it in attachment), yes it is moving continuously in one direction either back or forth, lightning condition will be normal or we could just make adjustments on top of tray to avoid shadows beside object, yes they are constant in unique place but orientation may change, objects will be same they won't differ, for example i got some parts from my garage ,can refer image attached.

Are these the objects you want to detect?? It is important to know exactly what objects you are dealing with before deciding for a particular solution. Unless you want to detect anything and you don't know what it is going to be. And what should be the output? Just some text each time an object is detected? What if we see more than one object? Do you need to count or recognize the objects or just detect that something is there?

@Witek, okay 'll answer you in detail, Lets say if i keep parts of my computer in front of camera then it should recognize(Not simply detect) what is what and these parts will be same they won't change, only their orientation and their place on tray may change. i want to recognize above objects for what i have now, of course there are 6 objects in above image and i just want to recognize all of 'em at the same time or i may place only 4 and show its name when it is recognized(in future i may want to display description), But let me recognize them first.

Ok, so now the situation is slightly different. You want to RECOGNIZE the objects and not only DETECT them. That is much more difficult. The key is to find some features that are easy to calculate and enable distinction between the objects at the same time. In your particular case, since you want to recognize these 6 objects that will just be oriented differently, it is quite simple.

1. Subtract the empty background from the image with objects.
2. Threshold the result
3. Find contours.
4. Calculate the area of the contours. You can recognize the four smallest ones just by their area.
5. As for the two biggest ones, find average color of a small patch of pixels in the center of each of them. You will find your red and blue objects.

@Witek, Thanks for your valuable suggestion, 'll try this if i am facing any prob. 'll post it.