The key is that you have to get the values from the x and y "sub" elements of the posted form values. When you have an image input field (let's call it "MyMap"), what gets submitted is actually two form fields: MyMap.x and MyMap.y. To be safe in .NET, you should construct the field names using the ClientID property of the InputImage control. Take a look at this example:
http://www.geekdork.com/samples/ImageMap.aspx
Here's the important code:
strXField = String.Format("{0}.x", ImageButton1.ClientID)
strYField = String.Format("{0}.y", ImageButton1.ClientID)
strX = Request.Form.Item(strXField)
strY = Request.Form.Item(strYField)
This will retrieve the coordinates of the mouse on the image. If you have an image that is some kind of grid, then you can just do some math based on the image size and returned coords to determine the grid location.