I'm using the Microsoft Office Document Imaging 11.0 Type Library and 
performing OCR on images.  What I'm doing is starting with a 1600x900 32-bit 
image then using the system.drawing.bitmap class to extract smaller images 
and then performing OCR on them.  When I select a 480x28 pixel image it works 
fine, but if I select a 400x24 pixel image I get the following message when I 
go to get the text from the layout object:

Error: -2147467261: Attempted to read or write protected memory. This is 
often a
n indication that other memory is corrupt.

Here's my code:
Dim tmpStr As String
Dim miDoc As MODI.Document
Dim miImage As MODI.Image
Dim miImageLayout As MODI.ILayout
miDoc = New MODI.Document
miDoc.Create(fileName)
miImage = miDoc.Images.Item(0)
miImage.OCR()
miImageLayout = miImage.Layout
tmpStr = miImageLayout.Text

Is there some restriction on the dimensions of the source image?