Advanced Despeckle image cleanup

The Advanced Despeckle image cleanup method removes unwanted blobs from an image. Within the Project Builder application image cleanup methods can be used only for the Advanced Zone Locator.

However, using script the Advanced Despeckle image cleanup method can also be applied elsewhere. To access the methods for the CscBlobRemoval object you have to add a reference to the CscForms library in the script, which provides the following methods:

AddBlobRule - use this method to define the constraints for a blob to be detected.
AddBlobRule2 - use this method instead of AddBlobRule if you want to use additional constraints, minimum mass and minimum proportion, for a blob to be detected.
DetectBlobs - use this method to detect all blobs that apply to the prior added blob rules on the given image without removing them from the image. The information about the detected blobs is stored.
RemoveBlobs - use this function to detect and remove all blobs that apply to the prior added blob rules from the given image and return the modified image without saving any information about the blobs.
DetectAndRemoveBlobs - use this function to detect and remove all blobs that apply to the prior added blob rules from the given image and return the modified image. The information about the detected blobs is stored.
BlobCount - use this function to return the number of detected blobs for DetectBlobs and DetectAndRemoveBlobs. The function returns zero if it is called after you have executed RemoveBlobs.
GetBlob - use this function to return a detected blob by its index. The CscBlobInfo object contains the information about the detected blob.

Detecting hand written signatures

You can use the Advanced Despeckle image cleanup to remove unwanted blobs from an image or to detect certain kinds of blobs, such as hand written signatures. Hand written signatures are usually a set of bound characters having a small blob proportion of 5-30%, in comparison to handprinted text (20-50%) or machine printed text (30-80%). In combination with other restrictions (for example on a 200 DPI image: minimum mass 250 pixels, width 40-500 pixels and height 10-300 pixels) signatures can easily be detected. The following script example displays the first five found blobs using message boxes.

' Add a reference to "Kofax Cascade Forms Processing" library to your script!

Private Sub Document_BeforeExtract(ByVal pXDoc As CASCADELib.CscXDocument)
    Dim br As New CSCFORMSLib.CscBlobRemoval
    Dim bi As CSCFORMSLib.CscBlobInfo
    Dim i As Integer

    br.AddBlobRule2(250, 0, 5, 30, 40, 500, 10, 300, 0, True) ' on a 200DPI image
    br.DetectBlobs(pXDoc.CDoc.Pages(0).GetBitonalImage(Project.ColorConversion))
    MsgBox("Found " + CStr(br.BlobCount) + " blobs.")
    For i = 0 To br.BlobCount-1
        If (i < 5) Then
            Set bi = br.GetBlob(i)
            MsgBox("Blob #" + CStr(i) + ": (" + CStr(bi.x) + "," + CStr(bi.y) + ") [" + _
                CStr(bi.Width) + "," + CStr(bi.Height) + "] Mass=" + CStr(bi.Mass) + " Prop=" + CStr(bi.Prop))
        End If
    Next
End Sub

Your search for returned results.

Search tips

Advanced Despeckle image cleanup

Detecting hand written signatures