ahmetozlu / signature_extractor Goto Github PK

hey there! amazing code.
I have a question, where do you get those constant_parameter_1, 2, 3 and 4 values? ? I'm still trying to understand where those values came from.

Thanks in advance!

Please, is difficult to develope a function that return true if a signature is present and false if not?

I need to understand if a document has been signed.

Thanks a lot

Error: 'image' and 'label' must be the same shape
While using the label2rgb function. The labels have (x,y,3) dimensions. How is that possible?

Hi.

Is it possible to retrieve the coordinates of the form (x,y,w,h)?? What??

Code identify table header and other shaped components. how to avoid those components?

I'm thinking to skip saving pre_version image and pass to the next step
So, I'm thinking to get a function to do this but not sure how to do this.
I tried to convert it as unit8 type but the colour is not correct.
I think it's realted to the types and normalization issues.
Could you advise how to get the function to do this? Thanks

img=function(pre_version) img = cv2.threshold(img, 0, 255, cv2.THRESH_BINARY_INV | cv2.THRESH_OTSU)[1] cv2.imwrite("output.png", img)

Excuse me,
After testing, I found that it didn't work well in many cases. Is there a better signature detection model?

In StrokeFilter.py file function def strokeFilter(im, orient, freq, kx, ky): is giving error on following line

sigmax = 1 / unfreq[0] * kx sigmay = 1 / unfreq[0] * ky

Error Stack -

File "D:\OCR\signature_extractor-master\signature_verifier\verifier\__init__.py", line 12, in verify
    kp1, des1 = extractKeypoints(img1)
  File "D:\OCR\signature_extractor-master\signature_verifier\verifier\ExtractKeypoints.py", line 42, in extractKeypoints
    img = preprocessing(img)
  File "D:\OCR\signature_extractor-master\signature_verifier\verifier\Preprocessing.py", line 26, in preprocessing
    newim = strokeFilter(normim, orientim, freq, kx, ky)
  File "D:\OCR\signature_extractor-master\signature_verifier\verifier\StrokeFilter.py", line 21, in strokeFilter
    sigmax = 1 / unfreq[0] * kx
IndexError: index 0 is out of bounds for axis 0 with size 0

Hi,
I don't understand why there are two constants when you do :
a4_small_size_outliar_constant = ((average / c1) * c2) + c3, why not a4_small_size_outliar_constant = ((average c) + c3 where c=c2/c1 ?

Thx