ImageEn for Delphi and C++ Builder ImageEn for Delphi and C++ Builder

 

ImageEn Forum
Profile    Join    Active Topics    Forum FAQ    Search this forumSearch
 All Forums
 ImageEn Library for Delphi, C++ and .Net
 ImageEn and IEvolution Support Forum
 How to detect vertical lines drawn by hand ?

Note: You must be registered in order to post a reply.
To register, click here. Registration is FREE!

View 
UserName:
Password:
Format  Bold Italicized Underline  Align Left Centered Align Right  Horizontal Rule  Insert Hyperlink   Browse for an image to attach to your post Browse for a zip to attach to your post Insert Code  Insert Quote Insert List
   
Message 

 

Emoji
Smile [:)] Big Smile [:D] Cool [8D] Blush [:I]
Tongue [:P] Evil [):] Wink [;)] Black Eye [B)]
Frown [:(] Shocked [:0] Angry [:(!] Sleepy [|)]
Kisses [:X] Approve [^] Disapprove [V] Question [?]

 
Check here to subscribe to this topic.
   

T O P I C    R E V I E W
jccruz Posted - Nov 28 2018 : 19:33:50
Dear all!
This image is part of an psychology assessment that I need to automate the analysis.
Each vertical lines is called "palos".
I need to detect all palos and be able to count them.
After that, I need to know the max distance between these palos.
Which palo is the largest and which is the smallest.

How can ImageEn or IEVision than help me with these chalenge?

Thanks
Jose Carlos







JCC
7   L A T E S T    R E P L I E S    (Newest First)
yogiyang Posted - Nov 30 2018 : 06:07:22
Hello,

if all else fails then try out Pattern Matching demos of IEVision. Though I have not tried it out but I think that will give you more accurate results.

Regards,


Yogi Yang
wesleybobato Posted - Nov 29 2018 : 15:27:26
Hi Jose!
You can try too!

https://www.imageen.com/help/TImageEnProc.ConvertToGray.html
https://www.imageen.com/help/TImageEnProc.RemoveNoise.html

Contrast
https://www.imageen.com/help/TImageEnProc.Contrast.html
https://www.imageen.com/help/TImageEnProc.Contrast2.html
https://www.imageen.com/help/TImageEnProc.Contrast3.html

Reduce Color
https://www.imageen.com/help/TImageEnProc.ConvertTo.html

Kind regards.
xequte Posted - Nov 29 2018 : 14:59:26
Hi Jose

Also, ensure you are prepping the image by increasing contrast and reducing colors.



Nigel
Xequte Software
www.imageen.com
jccruz Posted - Nov 29 2018 : 10:34:15
Hello Wesley and Nigel!
Thanks for your prompty answer!
Yesterday I tried BlobDetector and SeparateObjects but it did not work.
I tried Line Detection too, but not sucess.

Yes Wesley, I need count the "palos" by period. 152 means that there are 152 palos (vertical lines) in the first period (of 1 minute).
But I can select before detection to define a region.

What I need is a way to automaticaly detect and draw a vector exact over all palos in the region I selected.
If I have the vectors I can make any calculation.

Thanks again
Jose Carlos

JCC
wesleybobato Posted - Nov 29 2018 : 05:29:37
Hi Nigel

Are there regions that need to be isolated as it can do this?

He should stop counting every time he finds a number.

Will he need the OCR to identify these correct numerical characters? because it needs to compare the amount of "PALOS" with the number written

xequte Posted - Nov 28 2018 : 21:47:49
Hi

I would also try:

ImageEnView1.Proc.SeparateObjects();

https://www.imageen.com/help/TImageEnProc.SeparateObjects.html

And TIEVisionBlobDetector

https://www.imageen.com/help/TIEVisionBlobDetector.html

There are compiled demos for both on the Demos page.

Nigel
Xequte Software
www.imageen.com
wesleybobato Posted - Nov 28 2018 : 20:37:28
Hi Jose

I know this psychological test, much used by clinics of psychology in DETRANs of Brazil.

Have you tried using OCR? this is quite complex, and you forgot to explain that the numbers 152 and 140 is a pause ie you need to isolate regions from counts of PALOS.

Now calculations of distances and sizes are there many techniques have you explored all examples of IEVision?

https://www.imageen.com/demos/index.html

Drop the examples and take a test and share your results, I'm sure this is a great start.

Kind regards.