emrain:
Just as the topic implies, andela applicants might find this thread useful.

chigozie391:
am looking for a way to do ma IT at Andela self

emrain:

It's almost impossible, but just apply.

HottestFire:
@4kings got this article for you.


http://members.cbio.mines-paristech.fr/~nvaroquaux/formations/scipy-lecture-notes/advanced/image_processing/index.html

4kings:

Thanks. Can't believe I wasn't aware of scipy features for images.

But what I'm looking for is how to get the positions from the pixels detected as edges. The page just provides a function for normal edge detection, i can do this with other means.
And what about colour detection?

Can you also give me any hint on how you would go about it, what if you want the system to detect different edges and draw them individually with different colours.

M.Sc thesis, waiting for others.

adepeter2027:
This one pass my power. It's been long I left python for Java.

HottestFire:





M.Sc thesis, waiting for others.

4kings:

Lol... I have too many stuff on my plate, so can't focus that's why I posted it here.

---

Adepeter2027 do you work on image processing with Java?
Deicide why na?

4kings:
I have two algorithm requests.

1) Thresholding for multiple color detection.

2) Algorithm to extract location of edges after performing edge detection on images.

(Preferably without OpenCV just PIL and maybe numpy. But hey beggars are not choosers )

Cc: Emrain , UmaryUsuf , Seun Adepeter2027 , Deicide , sonofluc1fer , JonBellion osarenomaspecial , SoftEng , adejumoadeoluwa , melodyogonna classicdude Olyboy16 pcguru danielthegeek guru01 and others.

---

Non-pythonitas can also post their codes so long as it's explained.

from PIL import Image, ImageFilter
import numpy as np

im = Image.open('your-image.jpg')

# Edge detection: background should be black (i.e. pi    xel value = 0). only edges should be set to a non-zero pixel value
im = im.filter(ImageFilter.FIND_EDGES) 
im.save('output.jpg')

# Therefore to get the indexes of edges, retrieve the indexes of non-zero pixels.
np_im = np.array(im)
indexes = np.where(np_im != 0)  
print(indexes)