Global Name "Polygon" about advanced-lane-detection HOT 7 CLOSED

Run the cell where the global variable polygon is defined and make sure IMAGE = True.

from advanced-lane-detection.

I just run the cell which contains function 'pipeline' it define polygons as global, but not initialize it.
What you mean by IMAGE= true? Is it means I must check whether the image successfully loaded?

from advanced-lane-detection.

Once the global is defined it will stay active in your python kernel. IMAGE = True is a flag for processing the images and not the video. Makes a difference as you track frames for a video.

from advanced-lane-detection.

I had the same problem as you heyleeyandi! I think Jupyter was not able to understand the list comprehension.

I replaced these lines in the function lane_curve:
y = [x * (30/720.) for x in left_line[1]]
Instead:
y = []
for item in left_line[1]:
y.append(item*(30/720.))

There were some other errors after I replaced those lines, but I managed to get it running. It's really nice!

from advanced-lane-detection.

@frankzl I am running into same issue, even after making the fix you suggested:

Traceback (most recent call last):
  File "kylesf_lane_detect.py", line 519, in <module>
    Image_1 = pipeline(image_1, mtx, dist, dash)
  File "kylesf_lane_detect.py", line 272, in pipeline
    polygon = polygon
NameError: global name 'polygon' is not defined

In function impose_Lane_Area(), you might have to update
_unwarped = cv2.warpPerspective(**overlay**, M, img_size)_
since cv2.fillPoly doesn't returns None

Also, After making these 2 fixes, I am running into some new issue:
OpenCV Error: Assertion failed (contour1.checkVector(2) >= 0 && contour2.checkVector(2) >= 0 && (contour1.depth() == CV_32F || contour1.depth() == CV_32S) && contour1.depth() == contour2.depth()) in matchShapes

Any suggestion?

from advanced-lane-detection.

@kanishk-nd
It's been a while since I last looked at the code..
I just forked the repository and commited my modifications to the ipynb file for you. I only changed very little in the code. I tried running it again and the algorithm works on given the pictures.
Try it out and see if it works on your machine :)

from advanced-lane-detection.

@frankzl Thanks !! ^_^

I dumped the code in to a .py script, and realised that issue is with the Exception block (main exception block in pipeline() )which doesn't receive the 'polygon' variable. So I tweaked the code to not hit the last exception block for now.
You fixes work, as they also don't hit this Exception handler

from advanced-lane-detection.