Value Error: not enough values to unpack

[maxhallhatesm3]

Python 3 / OSX

$ python3 unproject_text/unproject_text.py samples/hd/IMG_1234.jpg
Traceback (most recent call last):
File "unproject_text/unproject_text.py", line 503, in
main()
File "unproject_text/unproject_text.py", line 490, in main
contours, hierarchy = get_contours(img)
File "unproject_text/unproject_text.py", line 135, in get_contours
cv2.CHAIN_APPROX_NONE)
ValueError: not enough values to unpack (expected 3, got 2)

[reddytocode]

in the 135 line
change this
_, contours, hierarchy = cv2.findContours(work, cv2.RETR_CCOMP, cv2.CHAIN_APPROX_NONE)
to this

contours, hierarchy = cv2.findContours(work, cv2.RETR_CCOMP, cv2.CHAIN_APPROX_NONE)

it's because in a previous version of opencv that function used to return 3 values, now it returns just 2 values

[shenxiancf12]

in the 135 line change this _, contours, hierarchy = cv2.findContours(work, cv2.RETR_CCOMP, cv2.CHAIN_APPROX_NONE) to this

contours, hierarchy = cv2.findContours(work, cv2.RETR_CCOMP, cv2.CHAIN_APPROX_NONE)

it's because in a previous version of opencv that function used to return 3 values, now it returns just 2 values

Still error as:
Traceback (most recent call last):
File "unproject_text.py", line 503, in
main()
File "unproject_text.py", line 495, in main
SRH = skew_detect(img, contours, RH)
File "unproject_text.py", line 467, in skew_detect
res = scipy.optimize.minimize_scalar(f, (-2.0, 0.0, 2.0))
File "\lib\site-packages\scipy\optimize_minimize.py", line 908, in minimize_scalar
return _minimize_scalar_brent(fun, bracket, args, **options)
File "\lib\site-packages\scipy\optimize_optimize.py", line 2606, in _minimize_scalar_brent
brent.optimize()
File "\lib\site-packages\scipy\optimize_optimize.py", line 2377, in optimize
xa, xb, xc, fa, fb, fc, funcalls = self.get_bracket_info()
File "\lib\site-packages\scipy\optimize_optimize.py", line 2361, in get_bracket_info
raise ValueError(
ValueError: Bracketing values (xa, xb, xc) do not fulfill this requirement: (f(xb) < f(xa)) and (f(xb) < f(xc))

It seems a little old project. Does this problem issue from the newest version of requiements (cv2 numpy scipy matplotlib) ?

[shenxiancf12]

in the 135 line change this _, contours, hierarchy = cv2.findContours(work, cv2.RETR_CCOMP, cv2.CHAIN_APPROX_NONE) to this
contours, hierarchy = cv2.findContours(work, cv2.RETR_CCOMP, cv2.CHAIN_APPROX_NONE)
it's because in a previous version of opencv that function used to return 3 values, now it returns just 2 values

Still error as: Traceback (most recent call last): File "unproject_text.py", line 503, in main() File "unproject_text.py", line 495, in main SRH = skew_detect(img, contours, RH) File "unproject_text.py", line 467, in skew_detect res = scipy.optimize.minimize_scalar(f, (-2.0, 0.0, 2.0)) File "\lib\site-packages\scipy\optimize_minimize.py", line 908, in minimize_scalar return _minimize_scalar_brent(fun, bracket, args, **options) File "\lib\site-packages\scipy\optimize_optimize.py", line 2606, in _minimize_scalar_brent brent.optimize() File "\lib\site-packages\scipy\optimize_optimize.py", line 2377, in optimize xa, xb, xc, fa, fb, fc, funcalls = self.get_bracket_info() File "\lib\site-packages\scipy\optimize_optimize.py", line 2361, in get_bracket_info raise ValueError( ValueError: Bracketing values (xa, xb, xc) do not fulfill this requirement: (f(xb) < f(xa)) and (f(xb) < f(xc))

It seems a little old project. Does this problem issue from the newest version of requiements (cv2 numpy scipy matplotlib) ?

Ok, I find the original code only works on deskew0.jpg. What should I need to do for apply this project on my own picture.