Opencv error -Unsupported depth of input image:

Question

I am trying to convert a normalised RGB image to HSV or LAB colour space. Here is the normalisation function:

here is the basic code

print ('original image shape: ', image.shape)
print ('normlaised image shape: ', needed_multi_channel_img.shape)
# Converting to LAB color space


lab_image = cv.cvtColor(needed_multi_channel_img, cv.COLOR_RGB2HSV)

cv.imshow('Lab_space.jpg',lab_image.astype('float32'))
cv.waitKey(0)
cv.destroyAllWindows()

Here is the output trace:

    /home/centura/gitlab/Acne_model/Acne Model/rosaceaexperiment1.py:82: RuntimeWarning: invalid value encountered in true_divide
  norm.append(image[i,j,a]/rooted_matrix[i,j])
/home/centura/gitlab/Acne_model/Acne Model/rosaceaexperiment1.py:82: RuntimeWarning: divide by zero encountered in true_divide
  norm.append(image[i,j,a]/rooted_matrix[i,j])
original image shape:  (375, 600, 3)
normlaised image shape:  (375, 600, 3)
Traceback (most recent call last):
  File "/home/centura/gitlab/Acne_model/Acne Model/rosaceaexperiment1.py", line 121, in <module>
    lab_image = cv.cvtColor(needed_multi_channel_img, cv.COLOR_RGB2HSV)
cv2.error: OpenCV(3.4.3) /io/opencv/modules/imgproc/src/color.hpp:257: error: (-2:Unspecified error) in function 'cv::CvtHelper<VScn, VDcn, VDepth, sizePolicy>::CvtHelper(cv::InputArray, cv::OutputArray, int) [with VScn = cv::Set<3, 4>; VDcn = cv::Set<3>; VDepth = cv::Set<0, 5>; cv::SizePolicy sizePolicy = (cv::SizePolicy)2u; cv::InputArray = const cv::_InputArray&; cv::OutputArray = const cv::_OutputArray&]'
> Unsupported depth of input image:
>     'VDepth::contains(depth)'
> where
>     'depth' is 6 (CV_64F)

For zero division error, I have replaced it with 0 and nan is also replaced with 0.

I also searched through StackOverflow but could not find any information to debug it. I do not understand the meaning of this error and how to rectify it.

your image type is CV_64F (float64) which is not supported by cvtColor function. Convert the image to an appropriate type, e.g. float32 — Miki
– Miki, Commented Mar 15, 2019 at 9:49
i tried it normal_image[:,:,0] = normal_image[:,:,0].astype(np.float32) normal_image[:,:,1] = normal_image[:,:,1].astype(np.float32) normal_image[:,:,2] = normal_image[:,:,2].astype(np.float32) still the error persists. — Danish Xavier
– Danish Xavier, Commented Mar 15, 2019 at 10:18

PJ127 · Accepted Answer · 2020-05-05 05:43:09Z

30

According to this answer https://stackoverflow.com/a/45956247/7683041, try:

img_float32 = np.float32(needed_multi_channel_img)
lab_image = cv.cvtColor(img_float32, cv.COLOR_RGB2HSV)

answered May 5, 2020 at 5:43

PJ127

1,33017 silver badges30 bronze badges

Sign up to request clarification or add additional context in comments.

Comments

ArieAI · Accepted Answer · 2022-05-19 17:17:42Z

1

I actually had a similar issue, the problem was that the input was in floats but it was expecting integers. when i converted to integers (.astype('int')) this error disappeared. maybe I'll help you too.

answered May 19, 2022 at 17:17

ArieAI

5124 silver badges23 bronze badges

1 Comment

Community Over a year ago

As it’s currently written, your answer is unclear. Please edit to add additional details that will help others understand how this addresses the question asked. You can find more information on how to write good answers in the help center.

Collectives™ on Stack Overflow

Opencv error -Unsupported depth of input image:

2 Answers 2

Comments

1 Comment

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

Comments

1 Comment

Your Answer

Sign up or log in

Post as a guest

Linked

Related