2

I am trying to make some transformations on an image with OpenCV and Python. I started by reading the image with cvLoadImage function, and then I got the image data with imageData function.

img = highgui.cvLoadImage("x.png",1)
data = img.imageData

The problem is, the imageData function returns a string data and when I try to do some calculations on the image data, it gives me error because e.g. it is not allowed to do substraction on strings in Python.

I have a C code as an example, and the following calculation works completely well:

x= data[100] + 4*data[40] -data[20]

But in Python, as I said, I can't do this. Any clue about this? What is the difference about Python vs C about this statement and how can apply this kind of calculations in Python?

Improve this question

3 Answers 3

Reset to default
1

As you've said, the imageData property returns a binary string containing the "raw image data" (I don't recall what format, though). Instead, you should access the image data by indexing into the img object:

>>> img = cv.CreateImage((10, 10), 8, 1)
>>> img[0, 0]
0.0
>>> img[0, 3] = 1.3
>>>
Improve this answer
Sign up to request clarification or add additional context in comments.

5 Comments

Oh, dang, sorry — didn't notice that you're using version 1 of the Python bindings, so my answer may not apply (I only know version 2).
An example of my imageData => "\x1e\x1e)))***///111666@@@IIIWWW]]]<<<:::TTT^^^}}}jjjQQQDDD<<<SSSYYYQQQNNN;;;MMMqqq\x7f\x7f\x7f\x8d\x8d\x8dhhhWWWRRRQQQPPPUUUgggvvv\x8c\x8c\x8c\x84"
The problem is, I am implementing a function by looking at a C code, and as I said they were able to do the transformation by using the ImageData. But I don't understand how they can do it. Is there a way to convert binary string to a integer in Python? Maybe in C language the chars (i don't have experiencve in C) it automatically converted to integer..
To turn that "string" into an "integer" you'd use something like this: data = [ord(ch) for ch in imageData] — then you could do stuff like data[3] = data[6] + data[8]. However, the conversion would be (comparatively) slow, so if I were you I'd look into the methods available on img to see if there's something like getPixel. You can list all the properties on img (methods and fields) using dir(img).
Of course, if it's possible to update your OpenCV bindings and use version 2, then you can use the syntax I've shown in my answer (and a bunch of other happy things)… But I realize that may not be possible.
0

If you're sure that the data you're getting as a string is actually an integer, you can cast it to an int.

i.e.

data = int(img.imageData)

http://docs.python.org/library/functions.html#int

This probably isn't the right way to achieve your goals however. Have you looked at the built-in library function examples?

http://opencv.willowgarage.com/documentation/python/operations_on_arrays.html

Improve this answer

2 Comments

I doubt int(imageData) is what you're looking for, as imageData will be a binary string…
Good call on the array operations, though — they are very helpful.
0

Try this:

Capture Image as Array with Python OpenCV

I encourage you to take a look at Python OpenCV Cookbook.

Improve this answer

Comments

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.