img = Image('logo')
l = img.findLines()
c = l.crop() #returns an np.array type of None

The desired functionality is to crop to the line feature. Also returning np.array of images for crop seems wrong. Seems like a list / image set would be the better approach.

http://developers.face.com/download/

The following code produces an error when finding a barcode. The barcode.png is from the sampleimages directory.

from SimpleCV import *
from zxing import *
i = Image("./barcode.png")
b = i.findBarcode("./zxing")
print b.data

The error produced is:

from SimpleCV import *
from zxing import *
i = Image("./barcode.png")
b = i.findBarcode("./zxing")
Exception in thread "main" java.lang.NoClassDefFoundError: com/google/zxing/clie
nt/j2se/CommandLineRunner
Caused by: java.lang.ClassNotFoundException: com.google.zxing.client.j2se.Comman
dLineRunner
at java.net.URLClassLoader$1.run(Unknown Source)
at java.security.AccessController.doPrivileged(Native Method)
at java.net.URLClassLoader.findClass(Unknown Source)
at java.lang.ClassLoader.loadClass(Unknown Source)
at sun.misc.Launcher$AppClassLoader.loadClass(Unknown Source)
at java.lang.ClassLoader.loadClass(Unknown Source)
Could not find the main class: com.google.zxing.client.j2se.CommandLineRunner.
Program will exit.
print b.data

Would really like readthedocs for autobuilding our stuff automatically.

The haarfile is being loaded each time in the loop when doing detection, we need a way to cache it after it has been loaded once.

This function works although the output doesn't appear to be what would be expected. Where as you would expect it to function similar to an alpha mask in photo shop where (black or white) get's passed through.

Hi,
being on OSX is just clicked on the download link.
Afterwards, I noticed:

1. it gets SimpleCV 1.1 even though 1.2. is out (but no superpack for that)
2. It wants to run a install script that puts all kinds of files all over the system .....
   (using sudo, making symlinks and stuff)
3. did not work with my distutils config file.

Why don't you give people a choice - inserting one more page after clicking on download,
explaining the options .....

• Sebastian Haase

I installed SimpleCV using the .PKG superpack for Mac, but Python can't find the SimpleCV module.

$ python
Python 2.7.1 (r271:86882M, Nov 30 2010, 10:35:34) 
[GCC 4.2.1 (Apple Inc. build 5664)] on darwin
Type "help", "copyright", "credits" or "license" for more information.
>>> from SimpleCV import Camera
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
ImportError: No module named SimpleCV

Specs

$ system_profiler SPSoftwareDataType | grep Mac
      System Version: Mac OS X 10.7.3 (11D50b)
$ python --version
Python 2.7.1

Current solution involves easy_install SimpleCV

The example at:

http://simplecv-examples.readthedocs.org/en/latest/index.html#features
blobs.draw() leads to an error

"AttributeError Traceback (most recent call last)
/usr/local/lib/python2.7/dist-packages/SimpleCV-1.2-py2.7.egg/SimpleCV/Shell/Shell.pyc in ()"
AttributeError: 'NoneType' object has no attribute 'draw'

I have installed SimpleCV-1.2 on Ubuntu 11.10.

Thank you for the great work, making live easier.

I'm following this example:

http://simplecv.org/doc/cookbook.html#using-a-camera-kinect-or-virtualcamera
Where is reads:
(red, green, blue) = Camera().getImage().channels()
I do it like this, as I already defined cam = Camera()
cam.getImage().channels()

but I get this error:

SimpleCV:21>(red, green, blue) = cam.getImage().channels()

AttributeError Traceback (most recent call last)

/usr/local/lib/python2.7/dist-packages/SimpleCV/Shell/Shell.pyc in ()
----> 1
2
3
4
5

AttributeError: Image instance has no attribute 'channels'

I'm trying SimpleCV 1.2 on Ubuntu 11.04
we know that we can load an image to SImpleCV by using the function Image(path)
eg: myImage= Image("/home/ubuntu/pics/face.jpeg")

then the image will appear in the python(v2.7) window if the image exists after using the function myImage.show()

In the picture window, the exit button is not working. I couldn't find any other method to close this window.

Hey,

   I am getting error after i enter following code.

'''
SimpleCV:1> im = Image('/home/vijay/Pictures/Screenshot.png')
SimpleCV:2> im.show()
'''
referring to jay's recent post, link: https://github.com/ingenuitas/SimpleCV/issues/88.
I am getting similar error in this case also.

Error log :
http://pastebin.com/hUwrahaz

For both the Microsoft Kinect, and other range or time-of-flight cameras, depth data comes in as 16-bit signed integers, standing in as fixed-point values for distances away from the device. The dynamic range and z-resolution is important to maintain, and crushing to 8-bit makes the data essentially unusable. Having 16-bit images supported transparently in SimpleCV would go a long, long way in this regard.

Would like a quick reference page like processing:

http://processing.org/reference/

http://simplecv.org/demos.html - videos are removed from YouTube and won't play.

I've tried to write a quick face detection script using SimpleCV. Although I love how simple it is, it seems to be really slow compared to the OpenCV Python version. Is there some API I'm using in the wrong way?

import sys
from SimpleCV import Camera, Color
from SimpleCV.Display import Display

def main():
    # Try to connect to the camera.
    cam  = Camera()

    if not cam:
        print 'Error opening camera. Exiting...'
        sys.exit(1)

    # Create PyGame display.
    disp = Display()

    # Main processing loop.
    while not disp.isDone():
        # Get an image from the camera.
        img = cam.getImage()

        faces = img.findHaarFeatures("haarcascade_frontalface_alt.xml")
        i = 0
        if not faces is None:
            for f in faces:
                f.draw(Color.GREEN)

        # Show the camera image.
        img.save(disp)

if __name__ == "__main__":
    try:
        main()
    except KeyboardInterrupt:
        print "\nStopping..."
        sys.exit(1)

After re-writing the test suite I noticed that a lot of the color conversions aren't handled correctly in the display and save methods. I think save/display should automagically do the conversion to RGB prior to saving or display the images.

I'm using Ubuntu 11.10

SimpleCV:1>cam = Camera()
mmap: Invalid argument
munmap: Invalid argument
munmap: Invalid argument
munmap: Invalid argument
munmap: Invalid argument
Unable to stop the stream.: Bad file descriptor
munmap: Invalid argument
munmap: Invalid argument
munmap: Invalid argument
munmap: Invalid argument
SimpleCV:2>cam = Camera()
libv4l2: error setting pixformat: Device or resource busy
HIGHGUI ERROR: libv4l unable to ioctl S_FMT

libv4l2: error setting pixformat: Device or resource busy
libv4l1: error setting pixformat: Device or resource busy
HIGHGUI ERROR: libv4l unable to ioctl VIDIOCSPICT

SystemError Traceback (most recent call last)

/usr/local/lib/python2.7/dist-packages/SimpleCV/Shell/Shell.pyc in ()
----> 1
2
3
4
5

/usr/local/lib/python2.7/dist-packages/SimpleCV/Camera.pyc in init(self, camera_index, prop_set, threaded, calibrationfile)
292 self.capture = pygame.camera.Camera("/dev/video" + str(camera_index))
293
--> 294 self.capture.start()
295 time.sleep(0)
296 self.pygame_buffer = self.capture.get_image()

SystemError: ioctl(VIDIOC_S_FMT) failure: no supported formats
SimpleCV:3>

I cannot get function sprite() to work.

from SimpleCV import *

cam = Camera()
j=0
fillimg = Image("img.jpg")
while (j<100):
i = cam.getImage()
i.dl().sprite(fillimg, pos=(0,0), scale=0.1)
j=j+1
i.show()

Traceback (most recent call last):
File "asdf.py", line 8, in
i.dl().sprite(fillimg, pos=(0,0), scale=0.1)
File "/usr/lib/pymodules/python2.7/SimpleCV/DrawingLayer.py", line 516, in sprite
image = image.convert(self._mSurface)
AttributeError: Image instance has no attribute 'convert'

Need to just setup a script so simplecv can update itself without having to use git (or just use git via python).

I would like something basically you fire up the simplecv shell and type:
update()

Let me start by saying Im not just new to programming, but also new to the mac OS. I watched the video you made for installing SimpleCV. I installed to all the default directories.
When I tried the camera command, or any I get device not defined. So it says camera not defined for the camera command.
What am i missing?
Jason

I'm using SimpleCV 1.2 with Ubuntu 12.04. Creating a camera without an argument results in a number of warnings:

>>> from SimpleCV import *
>>> cam = Camera()
VIDIOC_QUERYMENU: Invalid argument
VIDIOC_QUERYMENU: Invalid argument
VIDIOC_QUERYMENU: Invalid argument
>>> 

I would have expected this to be handled transparently (just give me the first camera you can find), unless I specify an argument (e.g. 0 or 1 for first or second camera). The same warnings show when I pass 0 as an argument, however.

On a side note, if you pass OpenCV's cv.CaptureFromCAM the argument -1, it should grab whatever camera it can get. I think SimpleCV instead passes OpenCV index 0 by default.

DICOM is a medical imaging format:
http://en.wikipedia.org/wiki/DICOM

There currently exist a library for it.
http://code.google.com/p/pydicom/

Would be nice to add as optional library with native support in SimpleCV.

Came across this while writing a test case. When you take a sublist of a featureset using index notation it returns a list of features when it should return a featureset. This makes it impossible to use featureset member functions on sub-lists.

Example:

SimpleCV:18> img = Image("lenna")
SimpleCV:19> l = img.findLines()
SimpleCV:20> l2 = l[0:10]
SimpleCV:21> type(l2)
SimpleCV:21: list
SimpleCV:22> l2.above((img.width/2,img.height/2)
....: )
ERROR

I'd like to blit Japanese text onto an image, but it shows up as a bunch of rectangles.

I try setting the font to Arial, a unicode font, but that results in an IOError.

Code:

SimpleCV.Font().setFont("/Users/andrew/Desktop/src/sensei/arial.ttf")

Trace:

<type 'int'>
Traceback (most recent call last):
  File "./sensei.py", line 51, in <module>
    main()
  File "./sensei.py", line 19, in main
    SimpleCV.Font().setFont("/Users/andrew/Desktop/src/sensei/arial.ttf")
  File "build/bdist.macosx-10.7-intel/egg/SimpleCV/Font.py", line 49, in __init__
  File "build/bdist.macosx-10.7-intel/egg/SimpleCV/Font.py", line 78, in setFont
  File "/usr/local/lib/python2.7/site-packages/PIL/ImageFont.py", line 218, in truetype
    return FreeTypeFont(filename, size, index, encoding)
  File "/usr/local/lib/python2.7/site-packages/PIL/ImageFont.py", line 134, in __init__
    self.font = core.getfont(file, size, index, encoding)
IOError: cannot open resource

I've installed "simplecv-setup-1.2.exe" on Windows 7 Professional SP1 and I'm getting a few errors.
First, iPython fails with the error *** run_installscript: internal error 0xFFFFFFFF ***

then Cython fails with: "Unable to find vcvarsall.bat"
Here's the full easy_install process:

Running Cython-0.16\setup.py -q bdist_egg --dist-dir c:\temp\easy_install-kmaamv
\Cython-0.16\egg-dist-tmp-siyvpj
Compiling module Cython.Plex.Scanners ...
Compiling module Cython.Plex.Actions ...
Compiling module Cython.Compiler.Lexicon ...
Compiling module Cython.Compiler.Scanning ...
Compiling module Cython.Compiler.Parsing ...
Compiling module Cython.Compiler.Visitor ...
Compiling module Cython.Compiler.FlowControl ...
Compiling module Cython.Compiler.Code ...
Compiling module Cython.Runtime.refnanny ...
warning: no files found matching '*.pyx' under directory 'Cython\Debugger\Tests'

warning: no files found matching '*.pxd' under directory 'Cython\Debugger\Tests'

warning: no files found matching '.h' under directory 'Cython\Debugger\Tests'
warning: no files found matching '.pxd' under directory 'Cython\Utility'
warning: no files found matching '*.h' under directory 'Cython\Utility'
warning: no files found matching '.cpp' under directory 'Cython\Utility'
warning: no files found matching 'bugs.txt' under directory 'tests'
error: Setup script exited with error: Unable to find vcvarsall.bat

Looks like Cython requires " Microsoft’s Visual C" to be installed or the open source MinGW (a Windows distribution of gcc).

Any suggestions as how to proceed if I don't have either?

Scikits-Learn has a bunch of awesome machine learning tools, it's open source, and cross platform. There are many examples they have like face recognition, etc that can be found here:
http://scikit-learn.sourceforge.net/dev/auto_examples/index.html

Instead of shouting in the docs that users must MUST MUST install SciPy/NumPy, consider making SciPy/NumPy a dependency in the egg. The whole point of SimpleCV is making opencv easy; this sort of thing should be automated.

The docs seem to indicate that either SciPy or NumPy is sufficient. Whichever is better suited for use with SimpleCV should have a dependency entry in the egg.

Ditto for PyGame.

I'm running Ubuntu 12.04, with SimpleCV version 1.2.

I can't seem to import the shell. I can show the simplecv logo from a normal Python shell, but this is quite annoying as it's impossible to close windows in this version.

>>> from SimpleCV import Shell
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/usr/lib/pymodules/python2.7/SimpleCV/Shell/__init__.py", line 1, in     <module>
    from Shell import *
  File "/usr/lib/pymodules/python2.7/SimpleCV/Shell/Shell.py", line 35, in <module>
    from IPython.Shell import IPShellEmbed
ImportError: No module named Shell

This is an image processing javascript library. It may go well with web based frontend.
https://github.com/jseidelin/pixastic

I have tried, but can't seem to get travis-cI (http://travis-ci.org/) working. This is due to the fact I can't seem to get it to install opencv on their servers even using apt-get and trying with opencv 2.1. The travis.yml file is already in the simplecv repo. I believe they have all the python required libs for simplecv to run, but feel free to add a pip requriements file if needed.

We currently need to load Haar Cascade files like:

face = cv.Load("face.xml")

It would probably be better not to make users work with the underlying opencv objects. Something like:

face = HaarCascade("face.xml")

would probably be more consistent with the way the rest of SimpleCV works.

http://sseemayer.github.com/Py2D/

This seems to be a good translation type of layer for handing data back to be rendered in a web interface in something like processing.js

I downloaded SimpleCV from GitHub just now. I installed it

sudo python setup.py install

I tried

i = Image("simplecv")

I am getting error

IOError: [Errno 2] No such file or directory: '/usr/local/lib/python2.7/dist-packages/SimpleCV-1.2-py2.7.egg/SimpleCV/sampleimages/simplecv.png'

sampleimages folder is not there in /usr/local/lib/python2.7/dist-packages/SimpleCV-1.2-py2.7.egg/SimpleCV
So, I looked at setup.py file, I couldn't find any problems.

package_data  = { #DO NOT REMOVE, NEEDED TO LOAD INLINE FILES i = Image('simplecv')
        'SimpleCV': ['sampleimages/*'],
        'SimpleCV':['Features/HaarCascades/*'],

HaarCascades are there but not sampleimages.

I tried re-installing SimpleCV from source that I had downloaded couple of weeks back. It's setup file includes only sampleimages.

package_data  = { #DO NOT REMOVE, NEEDED TO LOAD INLINE IMAGES
        'SimpleCV': ['sampleimages/*']

After installing from old version, I can access sampleimages

i = Image("simplecv")

Installed SimpleCV-e7665f5 > http://bpaste.net/show/28462/
Installed SimpleCV-3fea5c1 > http://bpaste.net/show/mcMUp9gjCWoWTvB5EBzX/

What does the below error message means?

And how can i solve that? I am running example MotionTracker.py

UserWarning: SimpleCV Find Blobs Failed - This could be an OpenCV python binding issue warnings.warn("SimpleCV Find Blobs Failed - This could be an OpenCV python binding issue")

The link from this page http://examples.simplecv.org/en/latest/index.html titles Tutorials leads to an inaccessible page.

Mahotas is another imaging library that is written in pure python. It would be nice to have support built into our Image Class so you have the option to use mahotas if say openCV isn't installed and falls back gracefully.

See:
https://github.com/luispedro/mahotas

ubuntu 10.10 fresh clean install / opencv 2.1 (python-opencv) / python-scipy / python-imaging / python-numpy /python-pygame / 1-click install of SimpleCV all installed.

trying to run:
from SimpleCV import *
mycam = Camera()
img = mycam.getImage()

getting:
Exception in thread Thread-1 (most likely raised during interpreter shutdown):
Traceback (most recent call last):
File "/usr/lib/python2.6/threading.py", line 532, in __bootstrap_inner
File "/usr/local/lib/python2.6/dist-packages/SimpleCV-1.2-py2.6.egg/SimpleCV/Camera.py", line 23, in run
<type 'exceptions.TypeError'>: 'NoneType' object is not iterable

also tried:
sudo easy_install simplecv
install_dir /usr/local/lib/python2.6/dist-packages/
Searching for simplecv
Best match: SimpleCV 1.2
Processing SimpleCV-1.2-py2.6.egg
SimpleCV 1.2 is already the active version in easy-install.pth

Using /usr/local/lib/python2.6/dist-packages/SimpleCV-1.2-py2.6.egg
Processing dependencies for simplecv
Finished processing dependencies for simplecv

help. thanks

Another finding from the improved unit tests -- color model doesn't like to take in tuples. Seems to work well with images. The following will return an all white image. The correct behavior is to color the red, green, and blue, sections black and have everything else be white.

SimpleCV:50> img = Image("logo")

SimpleCV:51> cm = ColorModel()

SimpleCV:52> cm.add((255,0,0))

SimpleCV:53> cm.add((0,255,0))

SimpleCV:54> cm.add((0,0,255))

SimpleCV:55> cm.theshold(img)

SimpleCV:56> test = cm.threshold(img)

SimpleCV:57> test.show()
SimpleCV:57: <SimpleCV.Display Object resolution:((250, 250)), Image Resolution: (250, 250) at memory location: (0x118a97128)>

SimpleCV:58> cm.add((0,0,0))

SimpleCV:59> test = cm.threshold(img)

SimpleCV:60> test = cm.threshold(img)

SimpleCV:61> cm.add((255,255,255))

SimpleCV:62> cm
SimpleCV:62: <SimpleCV.ColorModel.ColorModel instance at 0x118a92878>

SimpleCV:63> cm.mData
SimpleCV:63:
{'\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00': 1,
'\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x7f\x00\x00\x00\x00\x00\x00\x00': 1,
'\x00\x00\x00\x00\x00\x00\x00\x00\x7f\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00': 1,
'\x7f\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00': 1,
'\x7f\x00\x00\x00\x00\x00\x00\x00\x7f\x00\x00\x00\x00\x00\x00\x00\x7f\x00\x00\x00\x00\x00\x00\x00': 1}

This just returns false and needs to be more descriptive.

when building tests, i use featureset.show() a lot, and it is annoying when tuning parameters to have to run img.clearLayers() every time I want to change a number. featureset.show() should initialize a drawing layer, apply to image, and then discard it.

SimpleCV.npArray2cvMat apparently silently assumes that it get single precision float (np.float32)

In other case strange errors occur:

U.arrD((1,0,0),(0,1,0.))
array([[ 1., 0., 0.],
[ 0., 1., 0.]])
SimpleCV.npArray2cvMat( U.arrD((1,0,0),(0,1,0.)) )
<cvmat(type=42424005 32FC1 rows=2 cols=3 step=24 )>
N.asarray(_)
array([[ 0. , 1.875, 0. ],
[ 0. , 0. , 0. ]], dtype=float32)

When running:
nosetest SimpleCV/test/tests.py

Get a few errors on the blobs, and a couple other test. It appears to be with the OpenCV constants.
Currently on my Ubuntu 10.10 machine which everything is running fine, going to dump unit test output when I return home.

This function seems to die very often by hitting the max recursion depth (even on simple test cases).

Is there a better set of default arguments to prevent hitting this failure?

We need a way to do something like:
import SimpleCV
SimpleCV.system()

And have it output all the various system and library information. So the person can copy and paste when they submit bug reports so we don't have to elicit that information everytime someone submits a bug. (i.e. Did you install on windows? What version? What version of SimpleCV are you running? What pygame version? etc).

A friend wants to use SimpleCV to perform some tasks common performed by image magick. He uses image magick to go through a pile of images, check the jpeg quality setting, and if it is too high re-save the image at a lower quality setting. If the jpeg is already at a lower quality then nothing needs to be done. I noticed that this is possible via PIL but not exposed cleanly through the SimpleCV interface. You can see where this saving would happen here:

https://github.com/ingenuitas/SimpleCV/blob/master/SimpleCV/ImageClass.py#L1697

Here is the relevant PIL documenation:

http://www.pythonware.com/library/pil/handbook/image.htm

I think the code should look something like this:

fname = 'foo.jpg'
img = Image(fname)
if( img.quality > 80 ):
   img.save(fname,quality=80)

The default value for quality should be None, which defaults to the images quality at load time.

Would it make sense to have the Image class raise an IOError when the source is of type str() and empty '' ? This would simplify our (@alokshankar and @Soli2ude) code wrapping SimpleCV, although checking for empty strings is easy.

What does having the '' source do for things (use case)?

Things that get past IOError lead to us hitting this (if we don't trap before instantiating an Image):

Traceback (most recent call last):
  File "face_detector.py", line 36, in <module>
    for result in executor.map(find_faces, lines):
  File "/usr/local/lib/python2.7/dist-packages/futures-2.1.2-py2.7.egg/concurrent/futures/_base.py", line 550, in map
    yield future.result()
  File "/usr/local/lib/python2.7/dist-packages/futures-2.1.2-py2.7.egg/concurrent/futures/_base.py", line 398, in result
    return self.__get_result()
  File "/usr/local/lib/python2.7/dist-packages/futures-2.1.2-py2.7.egg/concurrent/futures/_base.py", line 357, in __get_result
    raise self._exception
TypeError: CvArr argument 'arr' must be IplImage, CvMat or CvMatND. Use fromarray() to convert numpy arrays to CvMat or cvMatND

_getRawKeypoints() function returns coordianates of detected keypoints in reversed order eg. (y,x) instead of (x,y).
If you use keypoints.pt the returned tuple needs to be reversed.

Need readme cleaned up, get rid of cookbook, move to examples.simplecv.org, etc.

https://github.com/ingenuitas/SimpleCV/blob/master/SimpleCV/ImageClass.py#L1834

Upload is using the file name. This is not cool as a lot of images don't have a filename, or the patch could change. Correct behavior is to save to a temporary file prior to upload.