I have a problem with my pixel_at function in my code. Here are all necessary code snippets.
Here is my Helper module:
import gtk.gdk
def pixel_at(x, y):
gtk.gdk
rw = gtk.gdk.get_default_root_window()
pixbuf = gtk.gdk.Pixbuf(gtk.gdk.COLORSPACE_RGB, False, 8, 1, 1)
pixbuf = pixbuf.get_from_drawable(rw, rw.get_colormap(), x, y, 0, 0, 1, 1)
return tuple(pixbuf.pixel_array[0, 0])
Here is the Controller class for initilaizing:
import Sensors as sensor
import Scanner as sc
import pyautogui
import time
class GameController(object):
COLOR_DINOSAUR = (83, 83, 83)
def __init__(self):
self.offset = (230, 270) #hardcoded Offset
self.width = 600 # hardcoded but alwways the same
# Stores points (jumps)
self.points = 0
#Listeners
self.onGameEnd = None
self.onGameStart = None
self.onSensorData = None
#Game State
self.gameState = 'OVER'
self.sensor = sensor.Sensors()
self.Scanner = sc.Scanner()
Here is my main function:
import pyautogui
import Helper
import GameController as GC
def main():
print Helper.pixel_at(25, 5)
Controller = GC.GameController()
# Just test values for this function
Controller.Scanner.scanUntil([75, 124],
(-2, 0),
(83, 83, 83),
75 / 2)
Here is the code for the Scanner:
import pyautogui
import Helper
def scanUntil(self, start, step, matchColor, iterLimit):
iterations = 0
current = self.makeInBounds(start)
if step[X] == 0 and step[Y] == 0:
return None
while not self.isOutOfBound(current):
color = Helper.pixel_at(current[X], current[Y])
# to slow color = pyautogui.pixel(current[X], current[Y])
if color == matchColor: # Tuple comparison
return current
current[X] += step[X]
current[Y] += step[Y]
iterations += 1
if iterations > iterLimit:
return None
return None
This line color = Helper.pixel_at(current[X], current[Y]) in the Scanner throws the error: Fatal IO error 0 (Success) on X server :0. But the print Helper.pixel_at(25, 5)call in main() returns me the correct RGB tuple.
The error is thrown at pixbuf = pixbuf.get_from_drawable(rw, rw.get_colormap(), x, y, 0, 0, 1, 1) in the pixel_at(x,y) so I think it has something to do with this.
From the documentation at http://www.pygtk.org/pygtk2reference/class-gdkpixbuf.html#method-gdkpixbuf--get-from-drawable I got this clue: In other words, copies image data from a server-side drawable to a client-side RGB(A) buffer.
But I don´t understand this server-side and client side and why is it working in my first print call but not for color = Helper.pixel_at(current[X], current[Y])?
Related
I am trying to read all the touching pixels with the same color in a image.
For that I use reccursive functions. When I check one pixel, I look on the right, left, top and bottom if the pixel close to it is the same color. If it is I add it to an array otherwise I don't.
The code is as follow:
vimport tkinter as tk
from PIL import Image
import sys
sys.setrecursionlimit(200000)
## WINDOWS
# to launch in debug mode
imgToDraw = Image.open('assets-test\\smile-face.png')
# to launch normaly
# imgToDraw = Image.open('..\\assets-test\\smile-face.png')
## LINUX
# imgToDraw = Image.open('../assets-test/smile-face.png')
imgPixels = imgToDraw.load()
imgWidth = imgToDraw.size[0]
imgHeight = imgToDraw.size[1]
# an element is a part of the image, it's a bunch of pixels with approximately the same color
# and each pixel touch at least one other pixel of the same element
elements = [];
isPixelChecked = [[ False for y in range( imgWidth ) ] for x in range( imgHeight )]
# min tolerable difference between two colors to consider them the same
# the higher the value is the more colors will be considered the same
COLOR_TOLERANCE = 10
reccursionCount = 0
class Element:
def __init__(self, color):
self.pixels = [];
self.color = color;
def addPixel(self, pixel):
self.pixels.append(pixel);
class Pixel:
def __init__(self, x, y, color):
self.x = x # x position of the pixel
self.y = y # y position of the pixel
self.color = color # color is a tuple (r,g,b)
def cutImageInElements():
global element
completeElement(element.pixels)
def completeElement(elemPixels):
global reccursionCount
global isPixelChecked
reccursionCount += 1
nbPixels = len(elemPixels);
xIndex = elemPixels[nbPixels - 1].x
yIndex = elemPixels[nbPixels - 1].y
xRightIdx = elemPixels[nbPixels - 1].x + 1
xLeftIdx = elemPixels[nbPixels - 1].x - 1
yBottomIdx = elemPixels[nbPixels - 1].y + 1
yTopIdx = elemPixels[nbPixels - 1].y - 1
isPixelChecked[xIndex][yIndex] = True
if((xRightIdx < imgWidth) and isPixelChecked[xRightIdx][yIndex] == False):
if(isColorAlmostSame(imgPixels[elemPixels[0].x, elemPixels[0].y], imgPixels[xRightIdx, yIndex])):
pixelAppended = Pixel(xRightIdx, yIndex, imgPixels[xRightIdx, yIndex])
elemPixels.append(pixelAppended)
completeElement(elemPixels)
if((xLeftIdx >= 0) and isPixelChecked[xLeftIdx][yIndex] == False):
if(isColorAlmostSame(imgPixels[elemPixels[0].x, elemPixels[0].y], imgPixels[xLeftIdx, yIndex])):
pixelAppended = Pixel(xLeftIdx, yIndex, imgPixels[xLeftIdx, yIndex])
elemPixels.append(pixelAppended)
completeElement(elemPixels)
if((yBottomIdx < imgHeight) and isPixelChecked[xIndex][yBottomIdx] == False):
if(isColorAlmostSame(imgPixels[elemPixels[0].x, elemPixels[0].y], imgPixels[xIndex, yBottomIdx])):
pixelAppended = Pixel(xIndex, yBottomIdx, imgPixels[xIndex, yBottomIdx])
elemPixels.append(pixelAppended)
completeElement(elemPixels)
if((yTopIdx >= 0) and isPixelChecked[xIndex][yTopIdx] == False):
if(isColorAlmostSame(imgPixels[elemPixels[0].x, elemPixels[0].y], imgPixels[xIndex, yTopIdx])):
pixelAppended = Pixel(xIndex, yTopIdx, imgPixels[xIndex, yTopIdx])
elemPixels.append(pixelAppended)
completeElement(elemPixels)
def isColorAlmostSame(pixel1, pixel2):
redDiff = abs(pixel1[0] - pixel2[0])
greenDiff = abs(pixel1[1] - pixel2[1])
blueDiff = abs(pixel1[2] - pixel2[2])
if(redDiff < COLOR_TOLERANCE and greenDiff < COLOR_TOLERANCE and blueDiff < COLOR_TOLERANCE):
return True
else:
return False
def printPixelsArr(pixelsArr):
for x in range(0, len(pixelsArr)):
print(pixelsArr[x].x, pixelsArr[x].y, pixelsArr[x].color)
if __name__ == '__main__':
pixel = Pixel(0, 0, imgPixels[0, 0]);
element = Element(pixel.color);
element.addPixel(pixel);
cutImageInElements();
print("NbReccursive call: ", reccursionCount)
This code works for small images of size 100x100 but crashes with an image of 400x400 with the error "terminated by signal SIGSEGV (Address boundary error)" when I launch the program on wsl2. When I run the program on cmd or powershell it just crashes but with no error code/msg.
I cannot understand why it would work with some size of images and not others. I can only think that the memory runs out or something but in the task manager the program uses almost no memory.
Not sure why that's failing, but that much recursion in Python isn't a great idea. I'd suggest reading about tail recursion that other languages use to make some recursive algorithms consume constant stack space. Note that your algorithm is not tail recursive, so this optimisation wouldn't help even if Python supported it.
I hacked together the following flood fill implementation. It uses Numpy so that it's only 10x slower than Pillow's ImageDraw.floodfill.
import numpy as np
def floodfill(im, row, col, threshold):
similar = np.mean(np.abs(im - im[row, col]), 2) < threshold
mask = np.zeros_like(similar)
mask[row, col] = 1
m2 = mask.copy()
while True:
m2[:,:] = mask
m2[1:,:] |= mask[:-1]
m2[:-1,:] |= mask[1:]
m2[:,1:] |= mask[:,:-1]
m2[:,:-1] |= mask[:,1:]
m2 &= similar
if np.all(m2 == mask):
return mask
mask[:,:] = m2
As an example of using this, you could do;
import requests
from io import BytesIO
res = requests.get("https://picsum.photos/300")
res.raise_for_status()
src = Image.open(BytesIO(res.content))
mask = floodfill(np.array(src, int), 10, 10, 40)
where the random image I got and the output mask are:
I have been, with pygame and pyopengl (oh and obviously Python), been trying to make a little tile-based 2D game, but I'm having trouble with the rendering using VBOs and glMultiDrawArray().
The program runs without errors, but I don't see any geometry drawn, so it's just a blank screen.
I've tried using glTranslate to see if maybe the geometry is being drawn, but I can't see it, as well as changing between using GluPerspective() and glOrthro2D(). No luck. I've pored over the code to see where it isn't working, but I have no clue what could be wrong. I'm still struggling to understand OpenGL and VBOs.
Here are the relevant bits of my code:
The Chunk class. Every chunk has its own VBO for vertices and textures (textures are currently unused)
class Chunk():
def __init__(self, position):
self.Position = position
self.VertexVBOId = _get_chunk_id()
self.VertexVBO = glGenBuffers(self.VertexVBOId)
self.TextureVBOId = _get_chunk_id()
self.TextureVBO = glGenBuffers(self.TextureVBOId)
Chunks[str(position)] = self
self.__updateVertexArray()
# glBindBuffer (GL_ARRAY_BUFFER, self.VertexVBO)
#self.__updateVertexArray()
def __getvertices(self):
vertices = []
for x in range(self.Position.x, self.Position.x + 16):
for y in range(self.Position.y, self.Position.y + 16):
pos = Vector2(x, y)
tile = GetTile(pos)
if tile != "air":
vertices.append(x+1)
vertices.append(y)
vertices.append(x+1)
vertices.append(y+1)
vertices.append(x)
vertices.append(y+1)
vertices.append(x)
vertices.append(y)
return vertices
def __updateVertexArray(self): #This will be called when a change is made the the chunk, as well as once initially
print("UPDATING VERTEX ARRAY")
vertices = self.__getvertices()
glBindBuffer (GL_ARRAY_BUFFER, self.VertexVBOId)
glBufferData (GL_ARRAY_BUFFER, len(vertices)*4, (c_float*len(vertices))(*vertices), GL_DYNAMIC_DRAW)
And here is the rendering loop:
def main():
print("Started")
pygame.init()
global displaySize
global SCREENSIZE
global PIXELS_PER_TILE
pygame.display.set_mode(displaySize, DOUBLEBUF|OPENGL)
#gluOrtho2D(-SCREENSIZE[0]/2, SCREENSIZE[0]/2, -SCREENSIZE[1]/2, SCREENSIZE[1]/2)
gluPerspective(180, 2, 0.1, 100)
... some other stuff ...
while True:
#Drawing
glClearColor(0.7, 0.7, 1, 0)
glClear(GL_COLOR_BUFFER_BIT|GL_DEPTH_BUFFER_BIT)
cameraTranslateX = (camera.Position.x % 1) * PIXELS_PER_TILE
cameraTranslateY = (camera.Position.y % 1) * PIXELS_PER_TILE
#Figure out which chunks to render
botLeft = camera.Position - Vector2(SCREENSIZE[0]/2, SCREENSIZE[1]/2) + Vector2(cameraTranslateX, cameraTranslateY)
topRight = camera.Position + Vector2(SCREENSIZE[0]/2, SCREENSIZE[1]/2) + Vector2(cameraTranslateX, cameraTranslateY)
FirstChunkPos = (botLeft/16).floor()
TotalChunksX = (topRight/16).ceil().x - FirstChunkPos.x
TotalChunksY = (topRight/16).ceil().y - FirstChunkPos.y
for x in range(TotalChunksX):
for y in range(TotalChunksY):
pos = Vector2(x + FirstChunkPos.x, y + FirstChunkPos.y)
chunk = Chunks.get(str(pos))
if not chunk:
chunk = Chunk(pos)
VertexVBO = chunk.VertexVBOId
glBindBuffer (GL_ARRAY_BUFFER, VertexVBO)
glVertexPointer (2, GL_FLOAT, 0, None)
TextureVBO = chunk.TextureVBOId
glMultiDrawArrays (GL_POLYGON, vertexArrayThingy1, vertexArrayThingy2, 255)
# glUnmapBuffer(GL_ARRAY_BUFFER,VertexVBO)
pygame.display.flip ()
The 2nd and 3rd arguments of glMultiDrawArrays are of type const GLint* and const GLsizei*. This function cannot draw from different buffers. There is no glDraw* command that can use multiple vertex buffers for drawing. All vertecx attributes must be in one and the same buffer. glMultiDrawArrays should draw different ranges from the buffer. Suppose you want to draw the following 3 attribute ranges [3:6], [18:27], [30:36]:
first = [3, 18, 30]
count = [3, 9, 6]
glMultiDrawArrays(GL_TRIANGLES, first, count, 3)
If you want to draw multiple lists of indices you have to use glMultiDrawElements and you have to create an array of pointers to arrays of indices:
import ctypes
ia1 = (GLuint * 6)(0, 1, 2, 0, 2, 3)
ia2 = (GLuint * 6)(12, 13, 14, 12, 14, 15)
counts = [6, 6]
indexPtr = (GLvoidp * 2)(ctypes.addressof(ia1), ctypes.addressof(ia2))
glMultiDrawElements(GL_TRIANGLES, counts, GL_UNSIGNED_INT, indexPtr, 2)
I am currently experimenting with the pytest module to create unit tests for a project I'm working on. I'm trying to test the 'add_point' method which draws an ellipse based on a set of pixels. What I want to do is inspect 'draw' to ensure that the ellipse has been created successfully. Unfortunately I don't know how to go about this, so any help will be appreciated. Here's my code so far:
(A) TheSlicePreviewMaker.py
import os, Image, ImageDraw, ImageFont
from json_importer import json_importer
class SlicePreviewer(object):
def __init__(self):
self.screen_size = (470, 470)
self.background_colour = (86,0,255)
self.platform_fill_colour = (100, 100, 100)
self.platform_outline_colour = (0, 0, 0)
self.platform_window = (0,0,469,469)
self.point_colour = (0,0,255)
self.config_object = json_importer("ConfigFile.txt")
self.image = None
def initialise_image(self):
self.image = Image.new('RGB',self.screen_size,self.background_colour)
draw = ImageDraw.Draw(self.image)
draw.rectangle(self.platform_window,outline=self.platform_outline_colour,fill=self.platform_fill_colour)
del draw
def add_point(self, px, py):
x1 = px - 1
y1 = py - 1
x2 = px + 1
y2 = py + 1
draw = ImageDraw.Draw(self.image)
draw.ellipse((x1,y1,x2,y2),outline=self.point_colour,fill=self.point_colour)
return draw #del draw
def save_image(self, file_name):
self.image.save(file_name, "BMP")
(B) test_TheSlicePreviewMaker.py
from TheSlicePreviewMaker import SlicePreviewer
slice_preview = SlicePreviewer()
class TestSlicePreviewer:
def test_init(self):
'''check that the config file object has been created on init'''
assert slice_preview.config_object != None
def test_initialise_image(self):
'''verify if the image has been successfully initialised'''
assert slice_preview.image.mode == 'RGB'
def test_add_point(self):
'''has the point been drawn successfully?'''
draw = slice_preview.add_point(196,273)
assert something
import pytest
if __name__ == '__main__':
pytest.main("--capture=sys -v")
SN: I've run TheSlicePreviewMaker.py separately to check the bitmap file it produces, so I know that the code works. What I want to achieve is unit test this so that each time I don't have to go check the bitmap.
One approach is to manually inspect the generated image and if looks OK to you, save it next to the test and use a image diffing algorithm (for example ImageChops.difference) to obtain a threshold value that you can use to make sure future test runs are still drawing the same image.
For example:
# contents of conftest.py
from PIL import ImageChops, ImageDraw, Image
import pytest
import os
import py.path
import math
import operator
def rms_diff(im1, im2):
"""Calculate the root-mean-square difference between two images
Taken from: http://snipplr.com/view/757/compare-two-pil-images-in-python/
"""
h1 = im1.histogram()
h2 = im2.histogram()
def mean_sqr(a,b):
if not a:
a = 0.0
if not b:
b = 0.0
return (a-b)**2
return math.sqrt(reduce(operator.add, map(mean_sqr, h1, h2))/(im1.size[0]*im1.size[1]))
class ImageDiff:
"""Fixture used to make sure code that generates images continues to do so
by checking the difference of the genereated image against known good versions.
"""
def __init__(self, request):
self.directory = py.path.local(request.node.fspath.dirname) / request.node.fspath.purebasename
self.expected_name = (request.node.name + '.png')
self.expected_filename = self.directory / self.expected_name
def check(self, im, max_threshold=0.0):
__tracebackhide__ = True
local = py.path.local(os.getcwd()) / self.expected_name
if not self.expected_filename.check(file=1):
msg = '\nExpecting image at %s, but it does not exist.\n'
msg += '-> Generating here: %s'
im.save(str(local))
pytest.fail(msg % (self.expected_filename, local))
else:
expected = Image.open(str(self.expected_filename))
rms_value = rms_diff(im, expected)
if rms_value > max_threshold:
im.save(str(local))
msg = '\nrms_value %s > max_threshold of %s.\n'
msg += 'Obtained image saved at %s'
pytest.fail(msg % (rms_value, max_threshold, str(local)))
#pytest.fixture
def image_diff(request):
return ImageDiff(request)
Now you can use the image_diff fixture in your tests. For example:
def create_image():
""" dummy code that generates an image, simulating some actual code """
im = Image.new('RGB', (100, 100), (0, 0, 0))
draw = ImageDraw.Draw(im)
draw.ellipse((10, 10, 90, 90), outline=(0, 0, 255),
fill=(255, 255, 255))
return im
def test_generated_image(image_diff):
im = create_image()
image_diff.check(im)
The first time your run this test, it will fail with this output:
================================== FAILURES ===================================
____________________________ test_generated_image _____________________________
image_diff = <test_foo.ImageDiff instance at 0x029ED530>
def test_generated_image(image_diff):
im = create_image()
> image_diff.check(im)
E Failed:
E Expecting image at X:\temp\sandbox\img-diff\test_foo\test_generated_image.png, but it does not exist.
E -> Generating here: X:\temp\sandbox\img-diff\test_generated_image.png
You can then manually check the image and if everything is OK, move it to a directory with the same name as the test file, with the name of the test as the file name plus ".png" extension. From now one whenever the test runs, it will check that the image is similar within an acceptable amount.
Suppose you change the code and produce a slightly different image, the test will now fail like this:
================================== FAILURES ===================================
____________________________ test_generated_image _____________________________
image_diff = <test_foo.ImageDiff instance at 0x02A4B788>
def test_generated_image(image_diff):
im = create_image()
> image_diff.check(im)
E Failed:
E rms_value 2.52 > max_threshold of 0.0.
E Obtained image saved at X:\temp\sandbox\img-diff\test_generated_image.png
test_foo.py:63: Failed
========================== 1 failed in 0.03 seconds ===========================
The code needs some polishing but should be a good start. You can find a version of this code here.
Cheers,
i would like to do some program by capture image from webcam, then cropped it. after crop, i do some image processing and from the process it will run my robots. Here the full program:
import cv2
from cv2 import *
import numpy as np
import pylab
import pymorph
import mahotas
from matplotlib import pyplot
from PIL import Image
# initialize the camera
cam = VideoCapture(0) # 0 -> index of camera
s, img = cam.read()
# frame captured without any errors
if s:
imwrite("img.jpg",img) #save image
#Crop Image
imageFile = "img.jpg"
im1 = Image.open(imageFile)
def imgCrop(im):
box = (0, 199, 640, 200)
region = im.crop(box)
region.save('crop.jpg')
cImg = imgCrop(im1)
#thresholding
def greyImg(im):
gray = im.convert('L')
bw = gray.point(lambda x: 0 if x<128 else 255, '1')
bw.save("bw.jpg")
tImg = greyImg(cImg )
#direction
def find_centroid(im, rez):
width, height = im.size
XX, YY, count = 0, 0, 0
for x in xrange(0, width, rez):
for y in xrange(0, height, rez):
if im.getpixel((x, y)) == 255:
XX += x
YY += y
count += 1
return XX/count, YY/count
print find_centroid(tImg, 1)
def robo_direct():
cen = find_centroid(im, 1)
diff = cen[0] - 320
if diff > 10:
print 'right'
if diff < -10:
print 'left'
else:
print 'straight'
print robo_direct()
The error was come out like this:
File "compile.py", line 32, in greyImg
gray = im.convert('L')
AttributeError: 'NoneType' object has no attribute 'convert'
That is because im is a None object.
Try again the code with:
print im is None
And you'll see. I don't know about threshold, but obviously you are creating the im object the wrong way.
Your function imgCrop(im1) has no return statement and as such returns None. And then your greyImg(im) function also has no return statement and also will return None.
To fix that add return statements to both functions that for the first return region and the second return bw.
Also your robo_direct() function should return and not print the direction so that the call to it in the statement print robo_direct() would print the direction.
I've gotten OpenCV working with Python and I can even detect a face through my webcam. What I really want to do though, is see movement and find the point in the middle of the blob of movement. The camshift sample is close to what I want, but I don't want to have to select which portion of the video to track. Bonus points for being able to predict the next frame.
Here's the code I have currently:
#!/usr/bin/env python
import cv
def is_rect_nonzero(r):
(_,_,w,h) = r
return (w > 0) and (h > 0)
class CamShiftDemo:
def __init__(self):
self.capture = cv.CaptureFromCAM(0)
cv.NamedWindow( "CamShiftDemo", 1 )
self.storage = cv.CreateMemStorage(0)
self.cascade = cv.Load("/usr/local/share/opencv/haarcascades/haarcascade_mcs_upperbody.xml")
self.last_rect = ((0, 0), (0, 0))
def run(self):
hist = cv.CreateHist([180], cv.CV_HIST_ARRAY, [(0,180)], 1 )
backproject_mode = False
i = 0
while True:
i = (i + 1) % 12
frame = cv.QueryFrame( self.capture )
if i == 0:
found = cv.HaarDetectObjects(frame, self.cascade, self.storage, 1.2, 2, 0, (20, 20))
for p in found:
# print p
self.last_rect = (p[0][0], p[0][1]), (p[0][2], p[0][3])
print self.last_rect
cv.Rectangle( frame, self.last_rect[0], self.last_rect[1], cv.CV_RGB(255,0,0), 3, cv.CV_AA, 0 )
cv.ShowImage( "CamShiftDemo", frame )
c = cv.WaitKey(7) % 0x100
if c == 27:
break
if __name__=="__main__":
demo = CamShiftDemo()
demo.run()
Found a solution at How do I track motion using OpenCV in Python?