We Keep Coding

UNABE to successfully LOAD GRAYSCALE IMAGE DATASETS in PYTHON via SPYDER IDE in ANACONDA

I need your help to modify the following code!
import os
from PIL import Image
from torch.utils.data import Dataset
import numpy as np
class CarvanaDataset(Dataset):
def __init__(self, image_dir, mask_dir, transform=None):
self.image_dir = image_dir
self.mask_dir = mask_dir
self.transform = transform
self.images = os.listdir(image_dir)
def __len__(self):
return len(self.images)
def __getitem__(self, index):
img_path = os.path.join(self.image_dir, self.images[index])
mask_path = os.path.join(self.mask_dir, self.images[index].replace(".jpg", "_mask.gif"))
image = np.array(Image.open(img_path).convert("RGB"))
mask = np.array(Image.open(mask_path).convert("L"), dtype=np.float32)
mask[mask == 255.0] = 1.0
if self.transform is not None:
augmentations = self.transform(image=image, mask=mask)
image = augmentations["image"]
mask = augmentations["mask"]
return image, mask
When I used color image datasets (from kaggle as recommended by the author of the code) the above code loads color image datasets successfully.
But, when I changed the dataset and used grayscale image datasets (both images and masks) the script generate the following error:
PIL.UnidentifiedImageError: cannot identify image file 'data/val_images/zoazoa_251.tiff'
Can you please help to fix this?
Note that the script file of interest "dataset.py" can be found here: CLIK

PIL doesn't paste properly

I'm trying to make a little weather forecast program which gives you an image with an overview of the weather in python
To get the weather, I'm using openweathermap and it works ok for me, but for the image I'm using PIL to paste the weather icon, but for some reason there's a part of it that's not being pasted, here you can see what the icon should be: https://openweathermap.org/img/wn/04n#2x.png, and here's how it appeared in the image that came out of my script:
Here's the part of the code that generates the image:
def drawImage(d):
img=PIL.Image.open("base.png")
url=f"https://openweathermap.org/img/wn/{d['icon']}#2x.png"
weatherIcon=Image.open(requests.get(url, stream=True).raw)
print(url)
img.paste(weatherIcon, (00, 10))
now=datetime.now()
name="boards/"+now.strftime('%d_%m_%Y_%H_%M_%S')+".png"
img.save(name)
return name
Some notes on this code:
The base.png is just a 720x720 blank image
The d that gets passed in is a dictionary with all the information, but here it only needs the icon, so I'll give this example: {"icon": "04n"}
I got the URL for the image from the website of OpenWeatherMap, see documentation: https://openweathermap.org/weather-conditions

This is happening because the icon image you download has transparency (an alpha channel). To remove that, you can use this answer.
I've simplified it slightly, define the following function:
def remove_transparency(im, bg_colour=(255, 255, 255)):
if im.mode in ('RGBA', 'LA') or (im.mode == 'P' and 'transparency' in im.info):
alpha = im.getchannel('A')
bg = Image.new("RGBA", im.size, bg_colour + (255,))
bg.paste(im, mask=alpha)
return bg
else:
return im
and call it in your code:
weatherIcon=Image.open(requests.get(url, stream=True).raw)
print(url)
weatherIcon = remove_transparency(weatherIcon)
img.paste(weatherIcon, (00, 10))
You might want to adjust that bg_colour parameter.

Python select ROI OpenCV

Sample Image
Hello,
I created an application in python that select the Region of Interest(ROI) of an image, record and label it. But I has a limit of one ROI per image, anyone know how to have multiple selection of ROI per image? Also on attached image, as you can see I have multiple window, I want it to be in one window with different options, what packages are use on this kind of application.
here's my code in python using opencv2. Thank you in advance for the help
for image in filelist:
img = cv2.imread(image)
fromCenter = False
r = cv2.selectROI(img, fromCenter)
lbl = simpledialog.askstring("Image Label", "Please Enter Label")
result = eTree.SubElement(results, "Image")
path = eTree.SubElement(result, 'Path')
roi = eTree.SubElement(result, 'ROI')
label = eTree.SubElement(result, 'Label')
path.text = str(image)
roi.text = str(r)
label.text = str(lbl)
tree = eTree.ElementTree(results)
i = i + 1
if i == count:
format = [('XML Files', '*.xml'), ('All Files', '*.*')]
save = filedialog.asksaveasfilename(filetype=format, defaultextension='*.xml')
tree.write(save, xml_declaration=True, encoding='utf-8', method="xml")

Well at least for the first part of the question, have you considered to try the cv2.createROIs() instead of cv2.createROI() ? When the image window is opened you then select your first ROI and press enter, then the second and press enter etc. And when you are finished then press the escape key. It returns x,y,w,h of each ROI. Note that you will have to change your code accordingly but it will allow you to select multiple ROI.
Input image:
Example:
import cv2
img = cv2.imread('rois.png')
fromCenter = False
ROIs = cv2.selectROIs('Select ROIs', img, fromCenter)
ROI_1 = img[ROIs[0][1]:ROIs[0][1]+ROIs[0][3], ROIs[0][0]:ROIs[0][0]+ROIs[0][2]]
ROI_2 = img[ROIs[1][1]:ROIs[1][1]+ROIs[1][3], ROIs[1][0]:ROIs[1][0]+ROIs[1][2]]
ROI_3 = img[ROIs[2][1]:ROIs[2][1]+ROIs[2][3], ROIs[2][0]:ROIs[2][0]+ROIs[2][2]]
cv2.imshow('1', ROI_1)
cv2.imshow('2', ROI_2)
cv2.imshow('3', ROI_3)
cv2.waitKey(0)
cv2.destroyAllWindows()
Result:

For custom ROI you can use EasyROI. It supports rectangle, line, circle and polygon.
For using it:
pip install EasyROI
from EasyROI import EasyROI
roi_helper = EasyROI()
roi = roi_helper.draw_rectangle(frame, quantity=2)

NAO fails to save captured image to local computer

I'm trying to save a captured 640x480 RGB image with NAO's front camera to my computer. I'm using python and PIL to do so. Unfortunately, the image just won't save on my computer, no matter what image type or path I use for the parameters of the Image.save()- Method. the image created with PIL contains valid RGB-information though. Here's my code sample from choregraphe:
import Image
def onInput_onStart(self):
cam_input = ALProxy("ALVideoDevice")
nameId = cam_input.subscribeCamera("Test_Cam", 1, 2, 13, 20)
image = cam_input.getImageRemote(nameId) #captures an image
w = image[0] #get the image width
h = image[1] #get the image height
pixel_array = image[6] #contains the image data
result = Image.fromstring("RGB", (w, h), pixel_array)
#the following line doesnt work
result.save("C:\Users\Claudia\Desktop\NAO\Bilder\test.png", "PNG")
cam_input.releaseImage(nameId)
cam_input.unsubscribe(nameId)
pass
Thank you so much for your help in advance!
- a frustrated student

In the comment, you say the code is pasted from choregraphe, so I guess you launch it using choregraphe.
If so, then the code is injected into your robot then started.
So your image is saved to the NAO hard drive and I guess your robot doesn't have a folder named: "C:\Users\Claudia\Desktop\NAO\Bilder\test.png".
So change the path to "/home/nao/test.png", start your code, then log into your NAO using putty or browse folder using winscp (as it looks like you're using windows).
And you should see your image-file.

In order for your code to run correctly it needs to be properly indented. Your code should look like this:
import Image
def onInput_onStart(self):
cam_input = ALProxy("ALVideoDevice")
nameId = cam_input.subscribeCamera("Test_Cam", 1, 2, 13, 20)
image = cam_input.getImageRemote(nameId) #captures an image
w = image[0] #get the image width
h = image[1] #get the image height
pixel_array = image[6] #contains the image data
...
Make sure to indent everything that's inside the def onInput_onStart(self): method.

Sorry for the late response, but it maybe helpful for someone. You should try it with naoqi. Here is the documentation for retriving images
http://doc.aldebaran.com/2-4/dev/python/examples/vision/get_image.html
The original code was not working for me so I made some tweeks.
parser = argparse.ArgumentParser()
parser.add_argument("--ip", type=str, default="nao.local.",
help="Robot IP address. On robot or Local Naoqi: use
'nao.local.'.")
parser.add_argument("--port", type=int, default=9559,
help="Naoqi port number")
args = parser.parse_args()
session = qi.Session()
try:
session.connect("tcp://" + args.ip + ":" + str(args.port))
except RuntimeError:
pass
"""
First get an image, then show it on the screen with PIL.
"""
# Get the service ALVideoDevice.
video_service = session.service("ALVideoDevice")
resolution = 2 # VGA
colorSpace = 11 # RGB
videoClient = video_service.subscribe("python_client",0,3,13,1)
t0 = time.time()
# Get a camera image.
# image[6] contains the image data passed as an array of ASCII chars.
naoImage = video_service.getImageRemote(videoClient)
t1 = time.time()
# Time the image transfer.
print ("acquisition delay ", t1 - t0)
#video_service.unsubscribe(videoClient)
# Now we work with the image returned and save it as a PNG using ImageDraw
# package.
# Get the image size and pixel array.
imageWidth = naoImage[0]
imageHeight = naoImage[1]
array = naoImage[6]
image_string = str(bytearray(array))
# Create a PIL Image from our pixel array.
im = Image.fromstring("RGB", (imageWidth, imageHeight), image_string)
# Save the image.
im.save("C:\\Users\\Lenovo\\Desktop\\PROJEKTI\\python2-
connect4\\camImage.png", "PNG")
Be careful to use Python 2.7.
The code runs on your computer not the NAO robot!

How to take snapshot from given region of the screen with python?

Here is my thoughts:
1) snap shot the given region from the screen like (100,100,80,60), save the result as image
2) process the image with OpenCV python interface
Just be first to python and wonder if this is good solution,be specific,wonder how to snapshot with python.
Thanks,

It's fairly simple using CGRectMake in Apple's CoreGraphics api:
CG.CGRectMake(x, y, w, h)
This allows you to define the horizontal/vertical positions, and width/height respectively.
Code:
#!/usr/bin/python
import Quartz
import LaunchServices
from Cocoa import NSURL
import Quartz.CoreGraphics as CG
def screenshot(path, dpi, region = None):
if region is None:
region = CG.CGRectInfinite
image = CG.CGWindowListCreateImage(
region,
CG.kCGWindowListOptionOnScreenOnly,
CG.kCGNullWindowID,
CG.kCGWindowImageDefault)
url = NSURL.fileURLWithPath_(path)
dest = Quartz.CGImageDestinationCreateWithURL(
url,
LaunchServices.kUTTypePNG, 1, None
)
prop = {
Quartz.kCGImagePropertyDPIWidth: dpi,
Quartz.kCGImagePropertyDPIHeight: dpi,
}
Quartz.CGImageDestinationAddImage(dest, image, prop)
Quartz.CGImageDestinationFinalize(dest)
spec = (0, 0, 800, 600) # x, y, w, h
path = '/path/to/screnshot_region.png' # save path
area = CG.CGRectMake(*spec) # set area to cgrect
screenshot(path, dpi=72, region=area) # call function
To use, just call the function:
screenshot(path, dpi=72, region=area)
*dpi will set the image output resolution; leave out the region argument for fullscreen capture. In regards to the OpenCV portion - I don't use it often enough to provide anything at this time.