Batch operation for generating Label image from ROIs and vice versa


I am trying to label cells in the microscopy image which can later be used as label image for training a deep learning model.

Label image: an image where all background pixels has zero value, and all other individual cell has a unique value. ()

While searching for many ways to do so, I came across a very useful plugin for ImageJ named [ijp-LaRoMe](https://BIOP/ijp-LaRoMe: some useful function to get Label from ROIs and vice versa, and more! (

This plugin makes it very easy to convert to and fro from label image to ROIs and measure many properties of ROIs. I was wondering if it’s possible to batch process many images (maybe all the images available in a specific folder)?

I already created an issue in the github repo. As pointed out by @romainGuiet, it will be useful for many other users who are active on So, I am creating a new topic here. @romainGuiet could you please have a look at this, and let me know if you can guide in any possible way.

Thank you and best regards,

Hi @its_jd ,

If you want to make the Labels from ROIs, you will need to have ROIs in the RoiManager and to have an image open. Indeed, we assume the ROIs were detected on an image and we’ll use the dimension of this image to create the Label image.

Please find below a macro language example that asks you the folder with the roisets and the size of the image (assume it’s a square image, and all images from a folder have same dimension).

#@File (style="directory") dir 
#@Integer (value=256) sizeImage

run("Close All");

newImage("empty", "8-bit black", sizeImage, sizeImage, 1);
emptyID= getImageID();

dir += File.separator
fileList = getFileList(dir)

for( i = 0 ; i <fileList.length ; i++){
	if ( endsWith(fileList[i], "zip")){
		roiManager( "open", dir+fileList[i] );
		run("ROIs to Label image", "");
		saveAs("tiff", dir+"label_"+fileList[i]+".tiff");

Please let me know if this works or if you need more help, in that case please provide me with a example of images+rois.





I have a jython script that does this, it has a simple to use GUI where you enter the directory path of raw images, roi sets and save path for the mask directory and label mask directory. It then creates binary images in the binary image folder and labelled images in label image folder that can be used for training.

#@ File(label='Choose Roi directory', style='directory') roidir

#@ File(label='Choose OriginalImage directory', style='directory') originaldir
#@ String(label='File types', value='tif') file_type_image
#@ String(label='Filter', value='._') filter_image
#@ Boolean(label='Recursive search', value=True) do_recursive
#@ File(label='Choose SaveMask directory', style='directory') maskdir
#@ File(label='Choose SaveLabelMask directory', style='directory') masklabeldir

import os
from import File

from ij import IJ
from ij.plugin.frame import RoiManager
from ij import WindowManager as wm

def batch_open_images(pathImage, pathRoi, pathMask, pathLabelMask, file_typeImage=None,  name_filterImage=None,  recursive=False):
    '''Open all files in the given folder.
    :param path: The path from were to open the images. String and are allowed.
    :param file_type: Only accept files with the given extension (default: None).
    :param name_filter: Reject files that contain the given string (default: wild characters).
    :param recursive: Process directories recursively (default: False).
    # Converting a File object to a string.
    if isinstance(pathImage, File):
        pathImage = pathImage.getAbsolutePath()

    def check_type(string):
        '''This function is used to check the file type.
        It is possible to use a single string or a list/tuple of strings as filter.
        This function can access the variables of the surrounding function.
        :param string: The filename to perform the check on.
        if file_typeImage:
            # The first branch is used if file_type is a list or a tuple.
            if isinstance(file_typeImage, (list, tuple)):
                for file_type_ in file_typeImage:
                    if string.endswith(file_type_):
                        # Exit the function with True.
                        return True
                        # Next iteration of the for loop.
            # The second branch is used if file_type is a string.
            elif isinstance(file_typeImage, string):
                if string.endswith(file_typeImage):
                    return True
                    return False
            return False
        # Accept all files if file_type is None.
            return True

    def check_filter(string):
        '''This function is used to check for a given filter.
        It is possible to use a single string or a list/tuple of strings as filter.
        This function can access the variables of the surrounding function.
        :param string: The filename to perform the filtering on.
        if name_filterImage:
            # The first branch is used if name_filter is a list or a tuple.
            if isinstance(name_filterImage, (list, tuple)):
                for name_filter_ in name_filterImage:
                    if name_filter_ in string:
                        # Exit the function with True.
                        return True
                        # Next iteration of the for loop.
            # The second branch is used if name_filter is a string.
            elif isinstance(name_filterImage, string):
                if name_filterImage in string:
                    return True
                    return False
            return False
        # Accept all files if name_filter is None.
            return True


    # We collect all files to open in a list.
    path_to_Image = []
    # Replacing some abbreviations (e.g. $HOME on Linux).
    path = os.path.expanduser(pathImage)
    # If we don't want a recursive search, we can use os.listdir().
    if not recursive:
        for file_name in os.listdir(pathImage):
            full_path = os.path.join(pathImage, file_name)
            if os.path.isfile(full_path):
                if check_type(file_name):
                    if check_filter(file_name):
    # For a recursive search os.walk() is used.
        # os.walk() is iterable.
        # Each iteration of the for loop processes a different directory.
        # the first return value represents the current directory.
        # The second return value is a list of included directories.
        # The third return value is a list of included files.
        for directory, dir_names, file_names in os.walk(pathImage):
            # We are only interested in files.
            for file_name in file_names:
                # The list contains only the file names.
                # The full path needs to be reconstructed.
                full_path = os.path.join(directory, file_name)
                # Both checks are performed to filter the files.
                if check_type(file_name):
                    if check_filter(file_name) is False:
                        # Add the file to the list of images to open.
                        path_to_Image.append([full_path, os.path.basename(os.path.splitext(full_path)[0])])
    # Create the list that will be returned by this function.
    Images = []
    Rois = []
    for img_path, file_name in path_to_Image:
        # IJ.openImage() returns an ImagePlus object or None.
        imp = IJ.openImage(img_path)
        if check_filter(file_name):
         print(file_name  ,  pathRoi)
        RoiName = str(pathRoi) + '/'+ file_name + '_rois' + '.zip'
        if os.path.exists(RoiName):
		         Roi =
		         # An object equals True and None equals False.
		         rm = RoiManager.getInstance()
		         if (rm==None):
		            rm = RoiManager()
		         rm.runCommand("Open", RoiName)
		         impMask = IJ.createImage("Mask", "8-bit grayscale-mode", imp.getWidth(), imp.getHeight(), imp.getNChannels(), imp.getNSlices(), imp.getNFrames())
		         IJ.setForegroundColor(255, 255, 255)
		         IJ.saveAs(impMask, '.tif', str(pathMask) + "/"  +  file_name);
		  "Find Connected Regions", "allow_diagonal display_one_image display_results regions_for_values_over=100 minimum_number_of_points=1 stop_after=-1")
		         IJ.saveAs("Tiff",str(pathLabelMask) + "/"  +  file_name);
		         #print(img_path, RoiName)
    return Images, Rois

def split_string(input_string):
    '''Split a string to a list and strip it
    :param input_string: A string that contains semicolons as separators.
    string_splitted = input_string.split(';')
    # Remove whitespace at the beginning and end of each string
    strings_striped = [string.strip() for string in string_splitted]
    return strings_striped

if __name__ in ['__builtin__','__main__']:
    # Run the batch_open_images() function using the Scripting Parameters.
    images = batch_open_images(originaldir,roidir ,maskdir,masklabeldir,
    for image in images:
        # Call the toString() method of each ImagePlus object.

It calls the Fiji plugin to get connected components, do you think skimage.measure can be used instead to label the regions instead though? This will be pure Fiji solution though.

1 Like