How to use the identify.O_FOREGROUND function in identify

new_setting_values.append(cps.NO)
            else:
                new_setting_values.append(cps.YES)
            setting_values = new_setting_values
            from_matlab = False
            variable_revision_number = 1
        if from_matlab:
            NotImplementedError("Don't know how to convert Matlab IdentifySecondary revision # %d"%(variable_revision_number))
        if variable_revision_number != self.variable_revision_number:
            NotImplementedError("Don't know how to handle IdentifySecondary revision # %d"%(variable_revision_number))
        if (not from_matlab) and variable_revision_number == 1:
            # Removed test mode
            # added Otsu parameters.
            setting_values = setting_values[:11]+setting_values[12:]
            setting_values += [cpmi.O_TWO_CLASS, cpmi.O_WEIGHTED_VARIANCE,
                               cpmi.O_FOREGROUND]
            variable_revision_number = 2
        if (not from_matlab) and variable_revision_number == 2:
            # Added discarding touching
            setting_values = setting_values + [cps.NO, cps.NO, "FilteredNuclei"]
            variable_revision_number = 3
        if (not from_matlab) and variable_revision_number == 3:
            # Added new primary outlines
            setting_values = setting_values + [cps.NO, "FilteredNucleiOutlines"]
            variable_revision_number = 4
        return setting_values, variable_revision_number, from_matlab

new_setting_values[UNCLUMP_METHOD_VAR] = UN_NONE
            if new_setting_values[WATERSHED_VAR] == cps.DO_NOT_USE:
                new_setting_values[WATERSHED_VAR] = WA_NONE
            variable_revision_number = 1
            from_matlab = False
        if (not from_matlab) and variable_revision_number == 1:
            # Added LOG method
            setting_values = list(setting_values)
            setting_values += [ cps.YES, ".5" ]
            variable_revision_number = 2
        
        if (not from_matlab) and variable_revision_number == 2:
            # Added Otsu options
            setting_values = list(setting_values)
            setting_values += [cpmi.O_TWO_CLASS, cpmi.O_WEIGHTED_VARIANCE,
                               cpmi.O_FOREGROUND]
            variable_revision_number = 3
        
        if (not from_matlab) and variable_revision_number == 3:
            # Added more LOG options
            setting_values = setting_values + [cps.YES, "5"]
            variable_revision_number = 4
        
        if (not from_matlab) and variable_revision_number == 4:
            # Added # of object limits
            setting_values = setting_values + [LIMIT_NONE, "500"]
            variable_revision_number = 5
            
        if (not from_matlab) and variable_revision_number == 5:
            # Changed object number limit option from "No action" to "Continue"
            if setting_values[-2] == "No action":
                setting_values[-2] = LIMIT_NONE

def calculate_threshold(self, image_group, workspace):
        '''Calculate a threshold for this image'''
        image_name = image_group.image_name.value
        image = workspace.image_set.get_image(image_name,
                                              must_be_grayscale = True)
        threshold_method = image_group.threshold_algorithm
        object_fraction = image_group.object_fraction.value
        two_class_otsu = (image_group.two_class_otsu == O_TWO_CLASS)
        use_weighted_variance = (image_group.use_weighted_variance == O_WEIGHTED_VARIANCE)
        assign_middle_to_foreground = (image_group.assign_middle_to_foreground == O_FOREGROUND)
        (local_threshold, global_threshold) = \
            (cpthresh.get_threshold(threshold_method,
                                    cpthresh.TM_GLOBAL,
                                    image.pixel_data,
                                    mask = image.mask,
                                    object_fraction = object_fraction,
                                    two_class_otsu = two_class_otsu,
                                    use_weighted_variance = use_weighted_variance,
                                    assign_middle_to_foreground = assign_middle_to_foreground)
             if image.has_mask
             else
             cpthresh.get_threshold(threshold_method,
                                    cpthresh.TM_GLOBAL,
                                    image.pixel_data,
                                    object_fraction = object_fraction,
                                    two_class_otsu = two_class_otsu,