ssd_batch_generator.py

'''
Includes:
* A batch generator for SSD model training and inference which can perform online data agumentation
* An offline image processor that saves processed images and adjusted labels to disk

Copyright (C) 2017 Pierluigi Ferrari

This program is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation, either version 3 of the License, or
(at your option) any later version.

This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
GNU General Public License for more details.

You should have received a copy of the GNU General Public License
along with this program.  If not, see <http://www.gnu.org/licenses/>.
'''

import numpy as np
import cv2
import random
from sklearn.utils import shuffle
from copy import deepcopy
from PIL import Image
import csv
import os
from bs4 import BeautifulSoup
import yaml
from pathlib import Path

# Image processing functions used by the generator to perform the following image manipulations:
# - Translation
# - Horizontal flip
# - Scaling
# - Brightness change
# - Histogram contrast equalization

def _translate(image, horizontal=(0,40), vertical=(0,10)):
    '''
    Randomly translate the input image horizontally and vertically.

    Arguments:
        image (array-like): The image to be translated.
        horizontal (int tuple, optinal): A 2-tuple `(min, max)` with the minimum
            and maximum horizontal translation. A random translation value will
            be picked from a uniform distribution over [min, max].
        vertical (int tuple, optional): Analog to `horizontal`.

    Returns:
        The translated image and the horzontal and vertical shift values.
    '''
    rows,cols,ch = image.shape

    x = np.random.randint(horizontal[0], horizontal[1]+1)
    y = np.random.randint(vertical[0], vertical[1]+1)
    x_shift = random.choice([-x, x])
    y_shift = random.choice([-y, y])

    M = np.float32([[1,0,x_shift],[0,1,y_shift]])
    return cv2.warpAffine(image, M, (cols, rows)), x_shift, y_shift

def _flip(image, orientation='horizontal'):
    '''
    Flip the input image horizontally or vertically.
    '''
    if orientation == 'horizontal':
        return cv2.flip(image, 1)
    else:
        return cv2.flip(image, 0)

def _scale(image, min=0.9, max=1.1):
    '''
    Scale the input image by a random factor picked from a uniform distribution
    over [min, max].

    Returns:
        The scaled image, the associated warp matrix, and the scaling value.
    '''

    rows,cols,ch = image.shape

    #Randomly select a scaling factor from the range passed.
    scale = np.random.uniform(min, max)

    M = cv2.getRotationMatrix2D((cols/2,rows/2), 0, scale)
    return cv2.warpAffine(image, M, (cols, rows)), M, scale

def _brightness(image, min=0.5, max=2.0):
    '''
    Randomly change the brightness of the input image.

    Protected against overflow.
    '''
    hsv = cv2.cvtColor(image,cv2.COLOR_RGB2HSV)

    random_br = np.random.uniform(min,max)

    #To protect against overflow: Calculate a mask for all pixels
    #where adjustment of the brightness would exceed the maximum
    #brightness value and set the value to the maximum at those pixels.
    mask = hsv[:,:,2] * random_br > 255
    v_channel = np.where(mask, 255, hsv[:,:,2] * random_br)
    hsv[:,:,2] = v_channel

    return cv2.cvtColor(hsv,cv2.COLOR_HSV2RGB)

def histogram_eq(image):
    '''
    Perform histogram equalization on the input image.

    See https://en.wikipedia.org/wiki/Histogram_equalization.
    '''

    image1 = np.copy(image)

    image1[:,:,0] = cv2.equalizeHist(image1[:,:,0])
    image1[:,:,1] = cv2.equalizeHist(image1[:,:,1])
    image1[:,:,2] = cv2.equalizeHist(image1[:,:,2])

    return image1

class BatchGenerator:
    '''
    A generator to generate batches of samples and corresponding labels indefinitely.

    Shuffles the dataset consistently after each complete pass.

    Currently provides two methods to parse annotation data: A general-purpose CSV parser
    and an XML parser for the Pascal VOC datasets. If the annotations of your dataset are
    in a format that is not supported by these parsers, you could just add another parser
    method and still use this generator.

    Can perform image transformations for data conversion and data augmentation,
    for details please refer to the documentation of the `generate()` method.
    '''

    def __init__(self,
                 images_path,
                 include_classes='all',
                 box_output_format=['class_id', 'xmin', 'xmax', 'ymin', 'ymax']):
        '''
        Arguments:
            images_path (str): The filepath to the image samples.
            include_classes (list, optional): Either 'all' or a list of integers containing the class IDs that
                are to be included in the dataset. Defaults to 'all', in which case all boxes will be included
                in the dataset.
            box_output_format (list, optional): A list of five strings representing the desired order of the five
                items class ID, xmin, xmax, ymin, ymax in the generated data. The expected strings are
                'xmin', 'xmax', 'ymin', 'ymax', 'class_id'. If you want to train the model, this
                must be the order that the box encoding class requires as input. Defaults to
                `['class_id', 'xmin', 'xmax', 'ymin', 'ymax']`. Note that even though the parser methods are
                able to produce different output formats, the SSDBoxEncoder currently requires the format
                `['class_id', 'xmin', 'xmax', 'ymin', 'ymax']`. This list only specifies the five box parameters
                that are relevant as training targets, a list of filenames is generated separately.
        '''
        # These are the variables we always need
        self.images_path = images_path
        self.include_classes = include_classes
        self.box_output_format = box_output_format

        # These are the variables that we only need if we want to use parse_csv()
        self.labels_path = None
        self.input_format = None

        # These are the variables that we only need if we want to use parse_xml()
        self.annotations_path = None
        self.image_set_path = None
        self.image_set = None
        self.classes = None

        # The two variables below store the output from the parsers. This is the input for the generate() method
        # `self.filenames` is a list containing all file names of the image samples. Note that it does not contain the actual image files themselves.
        self.filenames = [] # All unique image filenames will go here
        # `self.labels` is a list containing one 2D Numpy array per image. For an image with `k` ground truth bounding boxes,
        # the respective 2D array has `k` rows, each row containing `(xmin, xmax, ymin, ymax, class_id)` for the respective bounding box.
        self.labels = [] # Each entry here will contain a 2D Numpy array with all the ground truth boxes for a given image

    def parse_bosch_yaml(self, yaml_file=None, ret=False, force_dir=None):
      images = yaml.load(open(yaml_file, 'rb').read())
      self.filenames = []
      self.labels = []


      for i in range(len(images)):
        current_label = []
        image_path = images[i]['path']

        if (force_dir):
          image = Path(image_path).name
          image_path = force_dir + '/' + image

        for boxes in images[i]['boxes']:
          label = []
          # mark state
          state = 0
          if (boxes['label'] == 'Red'):
            state = 1
          elif (boxes['label'] == 'Yellow'):
            state = 2
          elif (boxes['label'] == 'Green'):
            state = 3

          label.append(state)
          label.append(int(boxes['x_min']))
          label.append(int(boxes['x_max']))
          label.append(int(boxes['y_min']))
          label.append(int(boxes['y_max']))
          current_label.append(label)
        if (len(current_label) != 0):
          self.filenames.append(image_path)
          self.labels.append(np.stack(current_label, axis=0))

      if ret:
        return self.filenames, self.labels


    def parse_csv(self,
                  labels_path=None,
                  input_format=None,
                  ret=False):
        '''
        Arguments:
            labels_path (str, optional): The filepath to a CSV file that contains one ground truth bounding box per line
                and each line contains the following six items: image file name, class ID, xmin, xmax, ymin, ymax.
                The six items do not have to be in a specific order, but they must be the first six columns of
                each line. The order of these items in the CSV file must be specified in `input_format`.
                The class ID is an integer greater than zero. Class ID 0 is reserved for the background class.
                `xmin` and `xmax` are the left-most and right-most absolute horizontal coordinates of the box,
                `ymin` and `ymax` are the top-most and bottom-most absolute vertical coordinates of the box.
                The image name is expected to be just the name of the image file without the directory path
                at which the image is located. Defaults to `None`.
            input_format (list, optional): A list of six strings representing the order of the six items
                image file name, class ID, xmin, xmax, ymin, ymax in the input CSV file. The expected strings
                are 'image_name', 'xmin', 'xmax', 'ymin', 'ymax', 'class_id'. Defaults to `None`.
            ret (bool, optional): Whether or not the image filenames and labels are to be returned.
                Defaults to `False`.

        Returns:
            None by default, optionally the image filenames and labels.
        '''

        # If we get arguments in this call, set them
        if not labels_path is None: self.labels_path = labels_path
        if not input_format is None: self.input_format = input_format

        # Before we begin, make sure that we have a labels_path and an input_format
        if self.labels_path is None or self.input_format is None:
            raise ValueError("`labels_path` and/or `input_format` have not been set yet. You need to pass them as arguments.")

        # Erase data that might have been parsed before
        self.filenames = []
        self.labels = []

        # First, just read in the CSV file lines and sort them.

        data = []

        with open(self.labels_path, newline='') as csvfile:
            csvread = csv.reader(csvfile, delimiter=',')
            k = 0
            for i in csvread: # For every line (i.e for every bounding box) in the CSV file...
                if k == 0: # Skip the header row
                    k += 1
                    continue
                else:
                    if self.include_classes == 'all' or int(i[self.input_format.index('class_id')].strip()) in self.include_classes: # If the class_id is among the classes that are to be included in the dataset...
                        obj = [] # Store the box class and coordinates here
                        obj.append(i[self.input_format.index('image_name')].strip()) # Select the image name column in the input format and append its content to `obj`
                        for item in self.box_output_format: # For each item in the output format...
                            obj.append(int(i[self.input_format.index(item)].strip())) # ...select the respective column in the input format and append it to `obj`
                        data.append(obj)

        data = sorted(data) # The data needs to be sorted, otherwise the next step won't give the correct result

        # Now that we've made sure that the data is sorted by file names,
        # we can compile the actual samples and labels lists

        current_file = '' # The current image for which we're collecting the ground truth boxes
        current_labels = [] # The list where we collect all ground truth boxes for a given image
        for idx, i in enumerate(data):
            if current_file == '': # If this is the first image file
                current_file = i[0]
                current_labels.append(i[1:])
                if len(data) == 1: # If there is only one box in the CVS file
                    self.labels.append(np.stack(current_labels, axis=0))
                    self.filenames.append(current_file)
            else:
                if i[0] == current_file: # If this box (i.e. this line of the CSV file) belongs to the current image file
                    current_labels.append(i[1:])
                    if idx == len(data)-1: # If this is the last line of the CSV file
                        self.labels.append(np.stack(current_labels, axis=0))
                        self.filenames.append(current_file)
                else: # If this box belongs to a new image file
                    self.labels.append(np.stack(current_labels, axis=0))
                    self.filenames.append(current_file)
                    current_labels = []
                    current_file = i[0]
                    current_labels.append(i[1:])

        if ret: # In case we want to return these
            return self.filenames, self.labels

    def parse_xml(self,
                  annotations_path=None,
                  image_set_path=None,
                  image_set=None,
                  classes=['background',
                           'aeroplane', 'bicycle', 'bird', 'boat',
                           'bottle', 'bus', 'car', 'cat',
                           'chair', 'cow', 'diningtable', 'dog',
                           'horse', 'motorbike', 'person', 'pottedplant',
                           'sheep', 'sofa', 'train', 'tvmonitor'],
                  exclude_truncated=False,
                  exclude_difficult=False,
                  ret=False):
        '''
        This is an XML parser for the Pascal VOC datasets. It might be applicable to other datasets with minor changes to
        the code, but in its current form it expects the data format and XML tags of the Pascal VOC datasets.

        Arguments:
            annotations_path (str, optional): The path to the directory that contains the annotation XML files for
                the images. The directory must contain one XML file per image and name of the XML file must be the
                image ID. The content of the XML files must be in the Pascal VOC format. Defaults to `None`.
            image_set_path (str, optional): The path to the directory that contains a text file with the image
                set to be loaded. Defaults to `None`.
            image_set (str, optional): The name of the image set text file to be loaded, ending in '.txt'.
                This text file simply contains one image ID per line and nothing else. Defaults to `None`.
            classes (list, optional): A list containing the names of the object classes as found in the
                `name` XML tags. Must include the class `background` as the first list item. The order of this list
                defines the class IDs. Defaults to the list of Pascal VOC classes in alphabetical order.
            exclude_truncated (bool, optional): If `True`, excludes boxes that are labeled as 'truncated'.
                Defaults to `False`.
            exclude_difficult (bool, optional): If `True`, excludes boxes that are labeled as 'difficult'.
                Defaults to `False`.
            ret (bool, optional): Whether or not the image filenames and labels are to be returned.
                Defaults to `False`.

        Returns:
            None by default, optionally the image filenames and labels.
        '''

        if not annotations_path is None: self.annotations_path = annotations_path
        if not image_set_path is None: self.image_set_path = image_set_path
        if not image_set is None: self.image_set = image_set
        if not classes is None: self.classes = classes

        # Erase data that might have been parsed before
        self.filenames = []
        self.labels = []

        # Parse the image set that so that we know all the IDs of all the images to be included in the dataset
        with open(os.path.join(self.image_set_path, self.image_set)) as f:
            image_ids = [line.strip() for line in f]

        # Parse the labels for each image ID from its respective XML file
        for image_id in image_ids:
            # Open the XML file for this image
            with open(os.path.join(self.annotations_path, image_id+'.xml')) as f:
                soup = BeautifulSoup(f, 'xml')

            folder = soup.folder.text # In case we want to return the folder in addition to the image file name. Relevant for determining which dataset an image belongs to.
            filename = soup.filename.text
            self.filenames.append(filename)

            boxes = [] # We'll store all boxes for this image here
            objects = soup.find_all('object') # Get a list of all objects in this image

            # Parse the data for each object
            for obj in objects:
                class_name = obj.find('name').text
                class_id = self.classes.index(class_name)
                # Check if this class is supposed to be included in the dataset
                if (not self.include_classes == 'all') and (not class_id in self.include_classes): continue
                pose = obj.pose.text
                truncated = int(obj.truncated.text)
                if exclude_truncated and (truncated ==1): continue
                difficult = int(obj.difficult.text)
                if exclude_difficult and (difficult == 1): continue
                xmin = int(obj.bndbox.xmin.text)
                ymin = int(obj.bndbox.ymin.text)
                xmax = int(obj.bndbox.xmax.text)
                ymax = int(obj.bndbox.ymax.text)
                item_dict = {'folder': folder,
                             'image_name': filename,
                             'image_id': image_id,
                             'class_name': class_name,
                             'class_id': class_id,
                             'pose': pose,
                             'truncated': truncated,
                             'difficult': difficult,
                             'xmin': xmin,
                             'ymin': ymin,
                             'xmax': xmax,
                             'ymax': ymax}
                box = []
                for item in self.box_output_format:
                    box.append(item_dict[item])
                boxes.append(box)

            self.labels.append(boxes)

        if ret:
            return self.filenames, self.labels


    def generate(self,
                 batch_size=32,
                 train=True,
                 ssd_box_encoder=None,
                 equalize=False,
                 brightness=False,
                 flip=False,
                 translate=False,
                 scale=False,
                 random_crop=False,
                 crop=False,
                 resize=False,
                 gray=False,
                 limit_boxes=True,
                 include_thresh=0.3,
                 diagnostics=False):
        '''
        Generate batches of samples and corresponding labels indefinitely from
        lists of filenames and labels.

        Returns two Numpy arrays, one containing the next `batch_size` samples
        from `filenames`, the other containing the corresponding labels from
        `labels`.

        Shuffles `filenames` and `labels` consistently after each complete pass.

        Can perform image transformations for data conversion and data augmentation.
        `resize`, `gray`, and `equalize` are image conversion tools and should be
        used consistently during training and inference. The remaining transformations
        serve for data augmentation. Each data augmentation process can set its own
        independent application probability. The transformations are performed
        in the order of their arguments, i.e. equalization is performed first,
        grayscale conversion is performed last.

        `prob` works the same way in all arguments in which it appears. It must be a float in [0,1]
        and determines the probability that the respective transform is applied to a given image.

        All conversions and transforms default to `False`.

        Arguments:
            batch_size (int, optional): The size of the batches to be generated. Defaults to 32.
            train (bool, optional): Whether or not the generator is used in training mode. If `True`, then the labels
                will be transformed into the format that the SSD cost function requires. Otherwise,
                the output format of the labels is identical to the input format. Defaults to `True`.
            ssd_box_encoder (SSDBoxEncoder, optional): Only required if `train = True`. An SSDBoxEncoder object
                to encode the ground truth labels to the required format for training an SSD model.
            equalize (bool, optional): If `True`, performs histogram equalization on the images.
                This can improve contrast and lead the improved model performance.
            brightness (tuple, optional): `False` or a tuple containing three floats, `(min, max, prob)`.
                Scales the brightness of the image by a factor randomly picked from a uniform
                distribution in the boundaries of `[min, max]`. Both min and max must be >=0.
            flip (float, optional): `False` or a float in [0,1], see `prob` above. Flip the image horizontally.
                The respective box coordinates are adjusted accordingly.
            translate (tuple, optional): `False` or a tuple, with the first two elements tuples containing
                two integers each, and the third element a float: `((min, max), (min, max), prob)`.
                The first tuple provides the range in pixels for horizontal shift of the image,
                the second tuple for vertical shift. The number of pixels to shift the image
                by is uniformly distributed within the boundaries of `[min, max]`, i.e. `min` is the number
                of pixels by which the image is translated at least. Both `min` and `max` must be >=0.
                The respective box coordinates are adjusted accordingly.
            scale (tuple, optional): `False` or a tuple containing three floats, `(min, max, prob)`.
                Scales the image by a factor randomly picked from a uniform distribution in the boundaries
                of `[min, max]`. Both min and max must be >=0.
            random_crop (tuple, optional): `False` or a tuple of four integers, `(height, width, min_1_object, max_#_trials)`,
                where `height` and `width` are the height and width of the patch that is to be cropped out at a random
                position in the input image. Note that `height` and `width` can be arbitrary - they are allowed to be larger
                than the image height and width, in which case the original image will be randomly placed on a black background
                canvas of size `(height, width)`. `min_1_object` is either 0 or 1. If 1, there must be at least one detectable
                object remaining in the image for the crop to be valid, and if 0, crops with no detectable objects left in the
                image patch are allowed. `max_#_trials` is only relevant if `min_1_object == 1` and sets the maximum number
                of attempts to get a valid crop. If no valid crop was obtained within this maximum number of attempts,
                the respective image will be removed from the batch without replacement (i.e. for each removed image, the batch
                will be one sample smaller). Defaults to `False`.
            crop (tuple, optional): `False` or a tuple of four integers, `(crop_top, crop_bottom, crop_left, crop_right)`,
                with the number of pixels to crop off of each side of the images.
                The targets are adjusted accordingly. Note: Cropping happens before resizing.
            resize (tuple, optional): `False` or a tuple of 2 integers for the desired output
                size of the images in pixels. The expected format is `(width, height)`.
                The box coordinates are adjusted accordingly. Note: Resizing happens after cropping.
            gray (bool, optional): If `True`, converts the images to grayscale.
            limit_boxes (bool, optional): If `True`, limits box coordinates to stay within image boundaries
                post any transformation. This should always be set to `True`, even if you set `include_thresh`
                to 0. I don't even know why I made this an option. If this is set to `False`, you could
                end up with some boxes that lie entirely outside the image boundaries after a given transformation
                and such boxes would of course not make any sense and have a strongly adverse effect on the learning.
            include_thresh (float, optional): Only relevant if `limit_boxes` is `True`. Determines the minimum
                fraction of the area of a ground truth box that must be left after limiting in order for the box
                to still be included in the batch data. If set to 0, all boxes are kept except those which lie
                entirely outside of the image bounderies after limiting. If set to 1, only boxes that did not
                need to be limited at all are kept. Defaults to 0.3.
            diagnostics (bool, optional): If `True`, yields three additional output items:
                1) A list of the image file names in the batch.
                2) An array with the original, unaltered images.
                3) A list with the original, unaltered labels.
                This can be useful for diagnostic purposes. Defaults to `False`. Only works if `train = True`.

        Yields:
            The next batch as a tuple containing a Numpy array that contains the images and a Python list
            that contains the corresponding labels for each image as 2D Numpy arrays. The output format
            of the labels is according to the `box_output_format` that was specified in the constructor.
        '''

        self.filenames, self.labels = shuffle(self.filenames, self.labels) # Shuffle the data before we begin
        current = 0

        # Find out the indices of the box coordinates in the label data
        xmin = self.box_output_format.index('xmin')
        xmax = self.box_output_format.index('xmax')
        ymin = self.box_output_format.index('ymin')
        ymax = self.box_output_format.index('ymax')

        while True:

            batch_X, batch_y = [], []

            #Shuffle the data after each complete pass
            if current >= len(self.filenames):
                self.filenames, self.labels = shuffle(self.filenames, self.labels)
                current = 0

            for filename in self.filenames[current:current+batch_size]:
                with Image.open('{}'.format(os.path.join(self.images_path, filename))) as img:
                    batch_X.append(np.array(img))
            batch_y = deepcopy(self.labels[current:current+batch_size])

            this_filenames = self.filenames[current:current+batch_size] # The filenames of the files in the current batch

            if diagnostics:
                original_images = np.copy(batch_X) # The original, unaltered images
                original_labels = deepcopy(batch_y) # The original, unaltered labels

            current += batch_size

            # At this point we're done producing the batch. Now perform some
            # optional image transformations:

            batch_items_to_remove = [] # In case we need to remove any images from the batch because of failed random cropping, store their indices in this list

            for i in range(len(batch_X)):

                img_height, img_width, ch = batch_X[i].shape
                batch_y[i] = np.array(batch_y[i]) # Convert labels into an array (in case it isn't one already), otherwise the indexing below breaks

                if equalize:
                    batch_X[i] = histogram_eq(batch_X[i])

                if brightness:
                    p = np.random.uniform(0,1)
                    if p >= (1-brightness[2]):
                        batch_X[i] = _brightness(batch_X[i], min=brightness[0], max=brightness[1])

                # Could easily be extended to also allow vertical flipping, but I'm not convinced of the
                # usefulness of vertical flipping either empirically or theoretically, so I'm going for simplicity.
                # If you want to allow vertical flipping, just change this function to pass the respective argument
                # to `_flip()`.
                if flip:
                    p = np.random.uniform(0,1)
                    if p >= (1-flip):
                        batch_X[i] = _flip(batch_X[i])
                        batch_y[i][:,[xmin,xmax]] = img_width - batch_y[i][:,[xmax,xmin]] # xmin and xmax are swapped when mirrored

                if translate:
                    p = np.random.uniform(0,1)
                    if p >= (1-translate[2]):
                        # Translate the image and return the shift values so that we can adjust the labels
                        batch_X[i], xshift, yshift = _translate(batch_X[i], translate[0], translate[1])
                        # Adjust the labels
                        batch_y[i][:,[xmin,xmax]] += xshift
                        batch_y[i][:,[ymin,ymax]] += yshift
                        # Limit the box coordinates to lie within the image boundaries
                        if limit_boxes:
                            before_limiting = deepcopy(batch_y[i])
                            x_coords = batch_y[i][:,[xmin,xmax]]
                            x_coords[x_coords >= img_width] = img_width - 1
                            x_coords[x_coords < 0] = 0
                            batch_y[i][:,[xmin,xmax]] = x_coords
                            y_coords = batch_y[i][:,[ymin,ymax]]
                            y_coords[y_coords >= img_height] = img_height - 1
                            y_coords[y_coords < 0] = 0
                            batch_y[i][:,[ymin,ymax]] = y_coords
                            # Some objects might have gotten pushed so far outside the image boundaries in the transformation
                            # process that they don't serve as useful training examples anymore, because too little of them is
                            # visible. We'll remove all boxes that we had to limit so much that their area is less than
                            # `include_thresh` of the box area before limiting.
                            before_area = (before_limiting[:,xmax] - before_limiting[:,xmin]) * (before_limiting[:,ymax] - before_limiting[:,ymin])
                            after_area = (batch_y[i][:,xmax] - batch_y[i][:,xmin]) * (batch_y[i][:,ymax] - batch_y[i][:,ymin])
                            if include_thresh == 0: batch_y[i] = batch_y[i][after_area > include_thresh * before_area] # If `include_thresh == 0`, we want to make sure that boxes with area 0 get thrown out, hence the ">" sign instead of the ">=" sign
                            else: batch_y[i] = batch_y[i][after_area >= include_thresh * before_area] # Especially for the case `include_thresh == 1` we want the ">=" sign, otherwise no boxes would be left at all

                if scale:
                    p = np.random.uniform(0,1)
                    if p >= (1-scale[2]):
                        # Rescale the image and return the transformation matrix M so we can use it to adjust the box coordinates
                        batch_X[i], M, scale_factor = _scale(batch_X[i], scale[0], scale[1])
                        # Adjust the box coordinates
                        # Transform two opposite corner points of the rectangular boxes using the transformation matrix `M`
                        toplefts = np.array([batch_y[i][:,xmin], batch_y[i][:,ymin], np.ones(batch_y[i].shape[0])])
                        bottomrights = np.array([batch_y[i][:,xmax], batch_y[i][:,ymax], np.ones(batch_y[i].shape[0])])
                        new_toplefts = (np.dot(M, toplefts)).T
                        new_bottomrights = (np.dot(M, bottomrights)).T
                        batch_y[i][:,[xmin,ymin]] = new_toplefts.astype(np.int)
                        batch_y[i][:,[xmax,ymax]] = new_bottomrights.astype(np.int)
                        # Limit the box coordinates to lie within the image boundaries
                        if limit_boxes and (scale_factor > 1): # We don't need to do any limiting in case we shrunk the image
                            before_limiting = deepcopy(batch_y[i])
                            x_coords = batch_y[i][:,[xmin,xmax]]
                            x_coords[x_coords >= img_width] = img_width - 1
                            x_coords[x_coords < 0] = 0
                            batch_y[i][:,[xmin,xmax]] = x_coords
                            y_coords = batch_y[i][:,[ymin,ymax]]
                            y_coords[y_coords >= img_height] = img_height - 1
                            y_coords[y_coords < 0] = 0
                            batch_y[i][:,[ymin,ymax]] = y_coords
                            # Some objects might have gotten pushed so far outside the image boundaries in the transformation
                            # process that they don't serve as useful training examples anymore, because too little of them is
                            # visible. We'll remove all boxes that we had to limit so much that their area is less than
                            # `include_thresh` of the box area before limiting.
                            before_area = (before_limiting[:,xmax] - before_limiting[:,xmin]) * (before_limiting[:,ymax] - before_limiting[:,ymin])
                            after_area = (batch_y[i][:,xmax] - batch_y[i][:,xmin]) * (batch_y[i][:,ymax] - batch_y[i][:,ymin])
                            if include_thresh == 0: batch_y[i] = batch_y[i][after_area > include_thresh * before_area] # If `include_thresh == 0`, we want to make sure that boxes with area 0 get thrown out, hence the ">" sign instead of the ">=" sign
                            else: batch_y[i] = batch_y[i][after_area >= include_thresh * before_area] # Especially for the case `include_thresh == 1` we want the ">=" sign, otherwise no boxes would be left at all

                if random_crop:
                    # Compute how much room we have in both dimensions to make a random crop.
                    # A negative number here means that we want to crop out a patch that is larger than the original image in the respective dimension,
                    # in which case we will create a black background canvas onto which we will randomly place the image.
                    y_range = img_height - random_crop[0]
                    x_range = img_width - random_crop[1]
                    # Keep track of the number of trials and of whether or not the most recent crop contains at least one object
                    min_1_object_fulfilled = False
                    trial_counter = 0
                    while (not min_1_object_fulfilled) and (trial_counter < random_crop[3]):
                        # Select a random crop position from the possible crop positions
                        if y_range >= 0: crop_ymin = np.random.randint(0, y_range + 1) # There are y_range + 1 possible positions for the crop in the vertical dimension
                        else: crop_ymin = np.random.randint(0, -y_range + 1) # The possible positions for the image on the background canvas in the vertical dimension
                        if x_range >= 0: crop_xmin = np.random.randint(0, x_range + 1) # There are x_range + 1 possible positions for the crop in the horizontal dimension
                        else: crop_xmin = np.random.randint(0, -x_range + 1) # The possible positions for the image on the background canvas in the horizontal dimension
                        # Perform the crop
                        if y_range >= 0 and x_range >= 0: # If the patch to be cropped out is smaller than the original image in both dimenstions, we just perform a regular crop
                            # Crop the image
                            patch_X = np.copy(batch_X[i][crop_ymin:crop_ymin+random_crop[0], crop_xmin:crop_xmin+random_crop[1]])
                            # Translate the box coordinates into the new coordinate system: Cropping shifts the origin by `(crop_ymin, crop_xmin)`
                            patch_y = np.copy(batch_y[i])
                            patch_y[:,[ymin,ymax]] -= crop_ymin
                            patch_y[:,[xmin,xmax]] -= crop_xmin
                            # Limit the box coordinates to lie within the new image boundaries
                            if limit_boxes:
                                # Both the x- and y-coordinates might need to be limited
                                before_limiting = np.copy(patch_y)
                                y_coords = patch_y[:,[ymin,ymax]]
                                y_coords[y_coords < 0] = 0
                                y_coords[y_coords >= random_crop[0]] = random_crop[0] - 1
                                patch_y[:,[ymin,ymax]] = y_coords
                                x_coords = patch_y[:,[xmin,xmax]]
                                x_coords[x_coords < 0] = 0
                                x_coords[x_coords >= random_crop[1]] = random_crop[1] - 1
                                patch_y[:,[xmin,xmax]] = x_coords
                        elif y_range >= 0 and x_range < 0: # If the crop is larger than the original image in the horizontal dimension only,...
                            # Crop the image
                            patch_X = np.copy(batch_X[i][crop_ymin:crop_ymin+random_crop[0]]) # ...crop the vertical dimension just as before,...
                            canvas = np.zeros((random_crop[0], random_crop[1], patch_X.shape[2]), dtype=np.uint8) # ...generate a blank background image to place the patch onto,...
                            canvas[:, crop_xmin:crop_xmin+img_width] = patch_X # ...and place the patch onto the canvas at the random `crop_xmin` position computed above.
                            patch_X = canvas
                            # Translate the box coordinates into the new coordinate system: In this case, the origin is shifted by `(crop_ymin, -crop_xmin)`
                            patch_y = np.copy(batch_y[i])
                            patch_y[:,[ymin,ymax]] -= crop_ymin
                            patch_y[:,[xmin,xmax]] += crop_xmin
                            # Limit the box coordinates to lie within the new image boundaries
                            if limit_boxes:
                                # Only the y-coordinates might need to be limited
                                before_limiting = np.copy(patch_y)
                                y_coords = patch_y[:,[ymin,ymax]]
                                y_coords[y_coords < 0] = 0
                                y_coords[y_coords >= random_crop[0]] = random_crop[0] - 1
                                patch_y[:,[ymin,ymax]] = y_coords
                        elif y_range < 0 and x_range >= 0: # If the crop is larger than the original image in the vertical dimension only,...
                            # Crop the image
                            patch_X = np.copy(batch_X[i][:,crop_xmin:crop_xmin+random_crop[1]]) # ...crop the horizontal dimension just as in the first case,...
                            canvas = np.zeros((random_crop[0], random_crop[1], patch_X.shape[2]), dtype=np.uint8) # ...generate a blank background image to place the patch onto,...
                            canvas[crop_ymin:crop_ymin+img_height, :] = patch_X # ...and place the patch onto the canvas at the random `crop_ymin` position computed above.
                            patch_X = canvas
                            # Translate the box coordinates into the new coordinate system: In this case, the origin is shifted by `(-crop_ymin, crop_xmin)`
                            patch_y = np.copy(batch_y[i])
                            patch_y[:,[ymin,ymax]] += crop_ymin
                            patch_y[:,[xmin,xmax]] -= crop_xmin
                            # Limit the box coordinates to lie within the new image boundaries
                            if limit_boxes:
                                # Only the x-coordinates might need to be limited
                                before_limiting = np.copy(patch_y)
                                x_coords = patch_y[:,[xmin,xmax]]
                                x_coords[x_coords < 0] = 0
                                x_coords[x_coords >= random_crop[1]] = random_crop[1] - 1
                                patch_y[:,[xmin,xmax]] = x_coords
                        else:  # If the crop is larger than the original image in both dimensions,...
                            patch_X = np.copy(batch_X[i])
                            canvas = np.zeros((random_crop[0], random_crop[1], patch_X.shape[2]), dtype=np.uint8) # ...generate a blank background image to place the patch onto,...
                            canvas[crop_ymin:crop_ymin+img_height, crop_xmin:crop_xmin+img_width] = patch_X # ...and place the patch onto the canvas at the random `(crop_ymin, crop_xmin)` position computed above.
                            patch_X = canvas
                            # Translate the box coordinates into the new coordinate system: In this case, the origin is shifted by `(-crop_ymin, -crop_xmin)`
                            patch_y = np.copy(batch_y[i])
                            patch_y[:,[ymin,ymax]] += crop_ymin
                            patch_y[:,[xmin,xmax]] += crop_xmin
                            # Note that no limiting is necessary in this case
                        # Some objects might have gotten pushed so far outside the image boundaries in the transformation
                        # process that they don't serve as useful training examples anymore, because too little of them is
                        # visible. We'll remove all boxes that we had to limit so much that their area is less than
                        # `include_thresh` of the box area before limiting.
                        if limit_boxes and (y_range >= 0 or x_range >= 0):
                            before_area = (before_limiting[:,xmax] - before_limiting[:,xmin]) * (before_limiting[:,ymax] - before_limiting[:,ymin])
                            after_area = (patch_y[:,xmax] - patch_y[:,xmin]) * (patch_y[:,ymax] - patch_y[:,ymin])
                            if include_thresh == 0: patch_y = patch_y[after_area > include_thresh * before_area] # If `include_thresh == 0`, we want to make sure that boxes with area 0 get thrown out, hence the ">" sign instead of the ">=" sign
                            else: patch_y = patch_y[after_area >= include_thresh * before_area] # Especially for the case `include_thresh == 1` we want the ">=" sign, otherwise no boxes would be left at all
                        trial_counter += 1 # We've just used one of our trials
                        # Check if we have found a valid crop
                        if random_crop[2] == 0: # If `min_1_object == 0`, break out of the while loop after the first loop because we are fine with whatever crop we got
                            batch_X[i] = patch_X # The cropped patch becomes our new batch item
                            batch_y[i] = patch_y # The adjusted boxes become our new labels for this batch item
                            # Update the image size so that subsequent transformations can work correctly
                            img_height = random_crop[0]
                            img_width = random_crop[1]
                            break
                        elif len(patch_y) > 0: # If we have at least one object left, this crop is valid and we can stop
                            min_1_object_fulfilled = True
                            batch_X[i] = patch_X # The cropped patch becomes our new batch item
                            batch_y[i] = patch_y # The adjusted boxes become our new labels for this batch item
                            # Update the image size so that subsequent transformations can work correctly
                            img_height = random_crop[0]
                            img_width = random_crop[1]
                        elif (trial_counter >= random_crop[3]) and (not i in batch_items_to_remove): # If we've reached the trial limit and still not found a valid crop, remove this image from the batch
                            batch_items_to_remove.append(i)

                if crop:
                    # Crop the image
                    batch_X[i] = np.copy(batch_X[i][crop[0]:img_height-crop[1], crop[2]:img_width-crop[3]])
                    # Translate the box coordinates into the new coordinate system if necessary: The origin is shifted by `(crop[0], crop[2])` (i.e. by the top and left crop values)
                    # If nothing was cropped off from the top or left of the image, the coordinate system stays the same as before
                    if crop[0] > 0:
                        batch_y[i][:,[ymin,ymax]] -= crop[0]
                    if crop[2] > 0:
                        batch_y[i][:,[xmin,xmax]] -= crop[2]
                    # Update the image size so that subsequent transformations can work correctly
                    img_height -= crop[0] + crop[1]
                    img_width -= crop[2] + crop[3]
                    # Limit the box coordinates to lie within the new image boundaries
                    if limit_boxes:
                        before_limiting = np.copy(batch_y[i])
                        # We only need to check those box coordinates that could possibly have been affected by the cropping
                        # For example, if we only crop off the top and/or bottom of the image, there is no need to check the x-coordinates
                        if crop[0] > 0:
                            y_coords = batch_y[i][:,[ymin,ymax]]
                            y_coords[y_coords < 0] = 0
                            batch_y[i][:,[ymin,ymax]] = y_coords
                        if crop[1] > 0:
                            y_coords = batch_y[i][:,[ymin,ymax]]
                            y_coords[y_coords >= img_height] = img_height - 1
                            batch_y[i][:,[ymin,ymax]] = y_coords
                        if crop[2] > 0:
                            x_coords = batch_y[i][:,[xmin,xmax]]
                            x_coords[x_coords < 0] = 0
                            batch_y[i][:,[xmin,xmax]] = x_coords
                        if crop[3] > 0:
                            x_coords = batch_y[i][:,[xmin,xmax]]
                            x_coords[x_coords >= img_width] = img_width - 1
                            batch_y[i][:,[xmin,xmax]] = x_coords
                        # Some objects might have gotten pushed so far outside the image boundaries in the transformation
                        # process that they don't serve as useful training examples anymore, because too little of them is
                        # visible. We'll remove all boxes that we had to limit so much that their area is less than
                        # `include_thresh` of the box area before limiting.
                        before_area = (before_limiting[:,xmax] - before_limiting[:,xmin]) * (before_limiting[:,ymax] - before_limiting[:,ymin])
                        after_area = (batch_y[i][:,xmax] - batch_y[i][:,xmin]) * (batch_y[i][:,ymax] - batch_y[i][:,ymin])
                        if include_thresh == 0: batch_y[i] = batch_y[i][after_area > include_thresh * before_area] # If `include_thresh == 0`, we want to make sure that boxes with area 0 get thrown out, hence the ">" sign instead of the ">=" sign
                        else: batch_y[i] = batch_y[i][after_area >= include_thresh * before_area] # Especially for the case `include_thresh == 1` we want the ">=" sign, otherwise no boxes would be left at all

                if resize:
                    batch_X[i] = cv2.resize(batch_X[i], dsize=resize)
                    batch_y[i][:,[xmin,xmax]] = (batch_y[i][:,[xmin,xmax]] * (resize[0] / img_width)).astype(np.int)
                    batch_y[i][:,[ymin,ymax]] = (batch_y[i][:,[ymin,ymax]] * (resize[1] / img_height)).astype(np.int)
                    img_width, img_height = resize # Updating these at this point is unnecessary, but it's one fewer source of error if this method gets expanded in the future

                if gray:
                    batch_X[i] = np.expand_dims(cv2.cvtColor(batch_X[i], cv2.COLOR_RGB2GRAY), 3)

            # If any batch items need to be removed because of failed random cropping, remove them now.
            for j in sorted(batch_items_to_remove, reverse=True):
                batch_X.pop(j)
                batch_y.pop(j) # This isn't efficient, but this should hopefully not need to be done often anyway

            if train: # During training we need the encoded labels instead of the format that `batch_y` has
                if ssd_box_encoder is None:
                    raise ValueError("`ssd_box_encoder` cannot be `None` in training mode.")
                y_true = ssd_box_encoder.encode_y(batch_y) # Encode the labels into the `y_true` tensor that the cost function needs

            # CAUTION: Converting `batch_X` into an array will result in an empty batch if the images have varying sizes.
            #          At this point, all images have to have the same size, otherwise you will get an error during training.
            if train:
                if diagnostics:
                    yield (np.array(batch_X), y_true, batch_y, this_filenames, original_images, original_labels)
                else:
                    yield (np.array(batch_X), y_true)
            else:
                yield (np.array(batch_X), batch_y, this_filenames)

    def get_filenames_labels(self):
        '''
        Returns:
            The list of filenames and the list of labels.
        '''
        return self.filenames, self.labels

    def get_n_samples(self):
        '''
        Returns:
            The number of image files in the initialized dataset.
        '''
        return len(self.filenames)

    def process_offline(self,
                        dest_path='',
                        start=0,
                        stop='all',
                        crop=False,
                        equalize=False,
                        brightness=False,
                        flip=False,
                        translate=False,
                        scale=False,
                        resize=False,
                        gray=False,
                        limit_boxes=True,
                        include_thresh=0.3,
                        diagnostics=False):
        '''
        Perform offline image processing.

        This function has mostly the same image processing capabilities as the generator function above,
        but it performs the processing on all items in `filenames` starting at index `start`
        until index `stop` and saves the processed images to disk. The labels are adjusted
        accordingly.

        Processing images offline is useful to reduce the amount of work done by the batch
        generator and thus can speed up training. For example, transformations that are performed
        on all images in a deterministic way, such as resizing or cropping, should be done offline.

        Arguments:
            dest_path (str, optional): The destination directory where the processed images
                and `labels.csv` should be saved, ending on a slash.
            start (int, optional): The inclusive start index from which onward to process the
                items in `filenames`. Defaults to 0.
            stop (int, optional): The exclusive stop index until which to process the
                items in `filenames`. Defaults to 'all', meaning to process all items until the
                end of the list.

        For a description of the other arguments, please refer to the documentation of `generate_batch()` above.

        Returns:
            `None`, but saves all processed images as JPEG files to the specified destination
            directory and generates a `labels.csv` CSV file that is saved to the same directory.
            The format of the lines in the destination CSV file is the same as that of the
            source CSV file, i.e. `[frame, xmin, xmax, ymin, ymax, class_id]`.
        '''

        import gc

        targets_for_csv = []
        if stop == 'all':
            stop = len(self.filenames)

        if diagnostics:
            processed_images = []
            original_images = []
            processed_labels = []

        # Find out the indices of the box coordinates in the label data
        xmin = self.box_output_format.index('xmin')
        xmax = self.box_output_format.index('xmax')
        ymin = self.box_output_format.index('ymin')
        ymax = self.box_output_format.index('ymax')

        for k, filename in enumerate(self.filenames[start:stop]):
            i = k + start
            with Image.open('{}'.format(os.path.join(self.images_path, filename))) as img:
                image = np.array(img)
            targets = np.copy(self.labels[i])

            if diagnostics:
                original_images.append(image)

            img_height, img_width, ch = image.shape

            if equalize:
                image = histogram_eq(image)

            if brightness:
                p = np.random.uniform(0,1)
                if p >= (1-brightness[2]):
                    image = _brightness(image, min=brightness[0], max=brightness[1])

            # Could easily be extended to also allow vertical flipping, but I'm not convinced of the
            # usefulness of vertical flipping either empirically or theoretically, so I'm going for simplicity.
            # If you want to allow vertical flipping, just change this function to pass the respective argument
            # to `_flip()`.
            if flip:
                p = np.random.uniform(0,1)
                if p >= (1-flip):
                    image = _flip(image)
                    targets[:,[0,1]] = img_width - targets[:,[1,0]] # xmin and xmax are swapped when mirrored

            if translate:
                p = np.random.uniform(0,1)
                if p >= (1-translate[2]):
                    image, xshift, yshift = _translate(image, translate[0], translate[1])
                    targets[:,[0,1]] += xshift
                    targets[:,[2,3]] += yshift
                    if limit_boxes:
                        before_limiting = np.copy(targets)
                        x_coords = targets[:,[0,1]]
                        x_coords[x_coords >= img_width] = img_width - 1
                        x_coords[x_coords < 0] = 0
                        targets[:,[0,1]] = x_coords
                        y_coords = targets[:,[2,3]]
                        y_coords[y_coords >= img_height] = img_height - 1
                        y_coords[y_coords < 0] = 0
                        targets[:,[2,3]] = y_coords
                        # Some objects might have gotten pushed so far outside the image boundaries in the transformation
                        # process that they don't serve as useful training examples anymore, because too little of them is
                        # visible. We'll remove all boxes that we had to limit so much that their area is less than
                        # `include_thresh` of the box area before limiting.
                        before_area = (before_limiting[:,1] - before_limiting[:,0]) * (before_limiting[:,3] - before_limiting[:,2])
                        after_area = (targets[:,1] - targets[:,0]) * (targets[:,3] - targets[:,2])
                        targets = targets[after_area >= include_thresh * before_area]

            if scale:
                p = np.random.uniform(0,1)
                if p >= (1-scale[2]):
                    image, M, scale_factor = _scale(image, scale[0], scale[1])
                    # Transform two opposite corner points of the rectangular boxes using the transformation matrix `M`
                    toplefts = np.array([targets[:,0], targets[:,2], np.ones(targets.shape[0])])
                    bottomrights = np.array([targets[:,1], targets[:,3], np.ones(targets.shape[0])])
                    new_toplefts = (np.dot(M, toplefts)).T
                    new_bottomrights = (np.dot(M, bottomrights)).T
                    targets[:,[0,2]] = new_toplefts.astype(np.int)
                    targets[:,[1,3]] = new_bottomrights.astype(np.int)
                    if limit_boxes and (scale_factor > 1): # We don't need to do any limiting in case we shrunk the image
                        before_limiting = np.copy(targets)
                        x_coords = targets[:,[0,1]]
                        x_coords[x_coords >= img_width] = img_width - 1
                        x_coords[x_coords < 0] = 0
                        targets[:,[0,1]] = x_coords
                        y_coords = targets[:,[2,3]]
                        y_coords[y_coords >= img_height] = img_height - 1
                        y_coords[y_coords < 0] = 0
                        targets[:,[2,3]] = y_coords
                        # Some objects might have gotten pushed so far outside the image boundaries in the transformation
                        # process that they don't serve as useful training examples anymore, because too little of them is
                        # visible. We'll remove all boxes that we had to limit so much that their area is less than
                        # `include_thresh` of the box area before limiting.
                        before_area = (before_limiting[:,1] - before_limiting[:,0]) * (before_limiting[:,3] - before_limiting[:,2])
                        after_area = (targets[:,1] - targets[:,0]) * (targets[:,3] - targets[:,2])
                        targets = targets[after_area >= include_thresh * before_area]

            if crop:
                image = image[crop[0]:img_height-crop[1], crop[2]:img_width-crop[3]]
                if limit_boxes: # Adjust boxes affected by cropping and remove those that will no longer be in the image
                    before_limiting = np.copy(targets)
                    if crop[0] > 0:
                        y_coords = targets[:,[2,3]]
                        y_coords[y_coords < crop[0]] = crop[0]
                        targets[:,[2,3]] = y_coords
                    if crop[1] > 0:
                        y_coords = targets[:,[2,3]]
                        y_coords[y_coords >= (img_height - crop[1])] = img_height - crop[1] - 1
                        targets[:,[2,3]] = y_coords
                    if crop[2] > 0:
                        x_coords = targets[:,[0,1]]
                        x_coords[x_coords < crop[2]] = crop[2]
                        targets[:,[0,1]] = x_coords
                    if crop[3] > 0:
                        x_coords = targets[:,[0,1]]
                        x_coords[x_coords >= (img_width - crop[3])] = img_width - crop[3] - 1
                        targets[:,[0,1]] = x_coords
                    # Some objects might have gotten pushed so far outside the image boundaries in the transformation
                    # process that they don't serve as useful training examples anymore, because too little of them is
                    # visible. We'll remove all boxes that we had to limit so much that their area is less than
                    # `include_thresh` of the box area before limiting.
                    before_area = (before_limiting[:,1] - before_limiting[:,0]) * (before_limiting[:,3] - before_limiting[:,2])
                    after_area = (targets[:,1] - targets[:,0]) * (targets[:,3] - targets[:,2])
                    targets = targets[after_area >= include_thresh * before_area]
                # Now adjust the box coordinates for the new image size post cropping
                if crop[0] > 0:
                    targets[:,[2,3]] -= crop[0]
                if crop[2] > 0:
                    targets[:,[0,1]] -= crop[2]
                img_height -= crop[0] - crop[1]
                img_width -= crop[2] - crop[3]

            if resize:
                image = cv2.resize(image, dsize=resize)
                targets[:,[0,1]] = (targets[:,[0,1]] * (resize[0] / img_width)).astype(np.int)
                targets[:,[2,3]] = (targets[:,[2,3]] * (resize[1] / img_height)).astype(np.int)

            if gray:
                image = np.expand_dims(cv2.cvtColor(image, cv2.COLOR_RGB2GRAY), 3)

            if diagnostics:
                processed_images.append(image)
                processed_labels.append(targets)

            img = Image.fromarray(image.astype(np.uint8))
            img.save('{}{}'.format(dest_path, filename), 'JPEG', quality=90)
            del image
            del img
            gc.collect()

            # Transform the labels back to the original CSV file format:
            # One line per ground truth box, i.e. possibly multiple lines per image
            for target in targets:
                target = list(target)
                target = [filename] + target
                targets_for_csv.append(target)

        with open('{}labels.csv'.format(dest_path), 'w', newline='') as csvfile:
            labelswriter = csv.writer(csvfile, delimiter=',', quotechar='|', quoting=csv.QUOTE_MINIMAL)
            labelswriter.writerow(['frame', 'xmin', 'xmax', 'ymin', 'ymax', 'class_id'])
            labelswriter.writerows(targets_for_csv)

        if diagnostics:
            print("Image processing completed.")
            return np.array(processed_images), np.array(original_images), np.array(targets_for_csv), processed_labels
        else:
            print("Image processing completed.")