# coding=utf-8 # Copyright 2022 The HuggingFace Inc. team. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. """Image processor class for Donut.""" from typing import Dict, List, Optional, Union import numpy as np from ...image_processing_utils import BaseImageProcessor, BatchFeature, get_size_dict from ...image_transforms import ( get_resize_output_image_size, pad, resize, to_channel_dimension_format, ) from ...image_utils import ( IMAGENET_STANDARD_MEAN, IMAGENET_STANDARD_STD, ChannelDimension, ImageInput, PILImageResampling, get_image_size, infer_channel_dimension_format, is_scaled_image, make_list_of_images, to_numpy_array, valid_images, validate_preprocess_arguments, ) from ...utils import TensorType, filter_out_non_signature_kwargs, logging from ...utils.import_utils import is_vision_available logger = logging.get_logger(__name__) if is_vision_available(): import PIL class DonutImageProcessor(BaseImageProcessor): r""" Constructs a Donut image processor. Args: do_resize (`bool`, *optional*, defaults to `True`): Whether to resize the image's (height, width) dimensions to the specified `size`. Can be overridden by `do_resize` in the `preprocess` method. size (`Dict[str, int]` *optional*, defaults to `{"shortest_edge": 224}`): Size of the image after resizing. The shortest edge of the image is resized to size["shortest_edge"], with the longest edge resized to keep the input aspect ratio. Can be overridden by `size` in the `preprocess` method. resample (`PILImageResampling`, *optional*, defaults to `Resampling.BILINEAR`): Resampling filter to use if resizing the image. Can be overridden by `resample` in the `preprocess` method. do_thumbnail (`bool`, *optional*, defaults to `True`): Whether to resize the image using thumbnail method. do_align_long_axis (`bool`, *optional*, defaults to `False`): Whether to align the long axis of the image with the long axis of `size` by rotating by 90 degrees. do_pad (`bool`, *optional*, defaults to `True`): Whether to pad the image. If `random_padding` is set to `True` in `preprocess`, each image is padded with a random amont of padding on each size, up to the largest image size in the batch. Otherwise, all images are padded to the largest image size in the batch. do_rescale (`bool`, *optional*, defaults to `True`): Whether to rescale the image by the specified scale `rescale_factor`. Can be overridden by `do_rescale` in the `preprocess` method. rescale_factor (`int` or `float`, *optional*, defaults to `1/255`): Scale factor to use if rescaling the image. Can be overridden by `rescale_factor` in the `preprocess` method. do_normalize (`bool`, *optional*, defaults to `True`): Whether to normalize the image. Can be overridden by `do_normalize` in the `preprocess` method. image_mean (`float` or `List[float]`, *optional*, defaults to `IMAGENET_STANDARD_MEAN`): Mean to use if normalizing the image. This is a float or list of floats the length of the number of channels in the image. Can be overridden by the `image_mean` parameter in the `preprocess` method. image_std (`float` or `List[float]`, *optional*, defaults to `IMAGENET_STANDARD_STD`): Image standard deviation. """ model_input_names = ["pixel_values"] def __init__( self, do_resize: bool = True, size: Dict[str, int] = None, resample: PILImageResampling = PILImageResampling.BILINEAR, do_thumbnail: bool = True, do_align_long_axis: bool = False, do_pad: bool = True, do_rescale: bool = True, rescale_factor: Union[int, float] = 1 / 255, do_normalize: bool = True, image_mean: Optional[Union[float, List[float]]] = None, image_std: Optional[Union[float, List[float]]] = None, **kwargs, ) -> None: super().__init__(**kwargs) size = size if size is not None else {"height": 2560, "width": 1920} if isinstance(size, (tuple, list)): # The previous feature extractor size parameter was in (width, height) format size = size[::-1] size = get_size_dict(size) self.do_resize = do_resize self.size = size self.resample = resample self.do_thumbnail = do_thumbnail self.do_align_long_axis = do_align_long_axis self.do_pad = do_pad self.do_rescale = do_rescale self.rescale_factor = rescale_factor self.do_normalize = do_normalize self.image_mean = image_mean if image_mean is not None else IMAGENET_STANDARD_MEAN self.image_std = image_std if image_std is not None else IMAGENET_STANDARD_STD def align_long_axis( self, image: np.ndarray, size: Dict[str, int], data_format: Optional[Union[str, ChannelDimension]] = None, input_data_format: Optional[Union[str, ChannelDimension]] = None, ) -> np.ndarray: """ Align the long axis of the image to the longest axis of the specified size. Args: image (`np.ndarray`): The image to be aligned. size (`Dict[str, int]`): The size `{"height": h, "width": w}` to align the long axis to. data_format (`str` or `ChannelDimension`, *optional*): The data format of the output image. If unset, the same format as the input image is used. input_data_format (`ChannelDimension` or `str`, *optional*): The channel dimension format of the input image. If not provided, it will be inferred. Returns: `np.ndarray`: The aligned image. """ input_height, input_width = get_image_size(image, channel_dim=input_data_format) output_height, output_width = size["height"], size["width"] if (output_width < output_height and input_width > input_height) or ( output_width > output_height and input_width < input_height ): image = np.rot90(image, 3) if data_format is not None: image = to_channel_dimension_format(image, data_format, input_channel_dim=input_data_format) return image def pad_image( self, image: np.ndarray, size: Dict[str, int], random_padding: bool = False, data_format: Optional[Union[str, ChannelDimension]] = None, input_data_format: Optional[Union[str, ChannelDimension]] = None, ) -> np.ndarray: """ Pad the image to the specified size. Args: image (`np.ndarray`): The image to be padded. size (`Dict[str, int]`): The size `{"height": h, "width": w}` to pad the image to. random_padding (`bool`, *optional*, defaults to `False`): Whether to use random padding or not. data_format (`str` or `ChannelDimension`, *optional*): The data format of the output image. If unset, the same format as the input image is used. input_data_format (`ChannelDimension` or `str`, *optional*): The channel dimension format of the input image. If not provided, it will be inferred. """ output_height, output_width = size["height"], size["width"] input_height, input_width = get_image_size(image, channel_dim=input_data_format) delta_width = output_width - input_width delta_height = output_height - input_height if random_padding: pad_top = np.random.randint(low=0, high=delta_height + 1) pad_left = np.random.randint(low=0, high=delta_width + 1) else: pad_top = delta_height // 2 pad_left = delta_width // 2 pad_bottom = delta_height - pad_top pad_right = delta_width - pad_left padding = ((pad_top, pad_bottom), (pad_left, pad_right)) return pad(image, padding, data_format=data_format, input_data_format=input_data_format) def pad(self, *args, **kwargs): logger.info("pad is deprecated and will be removed in version 4.27. Please use pad_image instead.") return self.pad_image(*args, **kwargs) def thumbnail( self, image: np.ndarray, size: Dict[str, int], resample: PILImageResampling = PILImageResampling.BICUBIC, data_format: Optional[Union[str, ChannelDimension]] = None, input_data_format: Optional[Union[str, ChannelDimension]] = None, **kwargs, ) -> np.ndarray: """ Resize the image to make a thumbnail. The image is resized so that no dimension is larger than any corresponding dimension of the specified size. Args: image (`np.ndarray`): The image to be resized. size (`Dict[str, int]`): The size `{"height": h, "width": w}` to resize the image to. resample (`PILImageResampling`, *optional*, defaults to `PILImageResampling.BICUBIC`): The resampling filter to use. data_format (`Optional[Union[str, ChannelDimension]]`, *optional*): The data format of the output image. If unset, the same format as the input image is used. input_data_format (`ChannelDimension` or `str`, *optional*): The channel dimension format of the input image. If not provided, it will be inferred. """ input_height, input_width = get_image_size(image, channel_dim=input_data_format) output_height, output_width = size["height"], size["width"] # We always resize to the smallest of either the input or output size. height = min(input_height, output_height) width = min(input_width, output_width) if height == input_height and width == input_width: return image if input_height > input_width: width = int(input_width * height / input_height) elif input_width > input_height: height = int(input_height * width / input_width) return resize( image, size=(height, width), resample=resample, reducing_gap=2.0, data_format=data_format, input_data_format=input_data_format, **kwargs, ) def resize( self, image: np.ndarray, size: Dict[str, int], resample: PILImageResampling = PILImageResampling.BICUBIC, data_format: Optional[Union[str, ChannelDimension]] = None, input_data_format: Optional[Union[str, ChannelDimension]] = None, **kwargs, ) -> np.ndarray: """ Resizes `image` to `(height, width)` specified by `size` using the PIL library. Args: image (`np.ndarray`): Image to resize. size (`Dict[str, int]`): Size of the output image. resample (`PILImageResampling`, *optional*, defaults to `PILImageResampling.BICUBIC`): Resampling filter to use when resiizing the image. data_format (`str` or `ChannelDimension`, *optional*): The channel dimension format of the image. If not provided, it will be the same as the input image. input_data_format (`ChannelDimension` or `str`, *optional*): The channel dimension format of the input image. If not provided, it will be inferred. """ size = get_size_dict(size) shortest_edge = min(size["height"], size["width"]) output_size = get_resize_output_image_size( image, size=shortest_edge, default_to_square=False, input_data_format=input_data_format ) resized_image = resize( image, size=output_size, resample=resample, data_format=data_format, input_data_format=input_data_format, **kwargs, ) return resized_image @filter_out_non_signature_kwargs() def preprocess( self, images: ImageInput, do_resize: Optional[bool] = None, size: Dict[str, int] = None, resample: PILImageResampling = None, do_thumbnail: Optional[bool] = None, do_align_long_axis: Optional[bool] = None, do_pad: Optional[bool] = None, random_padding: bool = False, do_rescale: Optional[bool] = None, rescale_factor: Optional[float] = None, do_normalize: Optional[bool] = None, image_mean: Optional[Union[float, List[float]]] = None, image_std: Optional[Union[float, List[float]]] = None, return_tensors: Optional[Union[str, TensorType]] = None, data_format: Optional[ChannelDimension] = ChannelDimension.FIRST, input_data_format: Optional[Union[str, ChannelDimension]] = None, ) -> PIL.Image.Image: """ Preprocess an image or batch of images. Args: images (`ImageInput`): Image to preprocess. Expects a single or batch of images with pixel values ranging from 0 to 255. If passing in images with pixel values between 0 and 1, set `do_rescale=False`. do_resize (`bool`, *optional*, defaults to `self.do_resize`): Whether to resize the image. size (`Dict[str, int]`, *optional*, defaults to `self.size`): Size of the image after resizing. Shortest edge of the image is resized to min(size["height"], size["width"]) with the longest edge resized to keep the input aspect ratio. resample (`int`, *optional*, defaults to `self.resample`): Resampling filter to use if resizing the image. This can be one of the enum `PILImageResampling`. Only has an effect if `do_resize` is set to `True`. do_thumbnail (`bool`, *optional*, defaults to `self.do_thumbnail`): Whether to resize the image using thumbnail method. do_align_long_axis (`bool`, *optional*, defaults to `self.do_align_long_axis`): Whether to align the long axis of the image with the long axis of `size` by rotating by 90 degrees. do_pad (`bool`, *optional*, defaults to `self.do_pad`): Whether to pad the image. If `random_padding` is set to `True`, each image is padded with a random amont of padding on each size, up to the largest image size in the batch. Otherwise, all images are padded to the largest image size in the batch. random_padding (`bool`, *optional*, defaults to `self.random_padding`): Whether to use random padding when padding the image. If `True`, each image in the batch with be padded with a random amount of padding on each side up to the size of the largest image in the batch. do_rescale (`bool`, *optional*, defaults to `self.do_rescale`): Whether to rescale the image pixel values. rescale_factor (`float`, *optional*, defaults to `self.rescale_factor`): Rescale factor to rescale the image by if `do_rescale` is set to `True`. do_normalize (`bool`, *optional*, defaults to `self.do_normalize`): Whether to normalize the image. image_mean (`float` or `List[float]`, *optional*, defaults to `self.image_mean`): Image mean to use for normalization. image_std (`float` or `List[float]`, *optional*, defaults to `self.image_std`): Image standard deviation to use for normalization. return_tensors (`str` or `TensorType`, *optional*): The type of tensors to return. Can be one of: - Unset: Return a list of `np.ndarray`. - `TensorType.TENSORFLOW` or `'tf'`: Return a batch of type `tf.Tensor`. - `TensorType.PYTORCH` or `'pt'`: Return a batch of type `torch.Tensor`. - `TensorType.NUMPY` or `'np'`: Return a batch of type `np.ndarray`. - `TensorType.JAX` or `'jax'`: Return a batch of type `jax.numpy.ndarray`. data_format (`ChannelDimension` or `str`, *optional*, defaults to `ChannelDimension.FIRST`): The channel dimension format for the output image. Can be one of: - `ChannelDimension.FIRST`: image in (num_channels, height, width) format. - `ChannelDimension.LAST`: image in (height, width, num_channels) format. - Unset: defaults to the channel dimension format of the input image. input_data_format (`ChannelDimension` or `str`, *optional*): The channel dimension format for the input image. If unset, the channel dimension format is inferred from the input image. Can be one of: - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format. - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format. - `"none"` or `ChannelDimension.NONE`: image in (height, width) format. """ do_resize = do_resize if do_resize is not None else self.do_resize size = size if size is not None else self.size if isinstance(size, (tuple, list)): # Previous feature extractor had size in (width, height) format size = size[::-1] size = get_size_dict(size) resample = resample if resample is not None else self.resample do_thumbnail = do_thumbnail if do_thumbnail is not None else self.do_thumbnail do_align_long_axis = do_align_long_axis if do_align_long_axis is not None else self.do_align_long_axis do_pad = do_pad if do_pad is not None else self.do_pad do_rescale = do_rescale if do_rescale is not None else self.do_rescale rescale_factor = rescale_factor if rescale_factor is not None else self.rescale_factor do_normalize = do_normalize if do_normalize is not None else self.do_normalize image_mean = image_mean if image_mean is not None else self.image_mean image_std = image_std if image_std is not None else self.image_std images = make_list_of_images(images) if not valid_images(images): raise ValueError( "Invalid image type. Must be of type PIL.Image.Image, numpy.ndarray, " "torch.Tensor, tf.Tensor or jax.ndarray." ) validate_preprocess_arguments( do_rescale=do_rescale, rescale_factor=rescale_factor, do_normalize=do_normalize, image_mean=image_mean, image_std=image_std, do_pad=do_pad, size_divisibility=size, # There is no pad divisibility in this processor, but pad requires the size arg. do_resize=do_resize, size=size, resample=resample, ) # All transformations expect numpy arrays. images = [to_numpy_array(image) for image in images] if do_rescale and is_scaled_image(images[0]): logger.warning_once( "It looks like you are trying to rescale already rescaled images. If the input" " images have pixel values between 0 and 1, set `do_rescale=False` to avoid rescaling them again." ) if input_data_format is None: # We assume that all images have the same channel dimension format. input_data_format = infer_channel_dimension_format(images[0]) if do_align_long_axis: images = [self.align_long_axis(image, size=size, input_data_format=input_data_format) for image in images] if do_resize: images = [ self.resize(image=image, size=size, resample=resample, input_data_format=input_data_format) for image in images ] if do_thumbnail: images = [self.thumbnail(image=image, size=size, input_data_format=input_data_format) for image in images] if do_pad: images = [ self.pad_image( image=image, size=size, random_padding=random_padding, input_data_format=input_data_format ) for image in images ] if do_rescale: images = [ self.rescale(image=image, scale=rescale_factor, input_data_format=input_data_format) for image in images ] if do_normalize: images = [ self.normalize(image=image, mean=image_mean, std=image_std, input_data_format=input_data_format) for image in images ] images = [ to_channel_dimension_format(image, data_format, input_channel_dim=input_data_format) for image in images ] data = {"pixel_values": images} return BatchFeature(data=data, tensor_type=return_tensors) __all__ = ["DonutImageProcessor"]