`i95 dZddlmZmZddlZddlmZmZm Z ddl m Z m Z m Z mZddlmZmZmZmZmZmZmZmZddlmZmZmZmZmZmZdd lm Z erddl!Z!erddl"Z"ej#e$Z%d Z& dd ej'd ee(d ee(deee(effdZ)e dGddeZ*dgZ+dS)z%Image processor class for LayoutLMv2.)OptionalUnionN)BaseImageProcessor BatchFeature get_size_dict)flip_channel_orderresizeto_channel_dimension_format to_pil_image)ChannelDimension ImageInputPILImageResamplinginfer_channel_dimension_formatmake_flat_list_of_imagesto_numpy_array valid_imagesvalidate_preprocess_arguments) TensorTypefilter_out_non_signature_kwargsis_pytesseract_availableis_vision_availableloggingrequires_backends)requiresctd|d|z ztd|d|z ztd|d|z ztd|d|z zgS)Nirr)int)boxwidthheights /home/jaya/work/projects/VOICE-AGENT/VIET/agent-env/lib/python3.11/site-packages/transformers/models/layoutlmv2/image_processing_layoutlmv2.py normalize_boxr$6sl DCFUN #$$ DCFVO $%% DCFUN #$$ DCFVO $%%  imagelangtesseract_configinput_data_formatc8||nd}t||}|j\}}tj||d|}|d|d|d|d |d f\}} } } } d t |Dfd t |D}fd t | D} fdt | D} fdt | D} fdt | D} g} t | | | | D](\}}}}||||z||zg}| |)g}| D]&}|t|||'t|t|ks Jd||fS)zdApplies Tesseract OCR on a document image, and returns recognized words + normalized bounding boxes.Nr)dict)r' output_typeconfigtextlefttopr!r"c@g|]\}}||S)strip).0idxwords r# z#apply_tesseract..Os)TTT)#ttzz||T#TTTr%c"g|] \}}|v | Sr4r4)r6r7r8irrelevant_indicess r#r9z#apply_tesseract..Ps( U U Uic4sBT7T7TT7T7T7Tr%c"g|] \}}|v | Sr4r4r6r7coordr;s r#r9z#apply_tesseract..Qs( U U Ujc5sBT7T7TE7T7T7Tr%c"g|] \}}|v | Sr4r4r=s r#r9z#apply_tesseract..Rs( S S SZS%S@R5R5R55R5R5Rr%c"g|] \}}|v | Sr4r4r=s r#r9z#apply_tesseract..Ss( W W WzsEDV9V9VU9V9V9Vr%c"g|] \}}|v | Sr4r4r=s r#r9z#apply_tesseract..Ts( Y Y Y U3FX;X;Xe;X;X;Xr%z-Not as many words as there are bounding boxes) r size pytesseract image_to_data enumeratezipappendr$len)r&r'r(r) pil_image image_width image_heightdatawordsr1r2r!r" actual_boxesxywh actual_boxnormalized_boxesr r;s @r#apply_tesseractrU?s,<+G''RU6GHHHI )K  $YTvVf g g gD&*6lDL$u+tT[}^bck^l&l#E4eVUTy/?/?TTT U U U U9U#3#3 U U UE U U U UIdOO U U UD S S S S9S>> S S SC W W W WYu%5%5 W W WE Y Y Y Yi&7&7 Y Y YFL$UF33(( 1aAE1q5) J''''OO c; M MNNNN u::-.. . . .0_ . . . " ""r%)vision)backendsc6eZdZdZdgZddejdddfdedee e e fded ed ee d ee d dffd Z ejddfde jde e e fdedeee efdeee efd e jf dZedddddddejdf dedeedee e e fdeed eed ee d ee deee efdedeee efd ejjfdZxZS)LayoutLMv2ImageProcessora Constructs a LayoutLMv2 image processor. Args: do_resize (`bool`, *optional*, defaults to `True`): Whether to resize the image's (height, width) dimensions to `(size["height"], size["width"])`. Can be overridden by `do_resize` in `preprocess`. size (`dict[str, int]` *optional*, defaults to `{"height": 224, "width": 224}`): Size of the image after resizing. Can be overridden by `size` in `preprocess`. resample (`PILImageResampling`, *optional*, defaults to `Resampling.BILINEAR`): Resampling filter to use if resizing the image. Can be overridden by the `resample` parameter in the `preprocess` method. apply_ocr (`bool`, *optional*, defaults to `True`): Whether to apply the Tesseract OCR engine to get words + normalized bounding boxes. Can be overridden by `apply_ocr` in `preprocess`. ocr_lang (`str`, *optional*): The language, specified by its ISO code, to be used by the Tesseract OCR engine. By default, English is used. Can be overridden by `ocr_lang` in `preprocess`. tesseract_config (`str`, *optional*, defaults to `""`): Any additional custom configuration flags that are forwarded to the `config` parameter when calling Tesseract. For example: '--psm 6'. Can be overridden by `tesseract_config` in `preprocess`. pixel_valuesTNr+ do_resizerBresample apply_ocrocr_langr(returnc tjdi|||nddd}t|}||_||_||_||_||_||_dS)N)r"r!r4) super__init__rr[rBr\r]r^r() selfr[rBr\r]r^r(kwargs __class__s r#rcz!LayoutLMv2ImageProcessor.__init__sv ""6"""'ttc-J-JT"""   "  0r%r& data_formatr)c t|}d|vsd|vr$td||d|df}t|f||||d|S)a Resize an image to `(size["height"], size["width"])`. Args: image (`np.ndarray`): Image to resize. size (`dict[str, int]`): Dictionary in the format `{"height": int, "width": int}` specifying the size of the output image. resample (`PILImageResampling`, *optional*, defaults to `PILImageResampling.BILINEAR`): `PILImageResampling` filter to use when resizing the image e.g. `PILImageResampling.BILINEAR`. data_format (`ChannelDimension` or `str`, *optional*): The channel dimension format for the output image. If unset, the channel dimension format of the input image is used. 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. 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. Returns: `np.ndarray`: The resized image. r"r!zFThe `size` dictionary must contain the keys `height` and `width`. Got )rBr\rgr))r ValueErrorkeysr )rdr&rBr\rgr)re output_sizes r#r zLayoutLMv2ImageProcessor.resizesFT"" 4  7$#6#6sfjfofofqfqsstt tH~tG}5   #/       r%imagesreturn_tensorsc  ||nj}njtnj||nj}||nj}||nj}t|}t|stdt|d|D} t|d |rYtdg} g} |D]B} t| || \}}| || |C|r fd|D} fd |D} fd |D}td |i| }|r | |d <| |d<|S)a Preprocess an image or batch of images. Args: images (`ImageInput`): Image to preprocess. do_resize (`bool`, *optional*, defaults to `self.do_resize`): Whether to resize the image. size (`dict[str, int]`, *optional*, defaults to `self.size`): Desired size of the output image after resizing. resample (`PILImageResampling`, *optional*, defaults to `self.resample`): Resampling filter to use if resizing the image. This can be one of the enum `PIL.Image` resampling filter. Only has an effect if `do_resize` is set to `True`. apply_ocr (`bool`, *optional*, defaults to `self.apply_ocr`): Whether to apply the Tesseract OCR engine to get words + normalized bounding boxes. ocr_lang (`str`, *optional*, defaults to `self.ocr_lang`): The language, specified by its ISO code, to be used by the Tesseract OCR engine. By default, English is used. tesseract_config (`str`, *optional*, defaults to `self.tesseract_config`): Any additional custom configuration flags that are forwarded to the `config` parameter when calling Tesseract. 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. NzkInvalid image type. Must be of type PIL.Image.Image, numpy.ndarray, torch.Tensor, tf.Tensor or jax.ndarray.)r[rBr\c,g|]}t|Sr4)r)r6r&s r#r9z7LayoutLMv2ImageProcessor.preprocess.. s <<.s> %dXYj kkr%c2g|]}t|S)r,)r )r6r&r)s r#r9z7LayoutLMv2ImageProcessor.preprocess.."s)eeeUZ$U>OPPPeeer%c4g|]}t|S))input_channel_dim)r )r6r&rgr)s r#r9z7LayoutLMv2ImageProcessor.preprocess..#s7   ej '{N_ ` ` `   r%rZ)rL tensor_typerMboxes)r[rBrr\r]r^r(rrrirrrrUrGr)rdrlr[rBr\r]r^r(rmrgr) words_batch boxes_batchr&rMrurLs` `` `` r# preprocessz#LayoutLMv2ImageProcessor.preprocesss8^"+!6IIDN 'ttTYT""'388!*!6IIDN '388/?/K++QUQf)&11F## :  &    =vay I I   * dM 2 2 2KK * *.uh@Pduvvv u""5)))""5))))  #F feee^deee     nt   .&!9~VVV  ('DM'DM r%)__name__ __module__ __qualname____doc__model_input_namesrBILINEARboolrr-strrrcnpndarrayrr r rFIRSTrrPILImagerx __classcell__)rfs@r#rYrYfsi.(()-'9'B"&*,111tCH~&1% 1  1 3- 1#3-1 1111114(:'B>BDH . . z. 38n. % . eC)9$9:; . $E#/?*?$@A .  . . . . `%$&&%))-15$("&*.;?(8(>DHdddD>dtCH~& d -. d D> d3-d#3-d!sJ!78d&d$E#/?*?$@Ad ddd'&dddddr%rY)NN),r|typingrrnumpyrimage_processing_utilsrrrimage_transformsr r r r image_utilsr rrrrrrrutilsrrrrrrutils.import_utilsrrrC get_loggerryloggerr$rrrUrY__all__r4r%r#rsS,+""""""""UUUUUUUUUUeeeeeeeeeeee                    +*****JJJ  H % %'+@D $#$# :$# 3-$#sm$# c+;&; <= $#$#$#$#N ;EEEEE1EE EP & &r%