qi7 dZddlZddlmZmZmZddlmZm Z m Z m Z ddl m Z mZmZmZmZmZmZmZddlmZddlmZmZmZmZmZmZdd lmZerddl Z erddl!Z!ejDe#Z$Gd d ed Z%dZ& ddejNde(dzde(dzde(e zdzfdZ)edGddeZ*dgZ+y)z%Image processor class for LayoutLMv2.N)BaseImageProcessor BatchFeature get_size_dict)flip_channel_orderresizeto_channel_dimension_format to_pil_image)ChannelDimension ImageInputPILImageResamplinginfer_channel_dimension_formatmake_flat_list_of_imagesto_numpy_array valid_imagesvalidate_preprocess_arguments) ImagesKwargs) TensorTypefilter_out_non_signature_kwargsis_pytesseract_availableis_vision_availableloggingrequires_backends)requiresc<eZdZUdZeed<edzed<edzed<y)LayoutLMv2ImageProcessorKwargsa apply_ocr (`bool`, *optional*, defaults to `True`): Whether to apply the Tesseract OCR engine to get words + normalized bounding boxes. Can be overridden by the `apply_ocr` parameter in the `preprocess` method. ocr_lang (`str`, *optional*): The language, specified by its ISO code, to be used by the Tesseract OCR engine. By default, English is used. Can be overridden by the `ocr_lang` parameter in the `preprocess` method. tesseract_config (`str`, *optional*): Any additional custom configuration flags that are forwarded to the `config` parameter when calling Tesseract. For example: '--psm 6'. Can be overridden by the `tesseract_config` parameter in the `preprocess` method. apply_ocrNocr_langtesseract_config)__name__ __module__ __qualname____doc__bool__annotations__strl/opt/pipecat/venv/lib/python3.12/site-packages/transformers/models/layoutlmv2/image_processing_layoutlmv2.pyrr4s" ODjDj r(rF)totalctd|d|z ztd|d|z ztd|d|z ztd|d|z zgS)Nirr)int)boxwidthheights r) normalize_boxr2Gs` DCFUN #$ DCFVO $% DCFUN #$ DCFVO $%  r(imagelangrinput_data_formatc||nd}t||}|j\}}tj||d|}|d|d|d|d|d f\}} } } } t |D cgc]\} }|j r| }} }t |D cgc] \} }| |vs |}} }t | D cgc] \} }| |vs |} } }t | D cgc] \} }| |vs |} } }t | D cgc] \} }| |vs |} } }t | D cgc] \} }| |vs |} } }g}t | | | | D]$\}}}}||||z||zg}|j|&g}|D]}|jt||| t|t|k(sJd ||fScc}} wcc}} wcc}} wcc}} wcc}} wcc}} w) zdApplies Tesseract OCR on a document image, and returns recognized words + normalized bounding boxes.r5dict)r4 output_typeconfigtextlefttopr0r1z-Not as many words as there are bounding boxes) r size pytesseract image_to_data enumeratestripzipappendr2len)r3r4rr5 pil_image image_width image_heightdatawordsr=r>r0r1idxwordirrelevant_indicescoord actual_boxesxywh actual_boxnormalized_boxesr/s r)apply_tesseractrWPs,<+G'RU6GHI )K  $ $YTvVf gD&*6lDL$u+tT[}^bck^l&l#E4eV09/?T)#ttzz|#TT#,U#3 Uic4sBT7TT UE U$-dO Ujc5sBT7TE UD U#,S> SZS%S@R5R5 SC S%.u%5 WzsEDV9VU WE W&/&7 Y U3FX;Xe YF YL$UF3( 1aAE1q5) J'( O c; MNO u:-. ._0__ . " "")U U U S W YsH&F.?F. F4!F46 F:F: G%G: GG G )G )vision)backendsceZdZdZdgZeZddejdddfde de e e fdzded e d e dzd e dzd dffd Z ejddfdejde e e fdede ezdzde ezdzd ejf dZedddddddej&df dede dzde e e fdzdedzd e dzd e dzd e dzde ezdzdede ezdzd ej.j.fdZxZS)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_valuesTNr7 do_resizer?resamplerrrreturnc t|di|||nddd}t|}||_||_||_||_||_||_y)N)r1r0r') super__init__rr]r?r^rrr) selfr]r?r^rrrkwargs __class__s r)rcz!LayoutLMv2ImageProcessor.__init__s[ "6"'tc-JT""   "  0r(r3 data_formatr5c t|}d|vsd|vrtd|j|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. r1r0zFThe `size` dictionary must contain the keys `height` and `width`. Got )r?r^rgr5)r ValueErrorkeysr)rdr3r?r^rgr5re output_sizes r)rzLayoutLMv2ImageProcessor.resizesyFT" 4 7$#6efjfofofqerst tH~tG}5   #/     r(imagesreturn_tensorsc 4||n |j}||n |j}t|}||n |j}||n |j}||n |j }||n |j }t|}t|s tdt||||D cgc] } t| }} | t|d} |rKt|dg} g} |D]6} t| ||| \}}| j|| j|8|r"|D cgc]} |j!| ||| }} |D cgc]} t#| | }} |D cgc]} t%| | | }} t'd|i| }|r  |d < |d <|Scc} wcc} wcc} wcc} w) 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.PYTORCH` or `'pt'`: Return a batch of type `torch.Tensor`. - `TensorType.NUMPY` or `'np'`: Return a batch of type `np.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. zSInvalid image type. Must be of type PIL.Image.Image, numpy.ndarray, or torch.Tensor)r]r?r^rr@r8)r3r?r^r5)input_channel_dimr\)rJ tensor_typerKboxes)r]r?rr^rrrrrrirrrrrWrErrr r)rdrlr]r?r^rrrrmrgr5r3 words_batch boxes_batchrKrqrJs r) preprocessz#LayoutLMv2ImageProcessor.preprocesssZ"+!6IDNN 'tTYYT"'38!*!6IDNN '38/?/K+QUQfQf)&1F#rs s% 6<vay I   dM 2KK *.uh@Pduv u""5)""5) * $ %dXYj kF _eeUZ$U>OPeent ej '{N_ `  .&!9~V 'DM'DM A=  f sFF 8FF)r r!r"r#model_input_namesr valid_kwargsr BILINEARr$r9r&r.rcnpndarrayr rrFIRSTr rPILImagert __classcell__)rfs@r)r[r[ws.((1L&*'9'B'B#')1138nt#1% 1  1 * 1*1 14(:'B'B59;? . zz. 38n. % . ++d2 . !11D8 .  . `%&"&&*.2!%#'+26(8(>(>;?__$;_38nt# _ %t+ _ $; _*_*_j(4/_&_!11D8_ _'_r(r[)NN),r#numpyrximage_processing_utilsrrrimage_transformsrrr r image_utilsr r r rrrrrprocessing_utilsrutilsrrrrrrutils.import_utilsrr{r@ get_loggerr loggerrr2ryr&rWr[__all__r'r(r)rs,UUee   -+   H %!\!&$(7; $# ::$# *$#Dj$#--4 $#N ;A1A AH & &r(