khV7XdZddlZddlZddlmZddlmZddlmZm Z m Z GddZ y) aExtract images from PDF. Both raster images and vector graphics are considered: * Normal images like jpeg or png could be extracted with method ``page.get_text('rawdict')`` and ``Page.get_images()``. Note the process for png images with alpha channel. * Vector graphics are actually composed of a group of paths, represented by operators like ``re``, ``m``, ``l`` and ``c``. They're detected by finding the contours with ``opencv``. N) Collection) BlockType)recursive_xy_cutinner_contoursxy_project_profileceZdZdejddfdZddej defdZddej defd Z ddefd Z d ed ed edefdZ e dejdej fdZe dejdefdZe dejfdZe dej$defdZe dejfdZy)ImagesExtractorpagereturnNc||_y)zwExtract images from PDF page. Args: page (fitz.Page): pdf page to extract images. N)_page)selfr s Z/var/www/teggl/fontify/venv/lib/python3.12/site-packages/pdf2docx/image/ImagesExtractor.py__init__zImagesExtractor.__init__s  bboxzoomcf|j|j||jj}n2|jjr||jjz}n|}||jjz}t j ||}|jj||S)aClip page pixmap (without text) according to ``bbox``. Args: bbox (fitz.Rect, optional): Target area to clip. Defaults to None, i.e. entire page. Note that ``bbox`` depends on un-rotated page CS, while clipping page is based on the final page. zoom (float, optional): Improve resolution by this rate. Defaults to 3.0. Returns: fitz.Pixmap: The extracted pixmap. )clipmatrix)_hide_page_textrrectrotationrotation_matrixfitzMatrix get_pixmap)rrr clip_bboxrs rclip_page_to_pixmapz#ImagesExtractor.clip_page_to_pixmaps TZZ( < IZZ tzz999II / T4(zz$$)F$CCrclip_image_res_ratiocL|j||}|j||S)arClip page pixmap (without text) according to ``bbox`` and convert to source image. Args: bbox (fitz.Rect, optional): Target area to clip. Defaults to None, i.e. entire page. clip_image_res_ratio (float, optional): Resolution ratio of clipped bitmap. Defaults to 3.0. Returns: list: A list of image raw dict. )rr)r _to_raw_dict)rrr!pixs rclip_page_to_dictz!ImagesExtractor.clip_page_to_dict?s-&&D7K&L  d++rcX|jj}|jj}t}|jj dD]}t |}d|d<|jj |}|jj}|D];}|jdkr|j|s)|j||f=d} |j| } g} | D]} t| dkDr6tj} | D] \}}| |z} |j| |}n_| d\}}|dd k(r|j||}n<|j!||}|j#||}|r|j%|| |d <| j|| S) aExtract normal images with ``Page.get_images()``. Args: clip_image_res_ratio (float, optional): Resolution ratio of clipped bitmap. Defaults to 3.0. Returns: list: A list of extracted and recovered image raw dict. .. note:: ``Page.get_images()`` contains each image only once, which may less than the real count of images in a page. T)fullrc0|dj|dS)Nr) intersects)abs rz0ImagesExtractor.extract_images..vs1Q4??1Q40rimage)rparentrr get_imageslistget_image_rectscropboxget_arear+appendgrouplenrRectr%_recover_pixmapr# _rotate_image)rr!docricitemrectsunrotated_page_bboxrfungroupsimagesr:rraw_dictr$s rextract_imageszImagesExtractor.extract_imagesMsjj::&&\JJ))t)4 (D:DDHJJ..t4E"&**"4"4  (==?A%x+55d;X 4,' ( ($1#( $E5zA~ IIK $),>sXI,N) MM( #Q( $T rmin_svg_gap_dxmin_svg_gap_dymin_wmin_hc ddl}|jd}|j|}|j||j}|j |dd|j \} } t| ||} | D cgc]} t| | ||} } tt| | }d}|rt| D]@\}\}}}}t|||||ddf| ||||f}|jd ||B|D]J\} }| \}}}}|j|||f||fd d |D] \}}}}|j|||f||fd d "L|jd ||jd|Scc} w)aFind contour of potential vector graphics. Args: min_svg_gap_dx (float): Merge svg if the horizontal gap is less than this value. min_svg_gap_dy (float): Merge svg if the vertical gap is less than this value. min_w (float): Ignore contours if the bbox width is less than this value. min_h (float): Ignore contours if the bbox height is less than this value. Returns: list: A list of potential svg region: (external_bbox, inner_bboxes:list). rN?)r)min_dxmin_dyFz sub-image-)rPrrr/)rrrPimg)cv2r _pixmap_to_cv_imagecvtColorCOLOR_BGR2GRAY thresholdTHRESH_BINARY_INVrrr5zip enumeraterimshow rectanglewaitKey)rrIrJrKrLcvpixmapsrcgray_binaryexternal_bboxesrgrouped_inner_bboxesrEdebugix0y0x1y1arr inner_bboxesu0v0u1v1s rdetect_svg_contoursz#ImagesExtractor.detect_svg_contourss ))s)3&&v.{{3 1 12LLsC1E1EF 6+6.Q_`XggtvtUE Jggc/+?@A '0'A 1##BB(RUBrE1_)=vbeRPRUl?ST Jqc*C0 1'- H"l!%BB S2r(RHiC'3HNBBLLr2hR)QGH  H IIeS ! JJqM 5 hs7E)r2ctjjt||j|j |j dS)zStore Pixmap ``image`` to raw dict. Args: image (fitz.Pixmap): Pixmap to store. bbox (fitz.Rect): Boundary box the pixmap. Returns: dict: Raw dict of the pixmap. )typerwidthheightr2)rIMAGEvaluetuplervrwtobytes)r2rs rr#zImagesExtractor._to_raw_dicts9OO))$K[[ll]]_   rr`rcddl}ddl}tj|}|jdd\}}|dz|dz}}d} |j ||f|| } |j | d} |j | d} t|| z|| zz} t|| z|| zz}| dxx| dz |z z cc<| dxx|dz |z z cc<|j|| | |f}|jd |\}}|jS) zRotate image represented by image bytes. Args: pixmap (fitz.Pixmap): Image to rotate. rotation (int): Rotation angle. Return: image bytes. rNrrN)rr)rr/)rr)r/rz.png) rTnumpyr rUshapegetRotationMatrix2Dabsint warpAffineimencoder{)r`rr_nprShwrirjscalercossinWH rotated_imgrcim_pngs rr>zImagesExtractor._rotate_images" 11&9yy!}1Aq!tB''R(EBffVD\"ffVD\" SQW% & SQW% & t Q" $ t Q" $ mmC!Q8 KK 4 6~~rc n|jDcgc] \}}}}| }}}}}|j|j|j}|D]U}|j |j ddj ddj dd}|j ||Wycc}}}}w)z$Hide page text before clipping page.sBTsBT 3 TrsTmsTm 3 TrsTdsTd 3 TrN) get_xobjectsextend get_contentsr3 xref_streamreplace update_stream)r xrefnameinvokerr xref_listr?streams rrzImagesExtractor._hide_page_text!s >B=N=N=PQQ9tT7DTQ Q**,-kk ,D__T*225*E.5geZ.H.5geZ.H    dF +  ,RsB/ r?rAc|d}|d}tj||}|dkDrtj||}|jrtj|d}d}|}|j|jk(r0|j|jk(rtj||}nt j d|d|djvr$tjtj|}|S)a&Restore pixmap with soft mask considered. References: * https://pymupdf.readthedocs.io/en/latest/document.html#Document.getPageImageList * https://pymupdf.readthedocs.io/en/latest/faq.html#how-to-handle-stencil-masks * https://github.com/pymupdf/PyMuPDF/issues/670 Args: doc (fitz.Document): pdf document. item (list): image instance of ``page.get_images()``. Returns: fitz.Pixmap: Recovered pixmap with soft mask considered. rr/NzCIgnore image due to inconsistent size of color and mask pixmaps: %sCMYKr0) rPixmapalpharvrwloggingwarninguppercsRGB)r?rAxsr$masktemps rr=zImagesExtractor._recover_pixmap6s& G Gkk#q! q5;;sA&Dyy{{3*yy$**$T[[)@kk#t, egkl T!W]]_ $++djj#.C rcddl}ddl}|j}|j|j ||j |j S)znConvert fitz Pixmap to opencv image. Args: pixmap (fitz.Pixmap): PyMuPDF Pixmap. rN)rTr}r{imdecode frombufferuint8 IMREAD_COLOR)r`r_rimg_bytes rrUz#ImagesExtractor._pixmap_to_cv_imagefs: >>#{{2==288rDocumentr5r=rUrrr r s7DII$ Dtyy DE DF ,TYY ,% ,X%Xv44u4TY4af4n 4;; TYY  &* T[[* 3* * Z,TYY,,(,DMM,,,^ O4;; O Orr ) __doc__rrcommon.Collectionr common.sharercommon.algorithmrrrr rrrrs, *$UU^O^Or