o 0jn@srgdZddlZddlmZddlmZmZmZmZm Z ddl Z ddl m Z ddlmZddlmZd d lmZmZd e jd e jd efddZ  dNdedededed ef ddZddZ  dOdeedeeded efdd Z dPde e jeefde e jeefd efd!d"Zd#d$Z d%d&Z!d'd(Z"d)d*Z#d+d,Z$d-d.Z%d/d0Z& dQde ee'ee'e'e'e'ffde ee'ee'e'e'e'ffd1ed2ed ee ee'ee'e'e'e'fff d3d4Z( dRd6eeeee'fd7ed2ed eeeeee'feeeee'fffd8d9Z)dSd;d<Z*d efd=d>Z+d?d@Z,dAedBe-fdCdDZ.dEdFZ/dGdHZ0dIdJZ1dTdKed efdLdMZ2dS)U)get_sub_regions_ocr_resget_show_colorsorted_layout_boxesN)deepcopy)DictListOptionalTupleUnion)Image)convert_points_to_boxes) OCRResult)BLOCK_LABEL_MAPREGION_SETTINGS src_boxes ref_boxesreturnc Csg}t|}|dkrot|dkrott|D]X}||}t|d|dddf}t|d|dddf}t|d|dddf}t|d|dddf} ||} | |} t| dk| dk@d} || q|S)a Get the indices of source boxes that overlap with reference boxes based on a specified threshold. Args: src_boxes (np.ndarray): A 2D numpy array of source bounding boxes. ref_boxes (np.ndarray): A 2D numpy array of reference bounding boxes. Returns: match_idx_list (list): A list of indices of source boxes that overlap with reference boxes. rNrr )lenrangenpmaximumminimumwhereextend) rrmatch_idx_listZ src_boxes_numZrnoZref_boxx1y1x2y2Zpub_wZpub_hZ match_idxr"q/var/www/html/Deteccion_Ine/venv/lib/python3.10/site-packages/paddlex/inference/pipelines/layout_parsing/utils.pyget_overlap_boxes_idx!s  r$TFoverall_ocr_res object_boxes flag_withinreturn_match_idxc Csi}g|d<g|d<g|d<g|d<|d}t||}tt|}tt|D]E}|r5||vr2d}n d}n ||vrsz%sorted_layout_boxes..)r1rTr2r rcS |ddSNr2rr"r3r"r"r#r5 cSr8r9r"r3r"r"r#r5r:)rsortedr-r/) reswZ num_boxesZ sorted_boxesZ_boxesZnew_resZres_leftZ res_rightir"r"r#rwsB   r horizontalunionbbox1bbox2 directionc Csd\}}|dkr d\}}t||||}t||||}||}|dkr(dS|dkr?t||||t||||} n4|dkrUt||||||||} n|dkrkt||||||||} ntd|d | dkr{|| Sd S) a Calculate the IoU of lines between two bounding boxes. Args: bbox1 (List[float]): First bounding box [x_min, y_min, x_max, y_max]. bbox2 (List[float]): Second bounding box [x_min, y_min, x_max, y_max]. direction (str): direction of the projection, "horizontal" or "vertical". Returns: float: Line overlap ratio. Returns 0 if there is no overlap. )rrr?)rr rr@smalllarge Invalid mode -, must be one of ['union', 'small', 'large'].)maxmin ValueError) rArBrCmode start_indexZ end_indexZintersection_startZintersection_endZoverlapZ ref_widthr"r"r#"calculate_projection_overlap_ratios0  rNc Cstj|tjd}tj|tjd}t|d|d}t|d|d}t|d|d}t|d|d}td||}td||}tj||tjd} t|} t|} |dkrd| | | } n|dkrot| | } n|dkrzt| | } ntd |d | dkrd S| | S) a Calculate the overlap ratio between two bounding boxes using NumPy. Args: bbox1 (np.ndarray, list or tuple): The first bounding box, format [x_min, y_min, x_max, y_max] bbox2 (np.ndarray, list or tuple): The second bounding box, format [x_min, y_min, x_max, y_max] mode (str): The mode of calculation, either 'union', 'small', or 'large'. Returns: float: The overlap ratio value between the two bounding boxes Zdtyperrr rr@rDrErFrGrH)rr0Zfloat64rrmultiplycalculate_bbox_arearK) rArBrL x_min_inter y_min_inter x_max_inter y_max_interZ inter_widthZ inter_heightZ inter_areaZ bbox1_areaZ bbox2_areaZref_arear"r"r#calculate_overlap_ratios. rVcCs|stdt|}t|dddf}t|dddf}t|dddf}t|dddf}t||||gS)a? Calculate the minimum enclosing bounding box for a list of bounding boxes. Args: bboxes (list): A list of bounding boxes represented as lists of four integers [x1, y1, x2, y2]. Returns: list: The minimum enclosing bounding box represented as a list of four integers [x1, y1, x2, y2]. z$The list of bounding boxes is empty.Nrrr r)rKrr0rJrI)ZbboxesZ bboxes_arrayZmin_xZmin_yZmax_xZmax_yr"r"r# calculate_minimum_enclosing_bboxs  rWcCttd|S)z#check if the char is english letterz ^[A-Za-z]$boolrematchcharr"r"r#is_english_letter(r_cCrX)zcheck if the char is numericz^[\d]+$rYr]r"r"r# is_numeric-r`racCshd}||vS)z check if the char is non-breaking punctuation Args: char (str): character to check Returns: bool: True if the char is non-breaking punctuation > ,u“;:u：u、"'u，-u；r")r^Znon_breaking_punctuationsr"r"r#is_non_breaking_punctuation2s rhc Cs:ttt|\}}}}d|d|d|d|d|d S)Nz imgs/img_in_Z_box__z.jpg)r-mapint)labelboxx_miny_minx_maxy_maxr"r"r#construct_img_pathMs$rrcCsr|||||f}|jdkrdS|jdkrt|S|jddkr4t|dddfddddfSt|S)z Crop `original_img` to [y_min:y_max, x_min:x_max], clamped to image bounds. Returns a C-contiguous array suitable for `PIL.Image.fromarray`. rNr r.)sizendimrZascontiguousarrayshape) original_imgrnrorprqcropr"r"r#_crop_image_region_for_pilRs   $ ryc Csg}t|jdt|jd}}|D]n}|dtdvr|d}ttt|d\}}} } tdt||}tdt| |} tdt||}tdt| |} | |ksV| |krWqt||d} t|||| | } | durkqt | } | | | |||| | f|ddq|S)NrrrlZ image_labels coordinatescore)pathimgrlrzr{) rkrvrr-rjrIrJrrryr Z fromarrayr/)rwZlayout_det_objsZ imgs_in_dochr=Zdet_objrlrnrorprqZimg_pathZ arr_for_pilr}r"r"r# gather_imgsas:    rratiosmallercCsJt|}t|}t||dd}||kr#||kr|s||kr!|s!dSdSdS)a Determine if the overlap area between two bounding boxes exceeds a given ratio and return the smaller (or larger) bounding box based on the `smaller` flag. Args: bbox1 (Union[List[int], Tuple[int, int, int, int]]): Coordinates of the first bounding box [x_min, y_min, x_max, y_max]. bbox2 (Union[List[int], Tuple[int, int, int, int]]): Coordinates of the second bounding box [x_min, y_min, x_max, y_max]. ratio (float): The overlap ratio threshold. smaller (bool): If True, return the smaller bounding box; otherwise, return the larger one. Returns: Optional[Union[List[int], Tuple[int, int, int, int]]]: The selected bounding box or None if the overlap ratio is not exceeded. rDrLrr N)rQrV)rArBrrZarea1Zarea2 overlap_ratior"r"r#_get_minbox_if_overlap_by_ratiosr?blocks thresholdcCst}t|}g}t|dD]^\}}t|dt|dD]N}|d|}||vs.||vr/qt|d|d||d} | durl|ddk} |ddk} | | kr_| rS|n|} ||d| n| dkre|n|} || qqt|dd D]} |d| =qt|S) a  Remove overlapping blocks based on a specified overlap ratio threshold. Args: blocks (List[Dict[str, List[int]]]): List of block dictionaries, each containing a 'block_bbox' key. threshold (float): Ratio threshold to determine significant overlap. smaller (bool): If True, the smaller block in overlap is removed. Returns: Tuple[List[Dict[str, List[int]]], List[Dict[str, List[int]]]]: A tuple containing the updated list of blocks and a list of dropped blocks. Zboxesrrz)rNrlimageT)reverse) r.r enumeraterrrr/addr;)rrrZdropped_indexesZoverlap_image_blocksr>Zblock1jZblock2Zoverlap_box_indexZis_block1_imageZis_block2_imageZ drop_indexindexr"r"r#remove_overlap_blockss6      rbboxc Cst|}t|}t|jdkr|nt|gd}t|jdkr#|nt|gd}t|d|d}t|d|d}t|d|d}t|d|d}||ksV||krXdS|dkret||||gS|dkr}tj||g||g||g||ggtjdStd ) a_ Compute the intersection of two bounding boxes, supporting both 4-coordinate and 8-coordinate formats. Args: bbox1 (tuple): The first bounding box, either in 4-coordinate format (x_min, y_min, x_max, y_max) or 8-coordinate format (x1, y1, x2, y2, x3, y3, x4, y4). bbox2 (tuple): The second bounding box in the same format as bbox1. return_format (str): The format of the output intersection, either 'bbox' or 'poly'. Returns: tuple or None: The intersection bounding box in the specified format, or None if there is no intersection. rrr rNrZpolyrOz.return_format must be either 'bbox' or 'poly'.) rr0rrvr rIrJZint16rK) rArBZ return_formatZrect1Zrect2rRrSrTrUr"r"r#get_bbox_intersections,    rc s|\}}}} |\} } } } ddddd}| ||| |||| || | |g}t|}t|}|||}t|dkrB|gfStdD]}||}|dd}||||<gg}}|D]&}t||dd}|td d krx||q_|td d kr||q_t|dkrt|dkr|D]M}|}|\}}}} |\} } } } | ||| |||| || | |g}t |}||}||}||||<t |||||\}}t|dkrqqfd d|D}t |}||fS| |t|}|||}qF||fS)aG Shrink the supplement region bbox according to the reference region bbox and match the block bboxes. Args: supplement_region_bbox (list): The supplement region bbox. ref_region_bbox (list): The reference region bbox. image_width (int): The width of the image. image_height (int): The height of the image. block_idxes_set (set): The indexes of the blocks that intersect with the region bbox. block_bboxes (dict): The dictionary of block bboxes. Returns: list: The new region bbox and the matched block idxes. r rrr)rrr rNrDrZ#match_block_overlap_ratio_thresholdg?Z#split_block_overlap_ratio_thresholdg?csg|]}|qSr"r").0idx block_bboxesr"r# Tsz1shrink_supplement_region_bbox..) rrJrrrrVrgetr/rIshrink_supplement_region_bboxrWremove)Zsupplement_region_bboxZref_region_bboxZ image_widthZ image_heightZblock_idxes_setrrrr r!Zx1_primeZy1_primeZx2_primeZy2_primeZindex_conversion_mapZedge_distance_listZedge_distance_list_tmpZ min_distanceZ src_indexriZ dst_indexZtmp_region_bboxZiner_block_idxesZsplit_block_idxesZ block_idxrZsplit_block_idxZsplit_block_bboxZ max_distanceZ iner_idxesZmatched_bboxesr"rr#rs                          rc Csl|dur|S|\}}}}|\}}}} tt||}tt||}tt||}tt|| } |||| g}|S)zUpdate region box with bboxN)rkrJrI) rZ region_boxrrr r!Z x1_regionZ y1_regionZ x2_regionZ y2_regionr"r"r#update_region_box^s   rformula_res_listocr_resc Cs|D]d}ttt|d\}}}}||f||f||f||fg}|d||d}|d||djdkrCt|dg|d<nt|d|dgf|d<|dd|d||d d qd S) zConvert formula result to OCR result format Args: formula_res_list (List): Formula results ocr_res (dict): OCR result Returns: ocr_res (dict): Updated OCR result Zdt_polysZ rec_formular*r,rZ rec_labelsformular)r+rN)r-rjrkr/rtrr0Zvstack) rrZ formula_resrnrorprqZ poly_pointsZformula_res_textr"r"r#!convert_formula_res_to_ocr_formatps& rcCs*tt|\}}}}t||||}|S)zCalculate bounding box area)rjfloatabs)rrrr r!Zarear"r"r#rQsrQcCs,|\}}|\}}||d||ddS)z/Calculate euclidean distance between two pointsr g?r")Zpoint1Zpoint2rrr r!r"r"r#caculate_euclidean_distsrcCsd}d}|j}|j}|j}|j}|durj|j}|j} t|j| dk} |dk} ||jko0||jk} | rJt|j|}t|j|}t|| dk} d} nt|j|j} ||dk}| ri|ri| ri| t|j|jkrid}n||dkrrd}||dkrzd}||fS)a Get segment start flag and end flag based on previous block Args: block (Block): Current block prev_block (Block): Previous block Returns: seg_start_flag (bool): Segment start flag seg_end_flag (bool): Segment end flag TN rrF) Zstart_coordinateZend_coordinateseg_start_coordinateseg_end_coordinateZ num_of_linesrrJrIwidth)blockZ prev_blockZseg_start_flagZ seg_end_flagZcontext_left_coordinateZcontext_right_coordinaterrZnum_of_prev_linesZpre_block_seg_end_coordinateZprev_end_space_smallZprev_lines_more_than_oneZoverlap_blocksZ edge_distanceZcurrent_start_space_smallr"r"r# get_seg_flagsR     rrlc Cs|rdddddddddd d }ns     <: 2 1$ #* 3. ZC