h~SrSSKJr SSKrSSKJr SSKJrJrJ r J r SSK J r SSK r SSKrSSKJs Js Jr SSKJr SSKJrJr SS KJrJrJr SS KJr SS K J!r! SS K"J#r#J$r$ "S S\ 5r%SSjr&SSjr'S Sjr(S!Sjr)S"Sjr*S#Sjr+S$Sjr,S%Sjr-S&Sjr.S'Sjr/S(Sjr0S)Sjr1S*Sjr2S+Sjr3S,Sjr4g)-zFunctional implementations for image mixing operations. This module provides utility functions for blending and combining images, such as copy-and-paste operations with masking. ) annotationsN)Sequence)AnyLiteral TypedDictcast)warn)Crop)LongestMaxSizeSmallestMaxSize) BboxProcessordenormalize_bboxesnormalize_bboxes)Compose)KeypointsProcessor)$NUM_BBOXES_COLUMNS_IN_ALBUMENTATIONS'NUM_KEYPOINTS_COLUMNS_IN_ALBUMENTATIONScB\rSrSr%SrS\S'S\S'S\S'S\S'S rg ) ProcessedMosaicItemzRepresents a single data item (primary or additional) after preprocessing. Includes the original image/mask and the *preprocessed* annotations. np.ndarrayimagenp.ndarray | Nonemaskbboxes keypointsN)__name__ __module__ __qualname____firstlineno____doc____annotations____static_attributes__rh/var/www/fran/franai/venv/lib/python3.13/site-packages/albumentations/augmentations/mixing/functional.pyrrs!     r%rcUupEUR5n[R"US:5nUUSUS4XgSU-USU-4'U$)a*Blend images by copying pixels from an overlay image to a base image using a mask. This function copies pixels from the overlay image to the base image only where the mask has non-zero values. The overlay is placed at the specified offset from the top-left corner of the base image. Args: base_image (np.ndarray): The destination image that will be modified. overlay_image (np.ndarray): The source image containing pixels to copy. overlay_mask (np.ndarray): Binary mask indicating which pixels to copy from the overlay. Pixels are copied where mask > 0. offset (tuple[int, int]): The (y, x) offset specifying where to place the top-left corner of the overlay relative to the base image. Returns: np.ndarray: The blended image with the overlay applied to the base image. r)copynpwhere) base_image overlay_image overlay_maskoffsety_offsetx_offset blended_image mask_indicess r&copy_and_paste_blendr4*sk0 HOO%M88L1,-LLYQQ MMq/H,l1o.HHI r%cdUupVUupxUupXW-n Xh-n U S-n XS-S-- nU S-nXS-S-- nX- S-nUU- S-nUR"U6nUR"U6nU [UU-5-nU[UU-5-n[U [UU55n[U[UU55nUU4$)aXCalculates the center point for the mosaic crop using proportional sampling within the valid zone. Ensures the center point allows a crop of target_size to overlap all grid cells, applying randomness based on center_range proportionally within the valid region where the center can lie. Args: grid_yx (tuple[int, int]): The (rows, cols) of the mosaic grid. cell_shape (tuple[int, int]): Shape of each cell in the mosaic grid. target_size (tuple[int, int]): The final output (height, width). center_range (tuple[float, float]): Range [0.0-1.0] for sampling center proportionally within the valid zone. py_random (random.Random): Random state instance. Returns: tuple[int, int]: The calculated (x, y) center point relative to the top-left of the conceptual large grid. r()uniformintmaxmin)grid_yx cell_shape target_size center_range py_randomrowscolscell_hcell_wtarget_htarget_w large_grid_h large_grid_wmin_cxmax_cxmin_cymax_cyvalid_wvalid_hrel_xrel_ycenter_xcenter_ys r&calculate_mosaic_center_pointrRMs4JDNF$H=L=L]F \a/ /F ]F \a/ /Fo!Gvo!G   | ,E   | ,EGeO,,HGeO,,H63x01H63x01H X r%c UupEUupgUupUup[R"US-5U-n [R"US-5U-n XS-- nXS-- nX-nX-nSSjnU"XU5nU"XU5n[U5S- n[U5S- n/n[U5HhnUUnUUS-n[U5HInUUnUUS-nUR [ U5[ U5[ U5[ U545 MK Mj U$)aCalculates placements by clipping arange-defined grid lines to the crop window. Args: grid_yx (tuple[int, int]): The (rows, cols) of the mosaic grid. cell_shape (tuple[int, int]): Shape of each cell in the mosaic grid. target_size (tuple[int, int]): The final output (height, width). center_xy (tuple[int, int]): The calculated (x, y) center of the final crop window, relative to the top-left of the conceptual large grid. Returns: list[tuple[int, int, int, int]]: A list containing placement coordinates `(x_min, y_min, x_max, y_max)` for each resulting cell part on the final output canvas. r(r6cb[R"XU5n[R"U5U- $)N)r*clipunique)coordsmin_valmax_valclipped_coordss r& _clip_coords/calculate_cell_placements.._clip_coordss(':yy(722r%)rWrrXr8rYr8returnr)r*arangelenrangeappendr8)r;r<r= center_xyr@rArBrCrDrErPrQy_coords_largex_coords_large crop_x_min crop_y_min crop_x_max crop_y_maxr[y_coords_clippedx_coords_clippednum_x_intervalsnum_y_intervalsresulty_idxy_miny_maxx_idxx_minx_maxs r&calculate_cell_placementsrts>*JDNF$H"HYYtax(61NYYtax(61NM)JM)J&J&J3 $N K#N K*+a/O*+a/O F' ' +?+E$U+E$UQY/E MM3u:s5z3u:s5zJ K,( Mr%c JUcUS:XaggUcgURnURnX4:wa'SSUSUSUSURS UR3 4$US :XaJURS nURS nXV:wa'SSUSUS USURS UR3 4$g) zFChecks if the dimensions and channels of item_data match primary_data.r)Fz$Item is missing required key 'image')TNFzItem 'z' has z dimensions, but primary has z. Primary shape: z, Item shape: z channels, but primary has )ndimshape) primary_data item_datadata_key primary_ndim item_ndimprimary_channels item_channelss r&_check_data_compatibilityrs w @ $$LI XJfYK/L\N[*001 ?P R  q'--b1!+  , &7RScRde""."4"4!5^IOOCTV  r%c "[U[5(d[SUS3[SS9 /$/nUR S5nUR S5n[ U5Hupg[U[ 5(d[SUSUS 3[S S9 M2S nSU4SU44HWupUR U 5n [U U [S U 55upU (aM8SUSUS U SU 3n[U[S S9 Sn O U(dMURU5 M U$)zVFilters a list of metadata dicts, keeping only valid ones based on data compatibility.zMetadata under key 'zU' is not a Sequence (e.g., list or tuple). Returning empty list for additional items.rv) stacklevelrrzItem at index z in 'z$' is not a dict and will be skipped.TLiteral['image', 'mask']z%' skipped due to incompatibility in 'z': F) isinstancerr UserWarningget enumeratedictrrra)metadata_inputmetadata_key_namedata valid_items primary_image primary_maskiitem item_is_valid target_keyprimary_target_dataitem_target_data is_compatible error_msgmsgs r&filter_valid_metadatarsZ nh / / "#4"569 :    KHHW%M88F#L^,$%%  5):(;;_`   m $ \ "0 +J $xx 3 '@# /<( $M !=$QCu->,?@..8\YKIS+!4 % '0 * =   t $A-D r%c0U(d0$[USS9nUS0nUVs/sH oUU:wdM UPM nn[[SU55nURU5 [ [ U5[ U55n[U5H n XyXFU 'M U$s snf)aAssigns item indices to placement coordinate tuples. Assigns the primary item (index 0) to the placement with the largest area, and assigns the remaining items (indices 1 to num_items-1) randomly to the remaining placements. Args: num_items (int): The total number of items to assign (primary + additional + replicas). cell_placements (list[tuple[int, int, int, int]]): List of placement coords (x1, y1, x2, y2) for cells to be filled. py_random (random.Random): Random state instance. Returns: dict[tuple[int, int, int, int], int]: Dict mapping placement coords (x1, y1, x2, y2) to assigned item index. c0USUS- USUS- -$)Nr6rrvr(r)rWs r&,assign_items_to_grid_cells..Cs#F1Iq 1fQi&)6KLr%)keyrr()r9listr`shuffler:r_) num_itemscell_placementsr?primary_placementplacement_to_item_indexrWremaining_placementsremaining_item_indices num_to_assignrs r&assign_items_to_grid_cellsr's,   L 1E 2A`vN_D_F`"%9"56 ,-0137M3NOM = !;Q;TQ 78" #"as BBc URU5nU(aX ;aURU5bURRnU(aTUVs/sH oUU;dM UPM sn(a5[SUS[ UR5R SUSUS3 5eSUSX U0nU(aUHnXP;dM XXe'M UR U5 URU5$U$s snf)zIHelper to preprocess annotations (bboxes or keypoints) for a single item.zItem contains 'z' but is missing required label fields: {[field for field in required_labels if field not in item]}. Ensure all label fields declared in z (z+) are present in the item dictionary when 'z ' is present.r)rparams label_fields ValueErrortyper preprocess)r processorr| original_datarequired_labelsfield temp_datas r&_preprocess_item_annotationsrXs HHX&MX%$((8*<*H#**77 ?X?%SWFW?X!(,77;IP>V>VW&667I[!&$W-#''/  r%cUbUOSn[R"XRS9n[R"XUS9nUR5H(upU R U5n U cMUuppXX2X24'M* U$)aAssembles the final mosaic image or mask from processed cell data onto a canvas. Initializes the canvas with the fill value and overwrites with processed segments. Handles potentially multi-channel masks. Addresses potential broadcasting errors if mask segments have unexpected dimensions. Assumes input data is valid and correctly sized. Args: processed_cells (dict[tuple[int, int, int, int], dict[str, Any]]): Dictionary mapping placement coords to processed cell data. target_shape (tuple[int, ...]): The target shape of the output canvas (e.g., (H, W) or (H, W, C)). dtype (np.dtype): NumPy dtype for the canvas. data_key (Literal["image", "mask"]): Specifies whether to assemble 'image' or 'mask'. fill (float | tuple[float, ...] | None): Value used to initialize the canvas (image fill or mask fill). Should be a float/int or a tuple matching the number of channels. If None, defaults to 0. Returns: np.ndarray: The assembled mosaic canvas. rr) fill_valuer)r*rfullitemsr)processed_cellsrrr|r actual_fillrcanvasr cell_datasegmentrrtgt_x2tgt_y2s r&$assemble_mosaic_from_processed_cellsrss:*$K+3JWW\ FF(7'<'<'>#--)  -= *FF3:6=&-/ 0(? Mr%c 0n UR5H8upX;n U upnnUU- nX- n[X5n[UU UU4UUUUUUS9 X'M: U $)aProcesses the geometry (cropping/padding) for all assigned mosaic cells. Iterates through assigned placements, applies geometric transforms via process_cell_geometry, and returns a dictionary mapping final placement coordinates to the processed item data. The bbox/keypoint coordinates in the returned dict are *not* shifted yet. Args: canvas_shape (tuple[int, int]): The shape of the canvas. cell_shape (tuple[int, int]): Shape of each cell in the mosaic grid. placement_to_item_index (dict[tuple[int, int, int, int], int]): Mapping from placement coordinates (x1, y1, x2, y2) to assigned item index. final_items_for_grid (list[ProcessedMosaicItem]): List of all preprocessed items available. fill (float | tuple[float, ...]): Fill value for image padding. fill_mask (float | tuple[float, ...]): Fill value for mask padding. fit_mode (Literal["cover", "contain"]): Fit mode for the mosaic. interpolation (int): Interpolation method for image. mask_interpolation (int): Interpolation method for mask. Returns: dict[tuple[int, int, int, int], ProcessedMosaicItem]: Dictionary mapping final placement coordinates (x1, y1, x2, y2) to the geometrically processed item data (image, mask, un-shifted bboxes/kps). ) r<rrrrrrrr)rget_cell_relative_positionr) canvas_shaper<rfinal_items_for_gridrrrrrprocessed_cells_geomritem_idxrrrrrrDrErs r&process_all_mosaic_geometriesr sdRT'>&C&C&E"#-)9&F??23CR 2G!"H-'1' 2 .'F* r%cUup#UupEpgUS- nUS- n XF-S- n XW-S- n X:aSn O X:aSn OSn X:aSn O X:aSn OSn SSS S S .n[S URX4S55$) aKDetermines the position of a cell relative to the center of the target canvas. Compares the cell center to the canvas center and returns its quadrant or "center" if it lies on or very close to a central axis. Args: placement_coords (tuple[int, int, int, int]): The (x_min, y_min, x_max, y_max) coordinates of the cell. target_shape (tuple[int, int]): The (height, width) of the overall target canvas. Returns: Literal["top_left", "top_right", "center", "bottom_left", "bottom_right"]: The position of the cell relative to the center of the target canvas. g@topbottomrleftrightrrrr))rr)rr)rr)rrILiteral['top_left', 'top_right', 'center', 'bottom_left', 'bottom_right'])rr)rrrDrEx1y1x2y2canvas_center_xcanvas_center_y cell_center_x cell_center_yv_posh_pos position_maps r&rrs&&H%NBBnOnOWOMWOM&  (&  ($%)+ L S%2 r%ch0nUup4UR5GHupVUSSupxUSUS- n USUS- n URS5n URS5n USURS 5S .n U bU RS:at[R"U 5n[ XU S .5n[R "XxXx/[RS 9n[R"UU5n[UX4S .5nUU S'O[R"S[45U S'U bdU RS:aT[R"U 5n[R "XxS/URS 9n[R"UU5nUU S'O[R"S[45U S'[!S U 5X%'GM U$)aShifts coordinates for all geometrically processed cells. Iterates through the processed cells (keyed by placement coords), applies coordinate shifting to bboxes/keypoints, and returns a new dictionary with the same keys but updated ProcessedMosaicItem values containing the *shifted* coordinates. Args: processed_cells_geom (dict[tuple[int, int, int, int], ProcessedMosaicItem]): Output from process_all_mosaic_geometries (keyed by placement coords). canvas_shape (tuple[int, int]): The shape of the canvas. Returns: dict[tuple[int, int, int, int], ProcessedMosaicItem]: Final dictionary mapping placement coords (x1, y1, x2, y2) to processed cell data with shifted coordinates. Nr6rrvr(rrrr)rr)heightwidthrr)rrrr*rrrfloat32rrremptyrrrrr)r rfinal_processed_cellscanvas_hcanvas_wrcell_data_geomrr cell_width cell_heightrrfinal_cell_datarbbox_denoramlizedrshifted_bboxes_denormalizedrrrrs r&shift_all_coordinatesr,.s(SU%H,@,F,F,H()"1-%a(+;A+>> &q),ztuple[float, float]r? random.Randomr]r-) r;r-r<r-r=r-rbr-r]list[tuple[int, int, int, int]])rzrr{rr|rr]ztuple[bool, str | None])rzSequence[dict[str, Any]] | Nonerstrrdict[str, Any]r]list[dict[str, Any]])rr8rr/r?r.r]$dict[tuple[int, int, int, int], int])rr1rz)BboxProcessor | KeypointsProcessor | Noner|zLiteral['bboxes', 'keypoints']r]r)rr2rzBboxProcessor | NonerzKeypointsProcessor | Noner]list[ProcessedMosaicItem])rr-rr-rrr]tuple[int, int, int, int])r<r-rrrr-rfloat | tuple[float, ...]rr6rLiteral['cover', 'contain']rr8rr8rrr]r)rrrr5r]r) rz/dict[tuple[int, int, int, int], dict[str, Any]]rztuple[int, ...]rznp.dtyper|rrz float | tuple[float, ...] | Noner]r)rr-r<r-rr3r r4rr6rr6rr7rLiteral[cv2.INTER_NEAREST, cv2.INTER_NEAREST_EXACT, cv2.INTER_LINEAR, cv2.INTER_CUBIC, cv2.INTER_AREA, cv2.INTER_LANCZOS4, cv2.INTER_LINEAR_EXACT]rr8r]4dict[tuple[int, int, int, int], ProcessedMosaicItem])rr5rr-r]r)r r9rr-r]r9)5r" __future__rrandomcollections.abcrtypingrrrrwarningsr cv2numpyr*1albumentations.augmentations.geometric.functional augmentations geometric functionalr-albumentations.augmentations.crops.transformsr -albumentations.augmentations.geometric.resizer r albumentations.core.bbox_utilsr rralbumentations.core.compositionr#albumentations.core.keypoints_utilsr$albumentations.core.type_definitionsrrrr4rRrtrrrrrrrrrr rr,rr%r&rKsv # $00 FF>Y^^3B !) !        F: ::!:& :  :  :z: ::!: : % :z%#% %'% %P5355 5 5p.#.#4.#.#* .#b) )8)-) )X,(2 @9&9&9&]9& 9&xii i"i $ i ) i * iii]iiX$,$/$$N/D/!/ /' / + /  /dJ !J J BJ 4 J $ J ) J *J J "#J 4:5J Z9/9!9O9x@!N@!!@!:@!r%