+ Ain1^RIt^RIHt^RIHt^RIHt^RIH t ^RI t ^RI H t ^RI Ht^RIHtRt]3R ltR t] P*!4R R R R/Rl4tR#)N)sparse)spsolve)laplace)utils)label)_build_matrix_innerc\P!W, ^4p\P!W,^,V4pW43#)r)npmaximumminimum)nd_idxradiusnd_shape bounds_lo bounds_his&&& Y/var/www/html/photoedit/myenv/lib/python3.14/site-packages/skimage/restoration/inpaint.py_get_neighborhoodr s4 6?A.I 6?Q.9I  c\P!WR7p^W1&\\V44p\P!V4pW4,p\P!V^R7pW4V3#)dtypeaxis)r zerosrwherestack)shapecenterr neigh_coefcoef_idx coef_valss&&& r_get_neigh_coefr"sZ%-JJ,-Jxx #H$Ixxq)H  **rc VPR ,pVP^,^,p\P!VP\R7p^V\ Ww)43VP ,&W,p V(V,p \P !V 4p \P!V 4p \P!V 4p \P!W34p\P !V 4p\\W4UUu.uFwpp\P!VV34NK upp4pV^8gp\P!VV\PRR7pVV,P4p\P!V4p\P!V4pV\P!VV8H4, p\P !V\P"R7p\P$!VV3VP&R7p\P !V\P"R7p\P !V\P"R7p\P !VVP&R7p/pVP)R 4p\P*!VP)R V344p ^p!^p"R p#\P,!V ^R7p \/V 4EFOwp#p$\1V$WqP4wp%p&\V&V%, 4p'\V$V%, 4p(VP3V'V(3R 4wp)p*V)f&\5V'V(VP&R7wp+p)p*V)V*3VV'V(3&V)V%R\P63,,p)\P8!V)VP4p,^p-\V*V,4F_wp.p/VV/,'dV#VV"&V/VV"&V.VV"&V"^, p"K/VV!R3;;,V.V V/R3,,,uu&V-^, p-Ka V-'gEKAV#VV!&V!^, p!EKR V#^,p0T!p1T"p2\;V0V1V2V VVVV VVVVV4 p3VRV3pVRV31R3,pVVP<3p4\>P@!VVV33V4R7p5V5RV3,p5\P$!VV3VP&R7p6VV6VR3&\CV5V6RR R 7p7V7P ^8XdV7R\P63,p7V7VV&V#uuppi) aSolve a (sparse) linear system corresponding to biharmonic inpainting. This function creates a linear system of the form: ``A @ u = b`` where ``A`` is a sparse matrix, ``b`` is a vector enforcing smoothness and boundary constraints and ``u`` is the vector of inpainted values to be (uniquely) determined by solving the linear system. ``A`` is a sparse matrix of shape (n_mask, n_mask) where ``n_mask`` corresponds to the number of non-zero values in ``mask`` (i.e. the number of pixels to be inpainted). Each row in A will have a number of non-zero values equal to the number of non-zero values in the biharmonic kernel, ``neigh_coef_full``. In practice, biharmonic kernels with reduced extent are used at the image borders. This matrix, ``A`` is the same for all image channels (since the same inpainting mask is currently used for all channels). ``u`` is a dense matrix of shape ``(n_mask, n_channels)`` and represents the vector of unknown values for each channel. ``b`` is a dense matrix of shape ``(n_mask, n_channels)`` and represents the desired output of convolving the solution with the biharmonic kernel. At mask locations where there is no overlap with known values, ``b`` will have a value of 0. This enforces the biharmonic smoothness constraint in the interior of inpainting regions. For regions near the boundary that overlap with known values, the entries in ``b`` enforce boundary conditions designed to avoid discontinuity with the known values. rconstant)outputmoderN:NNN)rFMMD_ATA) use_umfpack permc_spec)NN)"rr onesboolslicendimr flatnonzero concatenatetuplezipndiconvolveuint8sum count_nonzeroemptyintprrreshapeascontiguousarrayr enumeratergetr"newaxisravel_multi_indexrsizer csr_arrayr)8imagemaskoutneigh_coef_fullr!raveled_offsets n_channelsr edge_mask boundary_mask center_mask boundary_pts boundary_icenter_imask_i center_ptsbcmask_pts structuretmp nnz_matrixn_maskn_structnnz_rhs_vector_max row_idx_known data_knownrow_idx_unknowncol_idx_unknown data_unknown coef_cache mask_flatout_flat idx_known idx_unknown mask_pt_nr b_lob_hi coef_shape coef_centerr coefs_index1dnvalscoefi row_startknown_start_idxunknown_start_idxnnz_rhssp_shapematrix_unknownrhsresults8&&&&&& r!_inpaint_biharmonic_single_regionrv!sD2J  " "1 % *F $/I78IuVW%'$))34$M*t#K88M*L .J~~k*H ^^Z2 3F+&J\9VW9VAbnnaV,9VWXH 1$I ,,tYrxxj ICTJ  d #F *H""2"23(?"CCHH/rww?M-z:#))LJhhz9Ohhz9O88Jcii8LJ R I##CKKZ0@$ABHIKI88Lq1L&|4 6&vvzzB d4$;' FTM* $..*k)BLQ%  !0Ksyy" Ax6>u4EJ K0 1d1bjj=11&&x<5'*GD!||/8 ,/0 ,,0 [)q 9a<(D8AqD>,AA( + 5'0M) $ NI?5DA IO#!!G("(7+MHWHaK(J!H%% 9:(N $AvI.N ((FJ'syy 9C&C q^Se RF {{a2:: &CM JcXs $T split_into_regionsF channel_axisc `aVP^8d \R4hVRJpV'dVPRRM VPpWQP8wd \R4h\PP V4'd \ R4h\P!V4p\P!VP4pVPVRR7pVP\RR7pV'gVR\P3,p\P!VRR 7p^o^S,^,3VP,pS3VP,p \!WVPR 7wrp VP"R,p V S, pW(,pVP%^R 7VP'^R 73pV'Ed\(P*!VP^4p\(P,!VVR 7p\/V4pVV,p\(P0!V4p\3V^4EF'wpp\4;QJd+.V3R l\7VVP44FNK 5M$!V3R l\7VVP444pVV,V8HpV\9R43, pVV,P4p\P:!VR,P"Uu.uF pVV ,NK up4p\P<!VVR\P3,,^R 7p\?VV,VVV V V4VVV&EK* M\P:!VR,P"Uu.uF pVV ,NK up4p\P<!VVR\P3,,^R 7p\?WWzV V4\P@!VV^,V^,VR7V'g VR,pV#uupiuupi)aYInpaint masked points in image with biharmonic equations. Parameters ---------- image : (M[, N[, ..., P]][, C]) ndarray Input image. mask : (M[, N[, ..., P]]) ndarray Array of pixels to be inpainted. Have to be the same shape as one of the 'image' channels. Unknown pixels have to be represented with 1, known pixels - with 0. split_into_regions : bool, optional If True, inpainting is performed on a region-by-region basis. This is likely to be slower, but will have reduced memory requirements. channel_axis : int or None, optional If None, the image is assumed to be a grayscale (single channel) image. Otherwise, this parameter indicates which axis of the array corresponds to channels. .. versionadded:: 0.19 ``channel_axis`` was added in 0.19. Returns ------- out : (M[, N[, ..., P]][, C]) ndarray Input image with masked pixels inpainted. References ---------- .. [1] S.B.Damelin and N.S.Hoang. "On Surface Completion and Image Inpainting by Biharmonic Functions: Numerical Aspects", International Journal of Mathematics and Mathematical Sciences, Vol. 2018, Article ID 3950312 :DOI:`10.1155/2018/3950312` .. [2] C. K. Chui and H. N. Mhaskar, MRA Contextual-Recovery Extension of Smooth Functions on Manifolds, Appl. and Comp. Harmonic Anal., 28 (2010), 104-113, :DOI:`10.1016/j.acha.2009.04.004` Examples -------- >>> img = np.tile(np.square(np.linspace(0, 1, 5)), (5, 1)) >>> mask = np.zeros_like(img) >>> mask[2, 2:] = 1 >>> mask[1, 3:] = 1 >>> mask[0, 4:] = 1 >>> out = inpaint_biharmonic(img, mask) z!Input array has to be at least 1DNz&Input arrays have to be the same shapezMasked arrays are not supportedF)copy.C)orderrr)rSc3<"TFIwr\\VPS, ^4\VPS,V44xKK R#5i)rN)r-maxstartminstop).0slr@rs& r %inpaint_biharmonic..*sD AHBc"((V+Q/RWWv5Et1LMM AsAA)a_mina_maxrDr*).r)!r. ValueErrorrr ma isMaskedArray TypeErrorskimage img_as_floatr_supported_float_typerastyper,r>rzr"stridesrr~r3generate_binary_structurebinary_dilationr find_objectsr<r1r2r-arrayr6rvclip)rBrCrwrx multichannel img_baseshape float_dtyperDrfrgrEr r!channel_stride_bytesoffsets known_pointslimitskernel mask_dilated mask_labeled bbox_slices idx_regionbb_sliceroi_sl mask_regionotmpsostridesrFrs&&$$ @rinpaint_biharmonicrs}d zzA~<==t+L(4EKK$%++M "ABB uu5!!9::   'E--ekk:K LL5L 1E ;;t%; (D c2::o& ''%s #CFf*q."TYY.J)dii'K+:syy,(Oy ;;r?G>88F @S@ST@S1Q"666@STU&&8CO+D!D1M) y/ GGCvayq s; &k J;J Us 8P& P+)numpyr scipyrscipy.sparse.linalgr scipy.ndimagendimager3rr_sharedrmeasurer_inpaintrrfloatr"rvchannel_as_last_axisrrrrse'!) */ +aHN%NdNNr