1-Ph.ddlZddlmZmZmZddlmZddlm Z ddl m Z m Z ddl mZdd lmZmZmZdd lmZgd Zgd Zgd ZgdZdZdZdZdZGddeZdS)N)maximum_filterminimum_filterconvolve)integral_imagestructure_tensor)octagonstar)_censure_dob_loop)FeatureDetector_prepare_grayscale_input_2D_mask_border_keypoints)check_nD) rr)r) rr) r) )r r )r)r ) )rrrr)rrrr)rr)rr)rr)rrr#r)rrrrr#rr -.@Z) )rrr")rrr)rr)r%rr$)rr%)rr)r!rr)r&r!c tj|jd|jd||z dzftj}|dkr|j}||jdz|||jdz|jdzf|_t |}t||z dzD]F}||z}dd|zdzdzz } dd|dzzd|zzz } t|||dddd|f| | Gn|d krot||z dzD]X}t||zdz \} } t||zdz \} }t|t| | | ||dddd|f<Yn|d krt||z dzD]r}tt||zdz d}tt||zdz d}t|t|||dddd|f<s|S) Nrrdtypedob?rrrr r )npzerosshapefloat64itemsizestridesrranger OCTAGON_OUTER_SHAPEOCTAGON_INNER_SHAPEr_octagon_kernel STAR_SHAPESTAR_FILTER_SHAPE _star_kernel)image min_scale max_scalemoderesponse item_size integral_imgin inner_weight outer_weightmonominims W/var/www/html/test/jupyter/venv/lib/python3.11/site-packages/skimage/feature/censure.py _filter_imagerRBsdx QQY)>)BC2:H u}}% q) )  q) )HN1,= =  &e,, y9,q011  AA A !a%!)!11L"q!t)a!e"34L <!!!QQQ'!2L,     $   y9,q011 Q QA(Q):;FB(Q):;FB (BB0O0O P PHQQQ1W   Q y9,q011 D DA,Y]Q->?BCA,Y]Q->?BCA ( Q0B0B C CHQQQ1W   OcZ|d|zzdzd|z|dzzz }|d|zzdzd|z|dzzz }d||z z }d|z }|d|zz|d|zzz dz}t||} tj|d|zz|d|zzf} t||| || || f<|| z||z| zz } | S)Nrrr3)r r4r5) rLrMrNrOouterinnerrKrJc outer_oct inner_octbfilters rQr=r=us !b&[Q R26!2 2E !b&[Q R26!2 2E%%-(L;L q2v+"q2v+ &1,ABI"q2v+rAF{344I#BOOIadAqbDjY&,*E)RRG NrSc ||dzz|z |dzz }t|}tj|}t|||| || f<dtj||z z }dtj|z }||z||z|zz }|S)Nrr3)r r4 zeros_likesum)rPrIrW outer_star inner_starrKrJrZs rQr@r@s AF QaAaJz**J!!WWJq!tQrTz"&j!899:L +++LZ'<,+F**TTG NrScht||d\}}}d|||zdz|||z|dzz zk<dS)Nrc)orderFrr ) feature_maskrAsigmaline_thresholdArrArcAccs rQ_suppress_linesrisL$UE>>>MCcMRL#)!NcCi#q&6H$IIJJJrSc*eZdZdZ d dZdZd S) CENSUREa CENSURE keypoint detector. min_scale : int, optional Minimum scale to extract keypoints from. max_scale : int, optional Maximum scale to extract keypoints from. The keypoints will be extracted from all the scales except the first and the last i.e. from the scales in the range [min_scale + 1, max_scale - 1]. The filter sizes for different scales is such that the two adjacent scales comprise of an octave. mode : {'DoB', 'Octagon', 'STAR'}, optional Type of bi-level filter used to get the scales of the input image. Possible values are 'DoB', 'Octagon' and 'STAR'. The three modes represent the shape of the bi-level filters i.e. box(square), octagon and star respectively. For instance, a bi-level octagon filter consists of a smaller inner octagon and a larger outer octagon with the filter weights being uniformly negative in both the inner octagon while uniformly positive in the difference region. Use STAR and Octagon for better features and DoB for better performance. non_max_threshold : float, optional Threshold value used to suppress maximas and minimas with a weak magnitude response obtained after Non-Maximal Suppression. line_threshold : float, optional Threshold for rejecting interest points which have ratio of principal curvatures greater than this value. Attributes ---------- keypoints : (N, 2) array Keypoint coordinates as ``(row, col)``. scales : (N,) array Corresponding scales. References ---------- .. [1] Motilal Agrawal, Kurt Konolige and Morten Rufus Blas "CENSURE: Center Surround Extremas for Realtime Feature Detection and Matching", https://link.springer.com/chapter/10.1007/978-3-540-88693-8_8 :DOI:`10.1007/978-3-540-88693-8_8` .. [2] Adam Schmidt, Marek Kraft, Michal Fularz and Zuzanna Domagala "Comparative Assessment of Point Feature Detectors and Descriptors in the Context of Robot Navigation" http://yadda.icm.edu.pl/yadda/element/bwmeta1.element.baztech-268aaf28-0faf-4872-a4df-7e2e61cb364c/c/Schmidt_comparative.pdf :DOI:`10.1.1.465.1117` Examples -------- >>> from skimage.data import astronaut >>> from skimage.color import rgb2gray >>> from skimage.feature import CENSURE >>> img = rgb2gray(astronaut()[100:300, 100:300]) >>> censure = CENSURE() >>> censure.detect(img) >>> censure.keypoints array([[ 4, 148], [ 12, 73], [ 21, 176], [ 91, 22], [ 93, 56], [ 94, 22], [ 95, 54], [100, 51], [103, 51], [106, 67], [108, 15], [117, 20], [122, 60], [125, 37], [129, 37], [133, 76], [145, 44], [146, 94], [150, 114], [153, 33], [154, 156], [155, 151], [184, 63]]) >>> censure.scales array([2, 6, 6, 2, 4, 3, 2, 3, 2, 6, 3, 2, 2, 3, 2, 2, 2, 3, 2, 2, 4, 2, 2]) rrDoB333333?rc|}|dvrtd|dks|dks ||z dkrtd||_||_||_||_||_d|_d|_dS)N)r2r r z/`mode` must be one of 'DoB', 'Octagon', 'STAR'.rrz@The scales must be >= 1 and the number of scales should be >= 3.) lower ValueErrorrBrCrDnon_max_thresholdre keypointsscales)selfrBrCrDrqres rQ__init__zCENSURE.__init__szz|| 1 1 1NOO O q==IMMY-BQ-F-FU #" !2, rSc.t|d|j|jz }tjt |}t ||j|j|j}t|d|k}t|d|k}||z}d|||j k<td|D]7}t|dddd|f|d|j|zdz dz z|j 8tj|dd|f\}} } tj|| g} | |jzdz} |jdkr| |_| |_dStj| jd t( } |jd krtt|jdz|jD]U}t*|dz d dz dzt*|dz dz} | t-|j| | | |kzz} Vn|jd krt|jdz|jD]h}t.t0|dz d t.t0|dz d dzz} | t-|j| | | |kzz} i| | |_| | |_dS) zDetect CENSURE keypoints along with the corresponding scale. Parameters ---------- image : 2D ndarray Input image. r)rrrFrNg@.r2rr0r r )rrCrBr4ascontiguousarrayrrRrDrrrqr:rirenonzero column_stackrrrsr5r6boolr;rr>r?)rtrA num_scalesfilter_responseminimasmaximasrcrHrowscolsrsrrcumulative_maskrWs rQdetectzCENSURE.detects0 ^dn4 $%@%G%GHH( 4>4>49   !)<<O )<<O( AF _t'==>q*%%  A QQQ1W%dnq(1,33#      Z S!J,5F(GHHdFOT4L11 $.(1, 9  &DN DK F(9?1#5TBBB 9 ! !4>A-t~>>  (Q/2Q61A-t~>>  0Q7:; !21q5!9!!<=BC #9%+yRS#T#TaK$#?3_- rSN)rrrlrmr)__name__ __module__ __qualname____doc__rurrSrQrkrksZSSn  4U.U.U.U.U.rSrk)numpyr4 scipy.ndimagerrr transformrcornerr morphologyr r censure_cyr feature.utilrrr _shared.utilsrr;r<r>r?rRr=r@rirkrrSrQrsBBBBBBBBBB&&&&&&$$$$$$&&&&&&&&)))))) %$$$$$       M L L     000f   SSS E.E.E.E.E.oE.E.E.E.E.rS