Sr LO091- 344 ie C Specify the Need of aay hore in_short two Categories 2 se z ; ge Enhancement 2 Hence explain of spacial domain ‘enage eee [4 enhancement - The prim Loft : ad ' my need of image enhancement is bo _ : e visual appearance of an image and to _|male it more suitable for analysis by humans er £ machines: nes : “Image acquired by/ From sensors may suPPer __| Feorn low contrast, noise, poor iluminahion ox blurring _ _—_# {due to limitabons of imaging devices or envi condiWons- mental —_- —|___Trage enhancement hetps inh ighlighting (of __ _—#* —-tma ging. devices or environmental conditions) tmp. - Features, improving contrast, reducin emphasizing details such_as__edges and boundaries _ __| thereby tn creasin the effectveness of image > ace interpretation 4 subsequent processing tank: eee | Coteqorten of spahal Domain— Tenage—Fnhancerment: a ees Spabal. domain image enhancement technique Bee eee __| operate divectly on image pixels and are broadly classified into categories + __ ease a } porn’ processing techniques s ‘ ei Tn this, the enhancement of pixel depends on —__ __|_ its pwn intensity value: Each pixel is processed _—Lindependenty using_an_intensity transformation funchon. _|-eg-image neqakve, loqa ritomic transformaboen, eee ir power low transformaton, controot streching | These +echniquen_are— mainly used +o improve contraot.& brightness. __ eee eeeTechnique a depends on ik, or__kernel is perform operations 2) spatial Riltering | tn this, enhancemen Gitex mask neighboring. pixels: moved over the _ suchas smoothing image to ng harpeni : ors Uae "2 us imagen, | ss thi Rlters seduce Ae | while sharpening _Piters ann edge - | ine details. — — _ [ — ~| |@.4) Explain average Aiteseg ofan. image at | | @xomnple = pote seg | weve ys 3 Eee Aver cage. ‘Altering co CMean) ——— _4 Average filtering is a linear_spabial smoothing ng | Filter used to reduce naotse and smal\ “ee : variation into_an image: ——— £ach_pixel_is replaced by the. overage. of Neighbor n. ircelgs = _| "te talte cotied “Mean. Biter or Bor Alter 4 Maske eee es Cte fix example, a ae _jj | consider Bxd ‘eightor hood, so ae —|_pixel values: re + ——— eee a See ee jets Dees SO tea Slab. 60. Gol 2 7 7 who fede af | | Jaro i i _| step t= Add all values |stepa= Divide by gae (ot zor Bor20+y O* 604204 Gor gg> 360 so the olp. The center Pinel C40) remains 4o alter Ritering grta? 3c) Write expression, draw graph & oo Lin Short log transformation £ power law transformation, 4— ——— aouaeeees ae oe Eee _jLog transformation fee Expression - ——— ——— | where = SS __ = input gray level ss ef 3 a to qray level ae C= constant (seating factor) cee L-\ ae a: ——_1}2 —____— th = nput gray ver! * function expands dark pixel values. — i anc pixel values Te orapress bri nt p = Small sacle “in dark regions become : more_visibl a ae _ a - lo+ pet _| Power . ay __ Expression, Ae Get Y= gamma volute es —— jon Law Teansformatte rile gray. level — 3+ 0/p gray level” C=constank(gta) Obtain hi 2 a a : ae pie & Histogram equilizaton Mania AAG) bit Per Pixel, Given ‘yy Pere. mg “ i oe eT | | Gwen CAR, 4 bits lone) gray Nevel. crange fom Co-15) Lpeatele _ | “Tokal pixet N= 4 4 stg Z step 1- Histogram (ny) SSeeeceee eee geese | ¢ount octurentes of each gray level Gray level ry. count np oases Be i Oistribubon (PDF)| == Sk aes 7 oe | : ent . ; | evep 2 cumutatve distsbuben funchon (COR) ep > oY | : ce m4 375, 0: 8125, 0°81S, (+0 3 E cpl + 0:06.25, O1BTS, 0:4375, 0: 8125, x =o 7 - - - —— |step ates Tra nsforcnahen Faunetion a Chel) xCDe = 15 xeOF | —4r= i OO a > |, —_____cpe____ * |g 255 GtoGog 85 66g see [cast joo A:1976 Bp | to __ O:4375 = -l2 os gta so gts pe lio + is on is | a-4y) Explain -non-\inear Filtering method of. an | —image v —with example. ~ B +) | i= : Non linear filberin 4.6 a technique new _ FF | by a so ~oukput+—pixel value is not obtained + atinear crobinahon. 4 Tanti linear filters, non linear Alters aed Sen | linear operations such as median minim a, This, especially ePlectve For morse 4 parbcutart \ytrnpulse noise —Femovat | ree + - Most aeins non-linear Filter is. 5 Me edvan Biber bs The ae egan filter veplaces the vatue aa | \ \ median value of ifs neighboring Pixels. cram ple= = ane sides 3x3. neighbo hod—Limage processing __| decomposed. —| places) each repres Ses —_| Each pixel tgp an grbit {rn —_lin lotnary fore : —|_pinelvatu e2 150 ; @ ele ote 2 ee 9 255 10 : = ae ie 2 = = here 255 { salt noise : Step i- Arrange Values tn ascending order ——— | a Be Vor ia ig geo Pfter Sorting : [ BR, BANOO MN 1e\a,2s6 Step 2- And median CUAie alee eS (Step B= eplace center pixel — fae a ___ $0, 255 \$ replaced by Wp GQ -4c) Explain Git Plane Sti | importante. Bit plane slteing— is a technique in digital _____ where a grayscale image is _into_a_set of binary images Coit eoting one bit of pixel intenaily age can be wxiten | Binary represen tation = 15°Bk 1 =) : — —4 | Bit_o (188) * 9 ~ 4 “Fabel the bla Rama. separate image calle a _bit_plane: Tm portance of Bit. -plane_siteing : [2 tnagecentpeessian— —— 5— ee @ C6269) -349 - jase seenty the aes of Tmnage enhancement & explain _Spacial Ailtenng in tra aed Pedal ten a age enhancement: (57. gab what is necessity of High boost Altering? Hoo | this achieved? — cs) | wh eo we smooth an image to remove noise, the —_ [[Link] becomes, boyy soft = feb et as = _| = Blurry — ss 7 Edges berome duu Rut we don't want to Nose the whole tmage,we—— ‘ust want to make the details _shand out more, details not _the entire imag _—_| So we boost the— || Step-by- Step (Hous it is achieved) — ) Start with ort inal _irnage — | contains bacle ound, slow intensity change) ' — Detatls & edges (sudden intensity change) + |) Create Blurred versvon - 7 _| Use on average [Grousstan Alter: - a What hoppers, - - — Background aeemains : _ _ a aEdqes 2 fine details fade = __| | so ib contains —— = ~ i» Csrooo0*h) information. —4 | | ger_onl detar'\s 7 D Subtract Original = Blurred + Derails Troage Bedfb _— — aa + = af — ——— —~ | ee 2 aoa \ patois oy Sf _number Choast factor) _ Mu ply a - “Te Boost = 2 > details become Aoub e_stronge 5) Add Back fr_onginal_ _ ——— Fine Tmage = original + Goosted details. | @-3c) What_is_median iter? Apply B*S median _ | Piter to given 4*4 qraysece image & find the i qe gray g | processed “mage by considering _ouker row 8 Lcolumns as it is. a C5] A_medsan Filter is a non-l lina” s al Biber weds to_ reroove_impulse oie i ue : e. ' ‘ condition a tabile ~Prese ae ara outer rows and columns rem _—— —. ~ sremain uneha \ Panged — Sone only apply the Alter to nner a pind — position to process (2,0) (2,3) ett | (ap CyAsa Neighborhood = - _|[s ¢ 4 7 : coo qf] == - 4 5 6 isl vi 7 oe values? 5,6,7,6 0,7, 5,6, '8 : ~ WCested 05,5, 6,456.7 Tis - ; | median (Sth value) =6- : step 2- ev ginenGalayia oe Twa _ _| ¢ 2 9 Boe = J utes ad lo 7 3 oe — = G - N65 48 dient rit Reales 7G. ot gh On eR Gr ler ——— _| sorted > oO. 6 ee QQ, 9, te median: F — ae po stp A | medians 6 ee Shep = “pixel 23 2 moran 1,8, a \s 5 8,6 1,8 — _goxted 70,6, e141, (8,888 _ median: +Final processing image outer pixel nia aioe G eG é G 7 1 1 7 Qua s : “Blk of an ima with _ Alterin oO age G40) Explain. a i jp ege —_ Ter Repeated [eoog] - 234% 33h) eee ee lpm is is_an_en “enhancemen: ees —|an_ image — —}—particular features Such_as, ——{=Tumors in medic highlight a specie Sei oP Gray. lewels in Th is useful, ee woe ean bb nen phan al tian sea ae ee ~tuater bodies in Satellite ¢ = SpeciGe abjects in indus __ lg. 40) Oblain Wishogram and perform. hisagrarm [Satication. = si C89) bi i 3 4b) ae “the. he lowing. comep ini image enhancement: _ feta —used tofe eet. ee ae : oo : ‘Siept- Histogram _ a _| count freq® of each. gray feel Po Saute | Gray fe —— _ count ny a"Step Transfoernation fun” Lge = & Cen 2 Ee) ——— Step S- R eplace_Tnterned Intensities tmoapping __ tae 4>e hale! Ainol equolized image _