opencv - Scan Business Card Tesseract and Leptonica iOS -


i trying scan business card using tesseract ocr, doing sending image in no per-prossesing, heres code using.

 tesseract* tesseract = [[tesseract alloc] initwithlanguage:@"eng+ita"]; tesseract.delegate = self; [tesseract setvariablevalue:@"0123456789abcdefghijklmnopqrstuvwxyzabcdefghijklmnopqrstuvwxyz@.-()" forkey:@"tessedit_char_whitelist"];  [tesseract setimage:[uiimage imagenamed:@"card.jpg"]]; //image check [tesseract recognize];   nslog(@"here text %@", [tesseract recognizedtext]);

picture of card

this output

as can see accuracy not 100%, not concerned figure can fix simple per-processing. if notice mixes 2 text blocks @ bottom, splits address, , possibly other information on other cards.

how can possibly use leptonica(or else maybe opencv) group text somehow? possibly send regions of text on image individually tesseract scan? i've been stuck on problem while possible solutions welcome!

i recommend using algorithm called "run length smoothing algorithm" (rlsa). algorithm used in lot of document image processing systems, though not every system expose part of api.

the original paper published in 1982 , requires payment. however, same algorithm cited many other papers on document image processing, can find implementation details , improvements.

one such paper this: http://www.sciencedirect.com/science/article/pii/s0262885609002005

the basic idea scan document image row row, recording width of gaps between letters.

then, nearby text characters can combined filtering on width of gaps, , setting small gaps same color text. result large connected components represent:

  • words,
    • by closing gaps between characters,
  • text lines,
    • by closing gaps between words, ,
  • paragraphs
    • by scanning column column , closing vertical gaps between text lines.

if not have access document image analysis libraries expose functionality, can mimic effect by:

  • using morphological operations (morphological closing), ,
  • perform connected-component labeling on result.

most image processing libraries, such opencv, provides such functionality. might less efficient take approach because have re-run algorithm using different text gap sizes achieve different levels of clustering, unless user provides application text gap sizes.


-

Comments

Post a Comment

Popular posts from this blog

html - Sizing a high-res image (~8MB) to display entirely in a small div (circular, diameter 100px) -

first of all, thank being patient know simple question more couple of weeks' experience figure out. i've seen folks persevere , helpful on here , hope you're willing me. here fiddle: http://jsfiddle.net/65uzf/ here 1 of places searched answers on here: ( vertically center image thumbnails fix-sized div ) tried solutions on there , didn't work out me. i've searched ~4 or 5 others, , been unsuccessful those, too. what want take large image (what linked background-image #ysypic , #tabpic divs) , display thumbnail version of image in confines of divs have set it. is, using background properties have laid out in css, portion of image displays inside of circular divs. without these properties, display defaults smaller section of image [presumably pi*(50^2) area in top left] terrible resolution. i not care if entire image doesn't display inside of circles (it won't, given difference in shape), , overflow not being displayed fine. want able see figu
Read more

java - IntelliJ - No such instance method -

using intellij's 12 ultimate, i'm running following code in debugger: java import play.api.libs.json.jsvalue; public class foo { ... public jsvalue tojson() { ... } public class fooexample { ... foo foo = new foo(); system.out.println("...); //<-- breakpoint at breakpoint, right-clicked source code , picked "evaluate expression," typing in: foo.tojson() . but following error showed up: no such instance method: play.api.libs.json.jsvalue$class.com.foo.foo.tojson () am doing wrong? foo#tojson calls scala code, if matters. edit had breakpoint after instantiation of foo. downvoted it, deserved it. edit my answer no longer valid after op modified question (i.e. moved breakpoint actual is).. a breakpoint hit before line executed. in case foo has not yet been declared or instantiated (i.e. foo constructor has not yet been called). you'll need put breakpoint @ next line (or step on current line) if want evaluate foo .
Read more

identifier - Is it possible for an html5 document to have two ids? -

i wondering if html5 element can have 2 ids, this: <button id="thing1" id="thing2" name="button1" onclick="change()">click here</button> or this: <button id="thing1, thing2" name="button1" onclick="change()">click here</button> also, if 1 of ids said 1 thing, , other id said thing, 1 used? tl;dr : no. in versions of html, each element can have 1 id. see the spec : the id attribute specifies element's unique identifier (id). [dom] the value must unique amongst ids in element's home subtree , must contain @ least 1 character. the value must not contain space characters. note: there no other restrictions on form id can take; in particular, ids can consist of digits, start digit, start underscore, consist of punctuation , etc. note: element's unique identifier can used variety of purposes, notably way link specific parts of document usin
Read more