Page - 162 - in Document Image Processing

J. Imaging 2018,4, 39 scripts likeManipuri,GujaratiandUrdu, thusprovingtobe themodel tobeusedwhere thesescripts arewidelyused. Inorder tounderstandwhytheresults fromtheElliptical featuresetcombinesowellwith the twoother featuresets, correlationanalysis isperformedontheconfidencescoreoutputs. Spearman rankcorrelation isdoneontherank level informationprovidedbytheclassifiers toarriveatmean values for themeasureof thecorrelation.HOGandMLGshowanindexof0.619which isalmost the doubleof thescoresobtainedbycomparingtheElliptical featureswith these two.Withvaluesof0.32 and0.27, the lowcorrelation index isan indicationofbetterpossibilities for thecombinationprocesses. Thus, complementary information isprovidedbytheoutputofElliptical featuresetwhichhelps in the improvement theoverall combinedaccuracy. Secondaryclassifiersareapplied to learn thepatterns fromtheprimaryclassifieroutputsand developawaytocombinethem.Theconfidencescoresfromthethreesourcesareconcatenatedtoform alarger trainingsetwith its correct label. This set is thenewfeaturesetwhichundergoesclassification usingwell-knownalgorithms.Classifiers likek-NN,LogisticRegression,MLPandRandomForestare appliedtoreportfinal resultswhichare tabulated inTables14–17respectively. Theresultsarereported after3-foldcrossvalidationandtuningof theparameters involved. Thisprocess iscomputationally costlyandtakesaprocessingstepalongwithmuchhighercomplexitybut is compensatedbythehigh accuracyresults thatareobtained. 3-NNprovidesanaccuracyof98.30%,RandomForest classified 98.33%of the7200samplescorrectlyandLogisticRegressionattained98.48%accuracy.UsingMLP againas thesecondaryclassifier,98.36%accuracy isobtained.Devanagari is themostconfusedscript in all thecasesbutstillhasaccuracyover95%.Theotherscriptsarepredictedtoalmostcertainty. Table4.ClassificationresultsaftercombinationusingMajorityvotingprocedure. Class Class A B C D E F G H I J K L A 534 2 3 10 3 2 16 7 13 0 5 5 B 1 590 0 3 0 0 0 0 0 0 0 6 C 0 0 597 0 0 2 0 0 1 0 0 0 D 2 3 0 590 0 0 0 0 0 0 2 3 E 0 1 0 0 591 2 0 0 0 1 0 5 F 12 0 5 0 13 561 1 7 1 0 0 0 G 2 0 6 4 5 1 554 14 7 3 4 0 H 4 0 4 3 1 6 0 567 7 2 5 1 I 9 1 1 2 0 1 7 2 572 0 5 0 J 0 0 1 0 2 0 0 3 0 594 0 0 K 4 0 2 5 4 1 10 4 1 0 567 2 L 0 2 10 3 6 2 1 1 1 5 3 566 Table5.ClassificationresultsaftercombinationusingBordacountprocedurewithoutweight. Class Class A B C D E F G H I J K L A 567 0 5 7 0 0 5 4 7 0 3 2 B 16 580 0 4 0 0 0 0 0 0 0 0 C 1 0 586 0 0 5 0 4 3 0 0 1 D 25 1 0 572 0 0 0 0 0 0 1 1 E 6 0 0 0 466 108 0 11 1 0 0 8 F 16 0 2 0 7 571 0 1 1 1 0 1 G 25 0 4 2 0 2 548 3 1 0 15 0 H 30 0 5 0 0 21 0 533 6 0 1 4 I 39 0 2 1 0 2 6 6 540 0 4 0 J 0 0 2 0 2 0 0 7 0 589 0 0 K 5 0 0 3 0 0 12 0 1 0 579 0 L 4 0 10 2 2 4 0 1 3 3 3 568 162