Seite - 194 - in Document Image Processing

J. Imaging 2018,4, 32 Table3.Distributionof letters in theAcTiV-Rdataset. InArabic InArabicCharacter Label #of Occurrence I B M E CharacterLabel #of Occurrence I B M E Alif 28,433 - - HamzaAboveAlif 1653 - - Baa 7417 HamzaUnderAlif 1049 - - Taaa 8948 TildAboveAlif 87 - - Thaa 851 HamzaAboveAlifBroken 1022 - - Jiim 3270 HamzaAboveWaaw 268 - - Haaa 3976 LaamHamzaAboveAlif 925 - - Xaa 1345 LaamHamzaUnderAlif 563 - - Daal 6656 - - LaamTildAboveAlif 63 - - Thaal 459 - - Space 31,458 Raa 11,460 - - Digit_0 606 0 Zaay 1478 - - Digit_1 655 1 Siin 6348 ! " # $ Digit_2 475 2 Shiin 2353 ! " # $ Digit_3 276 3 Saad 2123 % & ' ( Digit_4 306 4 Daad 1085 % & ' ( Digit_5 248 5 Thaaa 2184 ) * + , Digit_6 203 6 Taa 481 ) * + , Digit_7 138 7 Ayn 5989 - . / 0 Digit_8 148 8 Ghayn 890 - . / 0 Digit_9 116 9 Faa 4942 1 2 3 4 Point 313 . Gaaf 4443 5 2 3 6 Colon 424 : Kaaf 2999 7 8 9 : Comma 118 ; Laam 18,868 < = > ? Slash 76 / Miim 11,907 @ A B C Percent 101 % Nuun 10,027 D E QuestionMark 8 ? Haa 2608 F G H I ExclamationMark 12 ! Waaw 10,614 - - Quote 445 “” Yaa 18,153 J J J J Hyphen 457 – AlifBroken 1211 - - ParenthesisO 30 ) Hamza 696 ParenthesisC 29 ( TaaaClosed 7239 F - - I Bar 13 | LaamAlif 1916 - - Overall 259,192 Figure5. Typicalvideo frames fromAcTiV-Ddataset. Fromleft to right: ExamplesofRussiaToday Arabic,France24Arabe,TunisiaNat1 (ElWataniya1)andAljazeeraHDframes. 194