Professional Documents
Culture Documents
fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/ACCESS.2017.2703155, IEEE Access
IEEE Access-2017-01632 1
2169-3536 (c) 2016 IEEE. Translations and content mining are permitted for academic research only. Personal use is also permitted, but republication/redistribution requires IEEE permission. See
http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/ACCESS.2017.2703155, IEEE Access
IEEE Access-2017-01632 2
2169-3536 (c) 2016 IEEE. Translations and content mining are permitted for academic research only. Personal use is also permitted, but republication/redistribution requires IEEE permission. See
http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/ACCESS.2017.2703155, IEEE Access
IEEE Access-2017-01632 3
This strategy takes into consideration the dots/diacritics information. Relying more on zonal and heuristics
even above zone 1. A recent study [9] also used zone based information for line and ligature segmentation becomes very
method [8] with dilating document images before much context and font dependent. Lines are segmented with
segmentation for getting more correct size of zones, which perfection [28], [29], but misplaced dots/diacritics are not
resulted in 98.7% line segmentation accuracy. This method allocated to their respective lines. Thus, our motivation is to
confronts difficulties for segmenting merged text lines. reduce the complex mechanism of zoning and their merging
decisions for line segmentation. In our approach, we rely
2) Ligature Segmentation
mainly on baseline information with three zonal method for
A ligature is composed of one or more characters joined
dots/diacritics allocation, and quantitative values (width,
together [8], and the combination of one or more ligatures
height, centroids, and overlapping) for ligature segmentation.
form words in Arabic script (Like Urdu language). A ligature
is comprised of two parts; the primary component is made up C. Contribution
of main stroke, while diacritics are nominated as secondary In this paper for Urdu Nastaleeq text, we present two
strokes as presented in Fig. 6. algorithms for line and ligature segmentation. These
algorithms can also be applied to other Nastaleeq based
scripted languages, such as Arabic, Persian, Pashto, and
Sindhi etc. Specifically, our proposed line segmentation
algorithm depends solely on horizontal projection, split
positions (propose in this paper and named as Curved Line
Split (CLS) algorithm) and base line information. In proposed
approach, the allocation of dots/diacritics to proper text line is
properly addressed by employing three-zonal method. For
ligature segmentation, the propose algorithm employs width,
height, centroids, and overlapping quantitative information as
opposed to heuristics [8], [9]. The accuracies achieved by
Fig. 6. Primary and secondary components of two ligatures
both the propose line and ligature segmentation algorithms are
Considerable efforts were devoted for line segmentation 99.33% and 99.08%, respectively, which are better than the
into ligatures like vertical histogram method [20] and [27], previous contributions [8], [9], [29].
horizontal projection method [4], and bounding box method The remaining paper is organized as follows. In Section 2,
[8], [9], [18], [28], [31]. line segmentation algorithm is discussed. While, in Section 3
Individual ligatures extracted by vertical histogram of text we discuss the ligature extraction algorithm. In Section 4,
line in Naskh as described in [20] and [27], but this method is results are analyzed for evaluating the performance of propose
not applicable for Nastaleeq because of inter-ligature algorithms. Finally, Section 5 includes the concluding
overlapping. In another work [4], vertical projection was remarks.
applied for getting ligatures and character on extracted lines
by horizontal projection method. Ligatures were also II. LINE SEGMENTATION
segmented by connected component method supplemented This section details about segmentation algorithm of a page
with baseline and centroid information, while achieving 94% into constituents text lines, which employs global horizontal
segmentation accuracy [31]. Connected components were projection method augmented with propose CLS algorithm.
extracted by bounding box methods from text lines [18] and Additionally, it contains steps for conflicts resolution
[28], and then secondary components were allocated to
regarding components crossing split lines based upon baseline
primary components based on the detected baseline. Text
information.
lines were segmented into connected components that were
categorized into 6 types [8], and then secondary components A. Curved Line Split Algorithm
were merged together with primary components to get the The propose CLS algorithm can even find out a curved (not
resultant ligatures based upon certain heuristics, thus, straight) demarcation line between every two consecutive text
achieving 99.02% accuracy. Another study [9] first segmented lines, when there is no straight demarcation line. The basic
lines into connected components, and then allocated idea of CLS algorithm is as follows:
secondary components to primary components with 92.5%
accuracy. Basic Idea: Algorithm starts a demarcation line from
middle of baselines of consecutive text lines. It traverses a
B. Motivation path from left to right on horizontal path, if it finds text pixels
The most recent line and ligature segmenting algorithms it moves up or down vertically until it finds a text pixel or
[8], [9] proposed the use of 4 zones for line segmentation, and baseline, and then from half of the vertical movement it starts
6 categories of connected components for constructing a final again horizontal traversal in between two lines.
ligature. They did not utilize the baseline information that is
proved to be very much decisive in our propose line The procedure of CLS is presented in Figs. 7 and 8. The
segmentation algorithm. Also, ligature segmentation [8] relies CLS algorithm detail is given as Algorithm 1.
on 6 heuristics, among which only one heuristic used baseline
2169-3536 (c) 2016 IEEE. Translations and content mining are permitted for academic research only. Personal use is also permitted, but republication/redistribution requires IEEE permission. See
http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/ACCESS.2017.2703155, IEEE Access
IEEE Access-2017-01632 4
2169-3536 (c) 2016 IEEE. Translations and content mining are permitted for academic research only. Personal use is also permitted, but republication/redistribution requires IEEE permission. See
http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/ACCESS.2017.2703155, IEEE Access
IEEE Access-2017-01632 5
peaks until we get a local highest valued peak. Then the value For visualization of this concept, four cases are briefed as
of next coming peaks gradually decrease until it reaches follows:
another trough. Such local highest valued peaks are termed as Case 1: Crossing component touching one base line
Line-Peaks, which are accompanied by gradual lower valued Move the crossing component to the text line where it is
peaks on both sides representing one line. Line-Peaks touching the baseline.
represent center/base with maximum pixel intensity of text Case 2: Crossing component not touching any base line
lines in a page image. a) Here, find the top row, bottom row, left most column and
right most column numbers for under consideration crossing
Line − Peaks: ∀𝑥 𝑓𝑖𝑛𝑑 𝑎𝑙𝑙 𝑥 ′′ 𝑐 𝑠. 𝑡 component.
𝑥𝑖−1 < 𝑥 ′′ 𝑐 > 𝑥𝑖+1 (4) b) Next, start search by moving above from the top row and
below from the bottom row while checking every pixel from
2) Average Text Line Height left most column to right most column for finding the nearest
Next, we find the average number of pixels’ rows per text text pixel of any other component.
line. c) Move this crossing component to the corresponding side
|𝑥|
Average Text Line(TL)Height: 𝐴𝑣𝑔𝐻 (𝑇𝐿) = ′′ , where text pixel is found earlier. In case of tie, move
|𝑥 |
component to the side where search found component is
Where, 𝐴𝑣𝑔𝐻 (𝑇𝐿) 𝑖𝑠 Average Text Line(TL)Height, touching base line. If both are not touching the baseline, then
|𝑥| 𝑟𝑒𝑝𝑟𝑒𝑠𝑒𝑛𝑡𝑠 𝑡𝑜𝑡𝑎𝑙 𝑟𝑜𝑤𝑠 𝑜𝑓 𝑝𝑖𝑥𝑒𝑙𝑠, perform step a, b and c for both newly found components, but
search is performed in the same direction where these new
𝑥 ′′ 𝑟𝑒𝑝𝑟𝑒𝑠𝑒𝑛𝑡𝑠 𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝐿𝑖𝑛𝑒_𝑃𝑒𝑎𝑘𝑠 , (5) components are found.
3) False Text Lines Case 3: Connected component touching neither base lines
Once again scan the peaks, if lowest troughs of some Line- nor split line
Peak come before ¾th of the average text line height then that Some diacritics/dots might neither crossing any base nor
line is considered as false Line-Peak and excluded from total split line. Such components might be segmented in wrong text
Line-Peaks. line. Therefore, scan every text line for any text component
∀𝑥 ′′ 𝑐 𝑓𝑖𝑛𝑑: 𝑎 = |𝑥 ′′ 𝑐𝑖−1 − 𝑥 ′′ 𝑐𝑖−2 |, measuring approximately ¼th height of each line. These areas
are referred as upper and lower diacritics/dots zones as
b= |𝑥 ′′ 𝑐𝑖 − 𝑥 ′′ 𝑐𝑖−1 |, presented in Fig. 9.
There may be possibilities of attaching diacritics to wrong
c= |𝑥 ′′ 𝑐𝑖+1 − 𝑥 ′′ 𝑐𝑖 |,
lines, most probably in upper and lower diacritics/dots zones
3
𝑖𝑓 𝑐 < 𝐴𝑣𝑔𝐻 (𝑇𝐿) ∧ 𝑎 < 𝑐, 𝑎 = 𝑎 + 𝑏, of every text line except first and last ones. First text line may
4
𝐸𝑙𝑖𝑚𝑖𝑛𝑎𝑡𝑒 𝑥 ′′ 𝑐𝑖 𝑓𝑟𝑜𝑚 Line − Peaks (6) only have lower diacritics/dots zone, while last one may have
only upper diacritics/dots zone.
4) Missed Text Lines In these portions, if some component is found then it is
All Line-Peak values must be greater than the highest moved to the line where it has minimum distance from the
trough value; otherwise the text lines with Line-Peaks less neighboring component. This step can easily be performed as
than the highest trough will be missed. After conforming step follows:
3, we hypothetically increase the value of all such Line-Peaks
more than the highest trough value. In this way, we fix the
problem of missed lines.
Highest Trough ∶ 𝑚𝑎 𝑥 𝑥𝑇′ = H 𝑇 ∀𝑥 𝑓𝑖𝑛𝑑 𝑥𝑖′ 𝑠. 𝑡
′
𝑥𝑖−1 > 𝑥𝑖′ < 𝑥𝑖+1
′
] (7)
′′
Missed Text Lines ∶ ∀𝑥 𝑐𝑖 < H𝑇 ,
𝑥 ′′ 𝑐𝑖 = 𝑥 ′′ 𝑐𝑖 + (|H 𝑇 − 𝑥 ′′ 𝑐𝑖 |) (8)
5) Segment Page Image into Text Lines Fig. 9. Upper and lower diacritics/dots zones scanning
Segment page image into text lines: Pass initial split row
numbers to CLS algorithm. The initial candidate split row a) For every component found in upper or lower diacritics/dots
numbers are the lowest trough values in between every two zone, where:
Line-Peaks. Convert the text page into connected components i. The component must not touch baseline.
format, so that every component is represented by a different ii. The component must not be greater in size than its
number. Then split this page according to the coordinates for respective zone.
demarcation of lines returned by CLS algorithm. b) Find the top and bottom row of each component. Also, find
6) Assignment of Components the left and right most column numbers of its maximum
Assignment of misplaced and crossing components to span/width.
proper text line may easily be done with the information of c) Start searching above and below from the component’s top
baseline of every text line. Crossing components are those coordinates and bottom coordinates, respectively, for
connected components that are crossing the demarcation line. finding the nearest text pixel.
2169-3536 (c) 2016 IEEE. Translations and content mining are permitted for academic research only. Personal use is also permitted, but republication/redistribution requires IEEE permission. See
http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/ACCESS.2017.2703155, IEEE Access
IEEE Access-2017-01632 6
Fig. 10. Sample Urdu Nastaleeq pages (a) from [28], (b) from
[29], (c) from [8], (d) from [9]
Fig. 15. Red encircled components of previous step are resolved in next step, encircled green, (a) to (b) crossing component touching one base line,
2169-3536 (c) crossing
(b) to (c) 2016 IEEE. Translations
component anddoes
contentnot
mining are permitted
touching for academic
any base research
line, (c) only.
to (d) Personal use isneither
component also permitted,
touchingbut republication/redistribution
base lines nor split lines, requires
(d)IEEE
to permission. See
(e) component
touching 2 baselines. http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/ACCESS.2017.2703155, IEEE Access
IEEE Access-2017-01632 7
B. Extract Information
Fig. 16. Example of crossing components, touching both base
lines
2169-3536 (c) 2016 IEEE. Translations and content mining are permitted for academic research only. Personal use is also permitted, but republication/redistribution requires IEEE permission. See
http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/ACCESS.2017.2703155, IEEE Access
IEEE Access-2017-01632 8
(b)
(c)
(d)
(e)
(f)
Fig. 18. Total 27, 12 and 21resultant ligatures of above
displayed sentences taken from UPTI data set, [28] and [29],
respectively. First sentence of UPTI dataset and its segmented
ligatures are represented in (a) and (b). A sentence of a text
page given in [28] and its segmented ligatures are represented
in (c) and (d), respectively. Similarly, a sentence of a text page
given in [29] and its segmented ligatures are represented in (e)
and (f), respectively.
2169-3536 (c) 2016 IEEE. Translations and content mining are permitted for academic research only. Personal use is also permitted, but republication/redistribution requires IEEE permission. See
http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/ACCESS.2017.2703155, IEEE Access
IEEE Access-2017-01632 9
2169-3536 (c) 2016 IEEE. Translations and content mining are permitted for academic research only. Personal use is also permitted, but republication/redistribution requires IEEE permission. See
http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/ACCESS.2017.2703155, IEEE Access
IEEE Access-2017-01632 10
Fig. 22. Urdu Nastaleeq pages taken from [29] and segmented by propose line segmentation algorithm
The highest intensity values at split positions pose C. Results of Propose Ligature Segmentation Algorithm
maximum obstacles in drawing boundary, whereas, a UPTI (Urdu Printed Text Images) dataset [18] is used for
straight line can demarcate a boundary in between two text checking the segmentation accuracy of the proposed
lines in case of minimum intensity value (i.e. 0). The segmentation algorithms. This dataset contains images of
propose line segmentation algorithm is applied on these 10063 sentences. Four degradation techniques that were
two complex pages, having extracted text lines accurately proposed in [32] are applied, which includes jitter, elastic
with only one diacritic out of place encircled red as shown elongation, threshold and sensitivity. Every degradation
in Fig. 22(b). Furthermore, some encircled dots/diacritics technique is applied with four different parameter values.
are misplaced by algorithm of [29] depicted in Fig.22 (e). Overall, UPTI contains 12 degradation versions of original
10063 sentence images.
TABLE 2 The sentence images of UPTI dataset from un-degraded
INITIAL SPLIT ROW NUMBERS version is segmented into ligatures as presented in Table 3.
Table 2(a) Initial split for Fig. Table 2(b) Initial split for Fig. There are total of 189584 ligatures as per split of ground
22(a) 22(c) truths of UPTI dataset. Accuracy of the implemented
Lines Intensity Split Lines Intensity Split
algorithms is even better than the best ligature segmentation
# Value Row # Value Row
Number Number
accuracy reported in [8] i.e. 99.02%.
1&2 42 93 1&2 3 61
2&3 83 164 2&3 19 120
3&4 82 248 3&4 27 178
4&5 79 325 4&5 24 224 Fig. 24. Wrongly connected components of ligatures
5&6 45 390 5&6 14 273
UPTI dataset contains incorrectly merged ligatures that
6&7 36 483 6&7 24 328
badly affect the accuracy, because the corresponding ligatures
7&8 10 383 in ground truths are not properly connected. Therefore, count
8&9 17 433 of image ligatures becomes less as compared to ligatures
segmented from text lines of ground truth, resulting in low
The propose line segmentation algorithm is also applied on accuracy. Examples of these incorrectly connected dots,
an Arabic language page provided in [29]. Our line diacritics and ligatures are depicted in Fig. 24.
segmentation algorithm misplaced only one diacritic, while
line segmentation algorithm of [29] misplaced 5 diacritics as First and second part of Fig. 24
shown in Figs. 23(b) and 23(c), respectively. Lines are represent two ligatures connected incorrectly. Fourth part
segmented by both algorithms with 100% accuracy.
consists of all secondary components of third part
Fig. 23. Arabic Nastaleeq page taken from [29] and segmented by propose line segmentation algorithm
2169-3536 (c) 2016 IEEE. Translations and content mining are permitted for academic research only. Personal use is also permitted, but republication/redistribution requires IEEE permission. See
http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/ACCESS.2017.2703155, IEEE Access
IEEE Access-2017-01632 11
TABLE 3
STATISTICS ABOUT COMPONENTS OF UPTI DATA SET
Primary Components Secondary Components Total Ligatures Segmentation
Components Segmented Accuracy
Touching Not Touching Not
Base Line Touching Base Touching
Base Line Line Base Line
UPTI Data Set
175431 13754 18253 126387 333825 189185 99.80%
Undegraded
TABLE 4
STATISTICS ABOUT COMPONENTS OF FIRST SENTENCE
OF UPTI DATA SET
Primary Secondary Total Ligatures
Comp. Comp. Comp. Segmented
Touching Touching
Baseline Baseline
Yes No Yes No
21 6 4 20 51 27
2169-3536 (c) 2016 IEEE. Translations and content mining are permitted for academic research only. Personal use is also permitted, but republication/redistribution requires IEEE permission. See
http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/ACCESS.2017.2703155, IEEE Access
IEEE Access-2017-01632 12
TABLE 5
COMPARISON OF URDU LANGUAGE LIGATURE SEGMENTATION ALGORITHMS
Ligature Primary Components Secondary Components
segmentation No With No With Total
Algorithm Total confusion confusion Total confusion confusion Comp. Ligatures Accuracy
[7] Javed et Corr./total
al 3375/3655 92%
[9] Israr et
al 6811/7364 99.49%
[8] Lehal 23555 18886 17565 12805 42441 23555 99.02%
This paper 189232 176353 12879 143260 18801 124459 332492 189232 99.80%
algorithms for line and ligature segmentation of Nastaleeq text [5] D. A. Satti and K. Saleem, “Complexities and implementation challenges
in offline Urdu Nastaleeq OCR,” in Proceedings of the Conference on
images. The proposed line segmentation algorithm places dots
Language & Technology, pp. 85–91, 2012.
and diacritics more accurately as compared to [28], [29].
Prevailing work [8], [9] relied more on zonal information and [6] S. Hussain, “Complexity of Asian writing systems: a case study of nafees
heuristics for line and ligature segmentation, respectively. In nasta’leeq for Urdu,” in Proceedings of the 12th AMIC Annual Conference
on e-Worlds: Governments, Business and Civil Society, Asian Media
our approach, we rely more on baseline for decision making
Information Center, Singapore, 2003.
as opposed to zonal information for line segmentation. Also,
proposed algorithm employs baseline, width, height, [7] S. T. Javed and S. Hussain, “Improving nastalique specific pre-recognition
centroids, and overlapping quantitative information for process for Urdu OCR,” in Multitopic Conference, 2009. INMIC 2009. IEEE
13th International, pp. 1–6, IEEE, 2009.
ligature segmentation algorithms. Furthermore, we tested our
ligature segmentation algorithms on a very large dataset with [8] G. S. Lehal, “Ligature segmentation for Urdu ocr,” in Document Analysis
improved accuracy as presented in Table 3. On average, for and Recognition (ICDAR), 2013 12th International Conference on,pp. 1130–
UPTI dataset, the propose algorithm segmented 189 thousand 1134, IEEE, 2013.
ligatures from 10063 text lines with 332 thousand connected
[9] I. U. Din, Z. Malik, I. Siddiqi, and S. Khalid, “Line and ligature
components. A total of about 142 thousand secondary segmentation in printed urdu document images,” J. Appl. Environ. Biol. Sci,
components have been successfully allocated to more than vol. 6, no. 3S, pp. 114–120, 2016.
189 thousand primary ligatures. The results of our line and
[10] I. Shamsher, Z. Ahmad, J. K. Orakzai, and A. Adnan, “Ocr for printed
ligature segmentation algorithms with accuracy of 99.33%
Urdu script using feed forward neural network,” in Proceedings of World
and 99.80% respectively, are thus, better than the precision Academy of Science, Engineering and Technology, vol. 23, pp. 172–175,
realized by the prevailing Urdu Nastaleeq segmentation 2007.
algorithms.
[11] I. K. Pathan, A. A. Ali, and R. Ramteke, “Recognition of offline
This work can be extended easily for other Nastaleeq script
handwritten isolated Urdu character,” Advances in Computational Research,
based languages like Persian, Pashto, Siraiki, and Panjabi etc. vol. 4, no. 1, 2012.
[2] S. Naz, K. Hayat, M. I. Razzak, M. W. Anwar, S. A. Madani, and S. [15] Z. Ahmad, J. K. Orakzai, I. Shamsher, and A. Adnan, “Urdu Nastaleeq
U.Khan, “The optical character recognition of urdu-like cursive scripts,” optical character recognition,” in Proceedings of world academy of science,
Pattern Recognition, vol. 47, no. 3, pp. 1229–1248, 2014. engineering and technology, vol. 26, pp. 249–252, Citeseer, 2007.
[3] G. S. Lehal and A. Rana, “Recognition of nastalique Urdu ligatures, “in [16] T. Nawaz, S. Naqvi, H. ur Rehman, and A. Faiz, “Optical character
Proceedings of the 4th International Workshop on Multilingual OCR, pp. 7, recognition system for urdu (naskh font) using pattern matching technique,”
ACM, 2013 International Journal of Image Processing (IJIP), vol. 3, no. 3, p. 92, 2009.
[4] U. Pal and A. Sarkar, “Recognition of printed Urdu script.,” in ICDAR, [17] S. Hussain, S. Ali, et al., “Nastalique segmentation-based approach for
vol. 2003, pp. 1183–1187, 2003. urdu ocr,” International Journal on Document Analysis and Recognition
(IJDAR), vol. 18, no. 4, pp. 357–374, 2015.
2169-3536 (c) 2016 IEEE. Translations and content mining are permitted for academic research only. Personal use is also permitted, but republication/redistribution requires IEEE permission. See
http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/ACCESS.2017.2703155, IEEE Access
IEEE Access-2017-01632 13
[18] N. Sabbour and F. Shafait, “A segmentation-free approach to Arabic and [27] T. M. Breuel, “Two geometric algorithms for layout analysis,” in
Urdu ocr.” in DRR, 2013. International workshop on document analysis systems, pp. 188–199,
Springer,2002.
[19] S. T. Javed and S. Hussain, “Segmentation based Urdu nastalique OCR,”
in Iberoamerican Congress on Pattern Recognition, pp. 41–49, Springer, [28] F. Shafait, A. Hasan, D. Keysers, and T. M. Breuel, “Layout analysis of
2013. Urdu document images,” in 10th IEEE Int. Multitopic Conference,
INMIC’06, Islamabad, Pakistan, Dec. 2006
[20] A. Ul-Hasan, S. B. Ahmed, F. Rashid, F. Shafait, and T. M. Breuel,
“Offline printed urdu nastaleeq script recognition with bidirectional lstm [29] S. S. Bukhari, F. Shafait, and T. M. Breuel, “High performance layout
networks,” in Document Analysis and Recognition (ICDAR), 2013 12th analysis of arabic and urdu document images,” in Document Analysis and
International Conference on, pp. 1061–1065, IEEE, 2013. Recognition (ICDAR), 2011 International Conference on, pp. 1275–1279,
IEEE, 2011.
[21] S. Naz, A. I. Umar, R. Ahmad, S. B. Ahmed, S. H. Shirazi, and M. I.
Razzak, “Urdu nastaliq text recognition system based on multidimensional [30] S. A. Husain, “A multi-tier holistic approach for urdu nastaliq
recurrent neural network and statistical features,” Neural Computing and recognition,” in Multi Topic Conference, 2002. Abstracts. INMIC 2002.
Applications, pp. 1–13, 2015. International, pp. 84, IEEE, 2002.
[22] S. Naz, A. I. Umar, R. Ahmad, S. B. Ahmed, S. H. Shirazi, I. Siddiqi, [31] S. T. Javed, S. Hussain, A. Maqbool, S. Asloob, S. Jamil, and H. Moin,
and M. I. Razzak, “Offline cursive urdu-nastaliq script recognition using “Segmentation free nastalique urdu ocr,” World Academy of Science,
multidimensional recurrent neural networks,” Neurocomputing, vol. 177,pp. Engineering and Technology, vol. 46, pp. 456–461, 2010.
228–241, 2016.
[32] H. S. Baird, “Document image defect models,” in Structured Document
[23] S. Naz, S. B. Ahmed, R. Ahmad, and M. I. Razzak, “Zoning features Image Analysis, pp. 546–556, Springer, 1992.
and 2DLSTM for Urdu text-line recognition,”. vol. 96, pp. 16-22, 2016.
[33] S. S. Bukhari, F. Shafait, and T. M. Breuel, “Towards generic text-line
[24] I. Ahmad, X. Wang, R. Li, and S. Rasheed, “Offline urdu Nastaleeq extraction,” in Document Analysis and Recognition (ICDAR), 2013 12 th
optical character recognition based on stacked denoising autoencoder,” China International Conference on, pp. 748–752, IEEE, 2013.
Communications, vol. 14, no. 1, pp. 146–157, 2017.
[34] S. S. Bukhari, F. Shafait, and T. M. Breuel, “Text-line extraction using a
[25] H. Malik and M. A. Fahiem, “Segmentation of printed urdu scripts using convolution of isotropic gaussian filter with a set of line filters,” in Document
structural features,” in Visualisation, 2009. VIZ’09. Second International Analysis and Recognition (ICDAR), 2011 Inte national Conference on, pp.
Confe rence in, pp. 191–195, IEEE, 2009. 579–583, IEEE, 2011.
[26] K. S. Kumar, S. Kumar, and C. Jawahar, “On segmentation of [35] S. S. Bukhari, T. M. Breuel, and F. Shafait, “Textline information
documents in complex scripts,” in Document Analysis and Recognition, extraction from grayscale camera-captured document images,” in Image
2007. ICDAR 2007. Ninth International Conference on, vol. 2, pp. 1243– Processing (ICIP), 2009 16th IEEE International Conference on, pp. 2013–
1247, IEEE, 2007. 2016, IEEE, 2009.
2169-3536 (c) 2016 IEEE. Translations and content mining are permitted for academic research only. Personal use is also permitted, but republication/redistribution requires IEEE permission. See
http://www.ieee.org/publications_standards/publications/rights/index.html for more information.