scholarly journals Bangla/English Script Identification Based on Analysis of Connected Component Profiles

2006 ◽  
pp. 243-254 ◽  
Author(s):  
Lijun Zhou ◽  
Yue Lu ◽  
Chew Lim Tan
Keyword(s):  
Connected Component ◽  
Script Identification

Metode Diplomatik dalam Mengidentifikasi Kandungan Isi Naskah Gulungan Berbahan Daluang Koleksi Candi Cangkuang

Panggung ◽  
2012 ◽  
Vol 22 (4) ◽  
Author(s):  
Tedi Permadi
Keyword(s):  
Direct Observation ◽  
Relevant Literature ◽  
Script Identification ◽  
Identification Techniques

ABSTRACTThis paper presents the results of the identification of rolled manuscripts made of daluang using diplomatic method. This method aims at getting the authenticity of the script based on the information that accompanies the text with the internal evidence contained in the manuscript. In terms of script identification techniques, diplomatic method utilizes direct observation techniques, assisted by other descriptions of contemporary manuscript as an evidence and support of the relevant literature. The use of diplomatic method in identifying rolled manuscripts produces the characteristics of the material, the literacy/language used in the text, and the editorial lapses contained in the text, but the identity of the author or the copyist and the time of the writing or copying manuscripts could not be found.Keywords: Manuscript identification, daluang, diplomatic method ABSTRAKTulisan ini menyajikan hasil identifikasi naskah gulungan berbahan daluang dengan menggunakan metode diplomatik. Metode diplomatik bertujuan untuk mendapatkan keaslian naskah berdasarkan informasi yang ada di dalam teks dengan bukti internal yang terkandung dalam naskah tersebut. Dalam hal teknik identifikasi naskah, metode diplomatik memanfaatkan teknik observasi langsung, dibantu dengan deskripsi dari naskah kontemporer lain sebagai bukti dan pendukung literatur yang relevan. Penggunaan metode diplomatik dalam mengidentifikasi naskah gulungan menghasilkan karakteristik material, huruf/bahasa yang digunakan dalam teks, dan penyimpangan editorial yang terkandung dalam teks, tetapi tidak bisa menemukan identitas penulis atau penyalin dan waktu penulisan atau penyalinan naskah.Kata kunci: Identifikasi naskah, daluang, metode diplomatik

scholarly journals Artistic multi-script identification at character level with extreme learning machine

2020 ◽  
Vol 167 ◽  
pp. 496-505
Author(s):  
Mridul Ghosh ◽  
Himadri Mukherjee ◽  
Sk. Md. Obaidullah ◽  
K.C. Santosh ◽  
Nibaran Das ◽  
...  
Keyword(s):  
Extreme Learning Machine ◽  
Script Identification ◽  
Learning Machine

scholarly journals Non Kählerian surfaces with a cycle of rational curves

2021 ◽  
Vol 8 (1) ◽  
pp. 208-222
Author(s):  
Georges Dloussky
Keyword(s):  
Deformation Theory ◽  
Complex Surface ◽  
Rational Curves ◽  
Intersection Form ◽  
Connected Component ◽  
Hirzebruch Surface ◽  
Compact Complex Surface ◽  
A Chain

Abstract Let S be a compact complex surface in class VII0 + containing a cycle of rational curves C = ∑Dj . Let D = C + A be the maximal connected divisor containing C. If there is another connected component of curves C ′ then C ′ is a cycle of rational curves, A = 0 and S is a Inoue-Hirzebruch surface. If there is only one connected component D then each connected component Ai of A is a chain of rational curves which intersects a curve Dj of the cycle and for each curve Dj of the cycle there at most one chain which meets Dj . In other words, we do not prove the existence of curves other those of the cycle C, but if some other curves exist the maximal divisor looks like the maximal divisor of a Kato surface with perhaps missing curves. The proof of this topological result is an application of Donaldson theorem on trivialization of the intersection form and of deformation theory. We apply this result to show that a twisted logarithmic 1-form has a trivial vanishing divisor.

Mining discriminative patches for script identification in natural scene images

2021 ◽  
Vol 40 (1) ◽  
pp. 551-563
Author(s):  
Liqiong Lu ◽  
Dong Wu ◽  
Ziwei Tang ◽  
Yaohua Yi ◽  
Faliang Huang
Keyword(s):  
Neural Networks ◽  
Experimental Results ◽  
The Other ◽  
Natural Scene ◽  
Fixed Size ◽  
Script Identification ◽  
Aspect Ratios ◽  
Novel Approach ◽  
Public Datasets ◽  
Natural Scene Images

This paper focuses on script identification in natural scene images. Traditional CNNs (Convolution Neural Networks) cannot solve this problem perfectly for two reasons: one is the arbitrary aspect ratios of scene images which bring much difficulty to traditional CNNs with a fixed size image as the input. And the other is that some scripts with minor differences are easily confused because they share a subset of characters with the same shapes. We propose a novel approach combing Score CNN, Attention CNN and patches. Attention CNN is utilized to determine whether a patch is a discriminative patch and calculate the contribution weight of the discriminative patch to script identification of the whole image. Score CNN uses a discriminative patch as input and predict the score of each script type. Firstly patches with the same size are extracted from the scene images. Secondly these patches are used as inputs to Score CNN and Attention CNN to train two patch-level classifiers. Finally, the results of multiple discriminative patches extracted from the same image via the above two classifiers are fused to obtain the script type of this image. Using patches with the same size as inputs to CNN can avoid the problems caused by arbitrary aspect ratios of scene images. The trained classifiers can mine discriminative patches to accurately identify some confusing scripts. The experimental results show the good performance of our approach on four public datasets.

Sign in / Sign up

Export Citation Format

Share Document