Deformable deep networks for instance segmentation of overlapping multi page handwritten documents

Abstract Script identification at character level in handwritten documents is a challenging task for Gurumukhi and Latin scripts due to the presence of slightly similar, quite similar or at times confusing character pairs. Hence, it is found to be inadequate to use single feature set or just traditional feature sets and classifier in processing the handwritten documents. Due to the evolution of deep learning, the importance of traditional feature extraction approaches is somewhere neglected which is considered in this paper. This paper investigates machine learning and deep learning ensemble approaches at feature extraction and classification level for script identification. The approach here is: i. combining traditional and deep learning based features ii. evaluating various ensemble approaches using individual and combined feature sets to perform script identification iii. evaluating the pre-trained deep networks using transfer learning for script identification ’iv. finding the best combination of feature set and classifiers for script identification. Three different kinds of traditional features like Gabor filter, Gray Level Co-Occurrence Matrix (GLCM), Histograms of Oriented Gradiants (HOG) are employed. For deep learning pretrained deep networks like VGG19, ResNet50 and LeNet5 have been used as feature extractor. These individual and combined features are trained using classifiers like Support Vector Machines (SVM) , K nearest neighbor (KNN), Random Forest (rf) etc. Further many ensemble approaches like Voting,Boosting and Bagging are evaluated for script classification. Exhaustive experimental work resulted into the highest accuracy of 98.82% with features extracted from ResNet50 using transfer learning and bagging based ensemble classifier which is higher as compared to previously reported work.

Download Full-text

Unknown Documents of 1734 About Swedish Citizens Taken Into Serfdom

SUMY HISTORICAL AND ARCHIVAL JOURNAL ◽

10.21272/shaj.2019.i32.p.12 ◽

2019 ◽

pp. 12-19

Author(s):

Serhii I. Degtyarev ◽

Violetta S. Molchanova

Keyword(s):

Social History ◽

Russian Empire ◽

District Court ◽

Peace Agreement ◽

The Caucasus ◽

Handwritten Documents ◽

Ancient Times ◽

Historical Phenomenon ◽

State Archives ◽

The Russian Empire

This work is devoted to the publication and analysis of two previously unknown handwritten documents of 1734. These documents contain information on several persons of Swedish nationality, which were illegally taken out by the Russian nobleman I. Popov during the Northern War from the territory of Sweden. Materials are stored in the State Archives of the Sumy region. They are part of the archival case of Okhtyrka District Court, but they are not thematically connected with it. These documents were once part of a much larger complex of materials. They refer to the request of former Swedish nationals to release them from serfdom from the Belgorod and Kursk landlords Popov and Dolgintsev. The further fate of these people remained unknown. But it is known that they were mistreated by their masters. Russian legislation at the time prohibited such treatment of persons of Swedish nationality. This was discussed in terms of the peace agreement Nishtadskoyi 1721. The two documents revealed illustrate the episodes of the lives of several foreigners who were captured. The analyzed materials give an opportunity to look at a historical phenomenon like a serfdom in the territory of the Russian Empire under a new angle. They allow us to study one of the ways to replenish the serfs. Documents can also be used as a source for the study of some aspects of social history, in biographical studies. The authors noted that the conversion to the property of the enslaved people of other nationalities was a very common practice in the XVII-XIX centuries. This source of replenishment of the dependent population groups were popular in many nations in Europe, Asia and Africa since ancient times. For example, in the Crimean Khanate, Turkey, Italy, Egypt, the nations of the Caucasus and many others. Кeywords: Sweden, Russian Empire, historical source, documents, Russo-Swedish War, Nistadt Treaty, Viborg, Swedish citizens, enslavement, serfdom.

Download Full-text