Script Identification for Printed and Handwritten Indian Documents: An Empirical Study of Different Feature Classifier Combinations

2021 ◽  
pp. 2140011
Author(s):  
Rajneesh Rani ◽  
Renu Dhir ◽  
Deepti Kakkar ◽  
Nonita Sharma
Keyword(s):  
Feature Extraction ◽  
Empirical Study ◽  
State Of The Art ◽  
Document Processing ◽  
Classification Techniques ◽  
Handwritten Documents ◽  
Script Identification ◽  
Basic Objective ◽  
To Come ◽  
Document Page

The identification of script in a document page image is the first step for an OCR system processing multi-script documents. In this multilingual/multiscript world, document processing systems relying on the OCR that need human involvement to select the appropriate OCR package is definitely undesirable and inefficient. The development of robust and efficient methods for automatic script identification of a document is a subject of major importance for automatic document processing in a multilingual/multiscript environment. Thus, the basic objective is to come up with some intuitive methods having straightforward implementation without compromising with efficiency. The aim of this work is to evaluate state-of-the-art feature extraction and classification techniques in the field of automatic script identification of printed and handwritten documents and to propose the best combination for the same.

scholarly journals Indian Handwritten Script Identification System based on Random Forest Tree Ensembles

2019 ◽  
Vol 8 (2) ◽  
pp. 2097-2103
Keyword(s):  
Random Forest ◽  
Processing System ◽  
Forest Tree ◽  
Structural Features ◽  
Identification System ◽  
Document Processing ◽  
Script Language ◽  
Handwritten Documents ◽  
Script Identification ◽  
Random Forest Tree

The work proposal addresses to introduce a methodology for Indian unconstrained handwritten script identification by practicing distinct features and classifiers. By utilizing classifiers like RF, SVM, k-NN, and LDA for Indian script identification using statistical, geometric, and structural features. To preserve all the information present on handwritten documents such as historical, medieval, inscription, financial administration, public records, government archives, letters, land councils, various agreements, etc. in digitalize form needs textual document processing system (e.g. OCR). To build a precise and productive multi-script/language textual document processing system must have script identification. For this study use, total 1288 (line wise) samples of ten scripts use in India are collected from different persons of different gender, age, education and region (rural or urban). After successful training and testing, 81.8% and 0.252 accuracies and the OOB error rate are achieved by Random Forest respectively. And 77.8%, 73.5%, and 65.5% accuracy is achieved in SVM, k-NN and LDA classifiers respectively

scholarly journals Script Identification in Handwritten Documents for Gurumukhi-Latin Script Using Transfer Learning with Deep and Shallow Classifiers

2021 ◽  
Author(s):  
Sukhandeep Kaur ◽  
Seema Bawa ◽  
Ravinder Kumar
Keyword(s):  
Feature Extraction ◽  
Deep Learning ◽  
Transfer Learning ◽  
Gabor Filter ◽  
Support Vector ◽  
K Nearest Neighbor ◽  
Handwritten Documents ◽  
Feature Sets ◽  
Script Identification ◽  
Deep Networks

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.

scholarly journals A Sign of Things to Come: Predicting the Perception of Above-the-Fold Time in Web Browsing

2021 ◽  
Vol 13 (2) ◽  
pp. 50
Author(s):  
Hamed Z. Jahromi ◽  
Declan Delaney ◽  
Andrew Hines
Keyword(s):  
Web Application ◽  
Web Applications ◽  
State Of The Art ◽  
Influencing Factor ◽  
The State ◽  
Experimental Result ◽  
Web Page ◽  
To Come ◽  
Web Quality ◽  
The Web

Content is a key influencing factor in Web Quality of Experience (QoE) estimation. A web user’s satisfaction can be influenced by how long it takes to render and visualize the visible parts of the web page in the browser. This is referred to as the Above-the-fold (ATF) time. SpeedIndex (SI) has been widely used to estimate perceived web page loading speed of ATF content and a proxy metric for Web QoE estimation. Web application developers have been actively introducing innovative interactive features, such as animated and multimedia content, aiming to capture the users’ attention and improve the functionality and utility of the web applications. However, the literature shows that, for the websites with animated content, the estimated ATF time using the state-of-the-art metrics may not accurately match completed ATF time as perceived by users. This study introduces a new metric, Plausibly Complete Time (PCT), that estimates ATF time for a user’s perception of websites with and without animations. PCT can be integrated with SI and web QoE models. The accuracy of the proposed metric is evaluated based on two publicly available datasets. The proposed metric holds a high positive Spearman’s correlation (rs=0.89) with the Perceived ATF reported by the users for websites with and without animated content. This study demonstrates that using PCT as a KPI in QoE estimation models can improve the robustness of QoE estimation in comparison to using the state-of-the-art ATF time metric. Furthermore, experimental result showed that the estimation of SI using PCT improves the robustness of SI for websites with animated content. The PCT estimation allows web application designers to identify where poor design has significantly increased ATF time and refactor their implementation before it impacts end-user experience.

scholarly journals Image Classification for the Automatic Feature Extraction in Human Worn Fashion Data

Mathematics ◽  
2021 ◽  
Vol 9 (6) ◽  
pp. 624
Author(s):  
Stefan Rohrmanstorfer ◽  
Mikhail Komarov ◽  
Felix Mödritscher
Keyword(s):  
Neural Networks ◽  
Feature Extraction ◽  
Image Classification ◽  
Convolutional Neural Networks ◽  
Data Augmentation ◽  
State Of The Art ◽  
Image Data ◽  
Classification Model ◽  
Upper Body ◽  
Automatic Feature Extraction

With the always increasing amount of image data, it has become a necessity to automatically look for and process information in these images. As fashion is captured in images, the fashion sector provides the perfect foundation to be supported by the integration of a service or application that is built on an image classification model. In this article, the state of the art for image classification is analyzed and discussed. Based on the elaborated knowledge, four different approaches will be implemented to successfully extract features out of fashion data. For this purpose, a human-worn fashion dataset with 2567 images was created, but it was significantly enlarged by the performed image operations. The results show that convolutional neural networks are the undisputed standard for classifying images, and that TensorFlow is the best library to build them. Moreover, through the introduction of dropout layers, data augmentation and transfer learning, model overfitting was successfully prevented, and it was possible to incrementally improve the validation accuracy of the created dataset from an initial 69% to a final validation accuracy of 84%. More distinct apparel like trousers, shoes and hats were better classified than other upper body clothes.

scholarly journals A Multi-Branch Feature Fusion Strategy Based on an Attention Mechanism for Remote Sensing Image Scene Classification

2021 ◽  
Vol 13 (10) ◽  
pp. 1950
Author(s):  
Cuiping Shi ◽  
Xin Zhao ◽  
Liguo Wang
Keyword(s):  
Remote Sensing ◽  
Feature Extraction ◽  
Classification Accuracy ◽  
Feature Fusion ◽  
State Of The Art ◽  
Rapid Development ◽  
Remote Sensing Image ◽  
Classification Performance ◽  
Attention Mechanism ◽  
Scene Classification

In recent years, with the rapid development of computer vision, increasing attention has been paid to remote sensing image scene classification. To improve the classification performance, many studies have increased the depth of convolutional neural networks (CNNs) and expanded the width of the network to extract more deep features, thereby increasing the complexity of the model. To solve this problem, in this paper, we propose a lightweight convolutional neural network based on attention-oriented multi-branch feature fusion (AMB-CNN) for remote sensing image scene classification. Firstly, we propose two convolution combination modules for feature extraction, through which the deep features of images can be fully extracted with multi convolution cooperation. Then, the weights of the feature are calculated, and the extracted deep features are sent to the attention mechanism for further feature extraction. Next, all of the extracted features are fused by multiple branches. Finally, depth separable convolution and asymmetric convolution are implemented to greatly reduce the number of parameters. The experimental results show that, compared with some state-of-the-art methods, the proposed method still has a great advantage in classification accuracy with very few parameters.

IT for Peace? Fighting Against Terrorism in Social Media – An Explorative Twitter Study

i-com ◽  
2017 ◽  
Vol 16 (2) ◽  
pp. 181-193 ◽  
Author(s):  
Christian Reuter ◽  
Katja Pätsch ◽  
Elena Runft
Keyword(s):  
Social Media ◽  
Information Systems ◽  
Empirical Study ◽  
State Of The Art ◽  
The State ◽  
The Internet ◽  
New Members ◽  
Starting Point ◽  
The Media

AbstractThe Internet and especially social media are not only used for supposedly good purposes. For example, the recruitment of new members and the dissemination of ideologies of terrorism also takes place in the media. However, the fight against terrorism also makes use of the same tools. The type of these countermeasures, as well as the methods, are covered in this work. In the first part, the state of the art is summarized. The second part presents an explorative empirical study of the fight against terrorism in social media, especially on Twitter. Different, preferably characteristic forms are structured within the scope with the example of Twitter. The aim of this work is to approach this highly relevant subject with the goal of peace, safety and safety from the perspective of information systems. Moreover, it should serve following researches in this field as basis and starting point.

scholarly journals The State-of-the-Art Technology of Currency Identification

2017 ◽  
Vol 9 (3) ◽  
pp. 58-72 ◽  
Author(s):  
Guangyu Wang ◽  
Xiaotian Wu ◽  
WeiQi Yan
Keyword(s):  
Feature Extraction ◽  
New Zealand ◽  
Comparative Study ◽  
State Of The Art ◽  
The State ◽  
Classification Algorithms ◽  
Security Issue ◽  
The Public ◽  
Us Dollar ◽  
Chinese Yuan

The security issue of currency has attracted awareness from the public. De-spite the development of applying various anti-counterfeit methods on currency notes, cheaters are able to produce illegal copies and circulate them in market without being detected. By reviewing related work in currency security, the focus of this paper is on conducting a comparative study of feature extraction and classification algorithms of currency notes authentication. We extract various computational features from the dataset consisting of US dollar (USD), Chinese Yuan (CNY) and New Zealand Dollar (NZD) and apply the classification algorithms to currency identification. Our contributions are to find and implement various algorithms from the existing literatures and choose the best approaches for use.

scholarly journals Human Behavior Prediction and Analysis Using Machine Learning-A Review

2021 ◽  
Vol 12 (5) ◽  
pp. 870-876
Author(s):  
Monali Gulhane, T.Sajana
Keyword(s):  
Machine Learning ◽  
Feature Extraction ◽  
Social Behavior ◽  
Human Behavior ◽  
Technical Difficulty ◽  
Vital Role ◽  
Machine Learning Algorithms ◽  
Disease Classification ◽  
Machine Learning Techniques ◽  
Classification Techniques

Nowadays many trends are being in the area of medicine to predict the human behaviour and analysis of patient behaviour is being studied but the technical difficulty of cost efficient method to predict the behaviour of user is overcome in the proposed researched methodology .The mental health of the used can lead to good immunity system to be healthy in this pandemic of COVID-19. Hence After a detailed study on different human health disease classification techniques it is found that machine learning techniques are reliable for the feature extraction and analysis of the different human parameters. CNN is the most optimum choice of classification of diseases. Feature extraction and feature selection is automatically managed by the CNN layers, which reduces the training speed. Techniques like sensor-based feature extraction like EEG, ECG, etc. will be further explored using machine learning algorithms for detection of early detections of diseases from human behavior on different platforms in this research. Social behavior and eating habits play a vital role in disease detection. A system that combines such a wide variety of features with effective classification techniques at each stage is needed. The research in this paper contributes the review of the human behavior analysis through different body parameters, food habits and social media influences with social behavior of the person. The main objective of research is to analysis theses different area parameters to predict the early signs of the diseases.

Sign in / Sign up

Export Citation Format

Share Document