Arpana Mahajan
Research Subjects
Arpana Mahajan has a total of 5 co-authors. Here are the top co-authors by number of shared publications:
Sanjay Chaudhary
Computer Engineering, Madhav University, Rajasthan, India
1 shared publication- Sanjay Gour
Gandhinagar University
1 shared publication Selvakuberan Karuppasamy
Accenture, Advanced Technology Centers in India,Chennai,India
1 shared publication- Sheshang Degadwala
Sigma University
1 shared publication Subhashini Lakshminarayanan
Accenture, Advanced Technology Centers in India,Chennai,India
1 shared publication
Latest Publications
- 4 May 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1109-1114
Improving Classification of Scanned Document Images using a Novel Combination of Pre-Processing Techniques
The accuracy of the optical character recognition (OCR) outputs is critical to the success of real-world applications. It is necessary to preprocess an image before an OCR system can correctly interpret the information contained therein. The technique presented employs a novel combination of preprocessing techniques to prepare document images for OCR. Text that has been extracted may be used in a range of ways, including business intelligence operations that often involve the extraction of meaningful semantic information from massive volumes of documents for subsequent strategic decision-making activities. This paper demonstrates a unique combination of image preprocessing on scanned documents which can be leveraged for semantic document understanding using LayoutLMv2. In the RVL-CDIP dataset's advertisement class, this study has included a few more samples. The proposed objective is to show that text classification systems can classify scanned documents accurately after applying proper preprocessing techniques. On the benchmark RVL-CDIP dataset, investigation indicates that the presented methodology outperforms the current state-of-the-art method by 2.34% of classification accuracy, achieving state-of-the-art results for document image classification.
- 4 May 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1109-1114
Improving Classification of Scanned Document Images using a Novel Combination of Pre-Processing Techniques
The accuracy of the optical character recognition (OCR) outputs is critical to the success of real-world applications. It is necessary to preprocess an image before an OCR system can correctly interpret the information contained therein. The technique presented employs a novel combination of preprocessing techniques to prepare document images for OCR. Text that has been extracted may be used in a range of ways, including business intelligence operations that often involve the extraction of meaningful semantic information from massive volumes of documents for subsequent strategic decision-making activities. This paper demonstrates a unique combination of image preprocessing on scanned documents which can be leveraged for semantic document understanding using LayoutLMv2. In the RVL-CDIP dataset's advertisement class, this study has included a few more samples. The proposed objective is to show that text classification systems can classify scanned documents accurately after applying proper preprocessing techniques. On the benchmark RVL-CDIP dataset, investigation indicates that the presented methodology outperforms the current state-of-the-art method by 2.34% of classification accuracy, achieving state-of-the-art results for document image classification.
- Retracted
- 1 June 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 327-330
Retracted: Categorical Image Classification Based On Representational Deep Network (RESNET)
Categorical Image Classification needs thousands of images to train. Also, System needs more time to extract the features as well as classification. In the Proposed Scenario we are going to describe different deep learning approaches for image classification. In the proposed System uses more, deep convolutional neural network to categorize thousands of high-resolution images into eight different classes. We have extracted image features from a pre-trained Representational deep Neural network (RESNET), and use that features to train machine learning Support vector machine (SVM) classifier. Representational deep networks makes feature extraction easiest and fastest way use than any other conventional network methods. In this research paper we are describe Image Categorical classification using proposed representational deep networks (RESNET).
- 1 June 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 327-330
RetractedRetracted: Categorical Image Classification Based On Representational Deep Network (RESNET)
Categorical Image Classification needs thousands of images to train. Also, System needs more time to extract the features as well as classification. In the Proposed Scenario we are going to describe different deep learning approaches for image classification. In the proposed System uses more, deep convolutional neural network to categorize thousands of high-resolution images into eight different classes. We have extracted image features from a pre-trained Representational deep Neural network (RESNET), and use that features to train machine learning Support vector machine (SVM) classifier. Representational deep networks makes feature extraction easiest and fastest way use than any other conventional network methods. In this research paper we are describe Image Categorical classification using proposed representational deep networks (RESNET).
- 20 November 2018
- Published by Springer Nature
- p. 189-196
Multiple Objects Tracking Under Occlusion Detection in Video Sequences
- 20 November 2018
- Published by Springer Nature
- p. 189-196
Multiple Objects Tracking Under Occlusion Detection in Video Sequences