Mita Nasipuri | Jadavpur University, Kolkata, India (original) (raw)
Papers by Mita Nasipuri
Advances in Intelligent Systems and Computing, 2017
ArXiv, 2021
Convolutional neural networks often generate multiple logits and use simple techniques like addit... more Convolutional neural networks often generate multiple logits and use simple techniques like addition or averaging for loss computation. But this allows gradients to be distributed equally among all paths. The proposed approach guides the gradients of backpropagation along weakest concept representations. A weakness scores defines the class specific performance of individual pathways which is then used to create a logit that would guide gradients along the weakest pathways. The proposed approach has been shown to perform better than traditional column merging techniques and can be used in several application scenarios. Not only can the proposed model be used as an efficient technique for training multiple instances of a model parallely, but also CNNs with multiple output branches have been shown to perform better with the proposed upgrade. Various experiments establish the flexibility of the learning technique which is simple yet effective in various multi-objective scenarios both em...
ArXiv, 2020
In the literature, many fusion techniques are registered for the segmentation of images, but they... more In the literature, many fusion techniques are registered for the segmentation of images, but they primarily focus on observed output or belief score or probability score of the output classes. In the present work, we have utilized inter source statistical dependency among different classifiers for ensembling of different deep learning techniques for semantic segmentation of images. For this purpose, in the present work, a class-wise Copula-based ensembling method is newly proposed for solving the multi-class segmentation problem. Experimentally, it is observed that the performance has improved more for semantic image segmentation using the proposed class-specific Copula function than the traditionally used single Copula function for the problem. The performance is also compared with three state-of-the-art ensembling methods.
The present work demonstrates a fast and improved technique for dewarping nonlinearly warped docu... more The present work demonstrates a fast and improved technique for dewarping nonlinearly warped document images. The images are first dewarped at the page-level by estimating optimum inverse projections using curvilinear homography. The quality of the process is then estimated by evaluating a set of metrics related to the characteristics of the text lines and rectilinear objects for measuring parallelism, orthogonality, etc. These are designed specifically to estimate the quality of the dewarping process without the need of any ground truth. If the quality is estimated to be unsatisfactory, the page-level dewarping process is repeated with finer approximations. This is followed by a line-level dewarping process that makes granular corrections to the warps in individual text-lines. The methodology has been tested on the CBDAR 2007 / IUPR 2011 document image dewarping dataset and is seen to yield the best OCR accuracy in the shortest amount of time, till date. The usefulness of the metho...
ArXiv, 2019
Deep neural network-based architectures give promising results in various domains including patte... more Deep neural network-based architectures give promising results in various domains including pattern recognition. Finding the optimal combination of the hyper-parameters of such a large-sized architecture is tedious and requires a large number of laboratory experiments. But, identifying the optimal combination of a hyper-parameter or appropriate kernel size for a given architecture of deep learning is always a challenging and tedious task. Here, we introduced a genetic algorithm-based technique to reduce the efforts of finding the optimal combination of a hyper-parameter (kernel size) of a convolutional neural network-based architecture. The method is evaluated on three popular datasets of different handwritten Bangla characters and digits. The implementation of the proposed methodology can be found in the following link: this https URL.
2020 25th International Conference on Pattern Recognition (ICPR), 2021
Sensing and Imaging, 2020
International Journal of Applied Pattern Recognition, 2018
Communication and Power Engineering, 2016
Advances in Intelligent and Soft Computing, 2012
2012 International Conference on Communications, Devices and Intelligent Systems (CODIS), 2012
ABSTRACT
2012 International Conference on Communications, Devices and Intelligent Systems (CODIS), 2012
ABSTRACT Proteins are responsible for all biological activities in a living object. With the adve... more ABSTRACT Proteins are responsible for all biological activities in a living object. With the advent of genome sequencing projects for different organisms, large amounts of DNA and protein sequence data is available, whereas their biological function is still un-annotated in the most of the cases. Predicting protein function is the most challenging problem in post-genomic era. Using sequence homology, phylogenetic profiles, gene expression data, and function of un-annotated protein can be predicted. Recently, the large interaction networks constructed from high throughput techniques like Yeast2Hybrid experiments are also used in prediction of protein function. Based on the concept that a protein performs similar function like its neighbor in protein Interaction network, two methods are proposed to predict protein function from protein interaction network using neighborhood properties. The first method uses neighborhood approach and second one is an intelligent technique which applies heuristic knowledge to find densely connected regions for better prediction accuracy. The overall match rate achieved in method-I is 95.8% and in method-II, it is 97.8% over 15 functional groups.
Journal of Intelligent Systems, 2013
Cellular and Molecular Biology Letters, 2011
Protein-protein interactions (PPI) control most of the biological processes in a living cell. In ... more Protein-protein interactions (PPI) control most of the biological processes in a living cell. In order to fully understand protein functions, a knowledge of protein-protein interactions is necessary. Prediction of PPI is challenging, especially when the three-dimensional structure of interacting partners is not known. Recently, a novel prediction method was proposed by exploiting physical interactions of constituent domains. We propose here a novel knowledge-based prediction method, namely PPI_SVM, which predicts interactions between two protein sequences by exploiting their domain information. We trained a two-class support vector machine on the benchmarking set of pairs of interacting proteins extracted from the Database of Interacting Proteins (DIP). The method considers all possible combinations of constituent domains between two protein sequences, unlike most of the existing approaches. Moreover, it deals with both single-domain proteins and multi domain proteins; therefore it ...
Page 1. Hierarchical Segmentation of Falsely Touching Characters from Camera Captured Degraded Do... more Page 1. Hierarchical Segmentation of Falsely Touching Characters from Camera Captured Degraded Document Images Satadal Saha 1, Subhadip Basu 2 and Mita Nasipuri 3 1 ECE Department, MCKV Institute of Engg. Liluah, Howrah, India ...
International Journal of Information, 2012
ABSTRACT: Binarization of document images is an extensively studied topic. Among the binarization... more ABSTRACT: Binarization of document images is an extensively studied topic. Among the binarization techniques, locally adaptive ones are most popular and majority of them are convolution-based. Computational requirements of such techniques make them unsuitable for low computing platforms and handheld mobile devices such as cell-phones, Personal Digital Assistants, etc. In this paper, we have presented a novel implementation approach for making convolution-based locally adaptive binarization techniques computationally ...
arXiv preprint arXiv:1003.0645, Mar 2, 2010
Abstract: Business card images are of multiple natures as these often contain graphics, pictures ... more Abstract: Business card images are of multiple natures as these often contain graphics, pictures and texts of various fonts and sizes both in background and foreground. So, the conventional binarization techniques designed for document images can not be directly applied on mobile devices. In this paper, we have presented a fast binarization technique for camera captured business card images. A card image is split into small blocks. Some of these blocks are classified as part of the background based on intensity variance. Then ...
Advances in Intelligent Systems and Computing, 2017
ArXiv, 2021
Convolutional neural networks often generate multiple logits and use simple techniques like addit... more Convolutional neural networks often generate multiple logits and use simple techniques like addition or averaging for loss computation. But this allows gradients to be distributed equally among all paths. The proposed approach guides the gradients of backpropagation along weakest concept representations. A weakness scores defines the class specific performance of individual pathways which is then used to create a logit that would guide gradients along the weakest pathways. The proposed approach has been shown to perform better than traditional column merging techniques and can be used in several application scenarios. Not only can the proposed model be used as an efficient technique for training multiple instances of a model parallely, but also CNNs with multiple output branches have been shown to perform better with the proposed upgrade. Various experiments establish the flexibility of the learning technique which is simple yet effective in various multi-objective scenarios both em...
ArXiv, 2020
In the literature, many fusion techniques are registered for the segmentation of images, but they... more In the literature, many fusion techniques are registered for the segmentation of images, but they primarily focus on observed output or belief score or probability score of the output classes. In the present work, we have utilized inter source statistical dependency among different classifiers for ensembling of different deep learning techniques for semantic segmentation of images. For this purpose, in the present work, a class-wise Copula-based ensembling method is newly proposed for solving the multi-class segmentation problem. Experimentally, it is observed that the performance has improved more for semantic image segmentation using the proposed class-specific Copula function than the traditionally used single Copula function for the problem. The performance is also compared with three state-of-the-art ensembling methods.
The present work demonstrates a fast and improved technique for dewarping nonlinearly warped docu... more The present work demonstrates a fast and improved technique for dewarping nonlinearly warped document images. The images are first dewarped at the page-level by estimating optimum inverse projections using curvilinear homography. The quality of the process is then estimated by evaluating a set of metrics related to the characteristics of the text lines and rectilinear objects for measuring parallelism, orthogonality, etc. These are designed specifically to estimate the quality of the dewarping process without the need of any ground truth. If the quality is estimated to be unsatisfactory, the page-level dewarping process is repeated with finer approximations. This is followed by a line-level dewarping process that makes granular corrections to the warps in individual text-lines. The methodology has been tested on the CBDAR 2007 / IUPR 2011 document image dewarping dataset and is seen to yield the best OCR accuracy in the shortest amount of time, till date. The usefulness of the metho...
ArXiv, 2019
Deep neural network-based architectures give promising results in various domains including patte... more Deep neural network-based architectures give promising results in various domains including pattern recognition. Finding the optimal combination of the hyper-parameters of such a large-sized architecture is tedious and requires a large number of laboratory experiments. But, identifying the optimal combination of a hyper-parameter or appropriate kernel size for a given architecture of deep learning is always a challenging and tedious task. Here, we introduced a genetic algorithm-based technique to reduce the efforts of finding the optimal combination of a hyper-parameter (kernel size) of a convolutional neural network-based architecture. The method is evaluated on three popular datasets of different handwritten Bangla characters and digits. The implementation of the proposed methodology can be found in the following link: this https URL.
2020 25th International Conference on Pattern Recognition (ICPR), 2021
Sensing and Imaging, 2020
International Journal of Applied Pattern Recognition, 2018
Communication and Power Engineering, 2016
Advances in Intelligent and Soft Computing, 2012
2012 International Conference on Communications, Devices and Intelligent Systems (CODIS), 2012
ABSTRACT
2012 International Conference on Communications, Devices and Intelligent Systems (CODIS), 2012
ABSTRACT Proteins are responsible for all biological activities in a living object. With the adve... more ABSTRACT Proteins are responsible for all biological activities in a living object. With the advent of genome sequencing projects for different organisms, large amounts of DNA and protein sequence data is available, whereas their biological function is still un-annotated in the most of the cases. Predicting protein function is the most challenging problem in post-genomic era. Using sequence homology, phylogenetic profiles, gene expression data, and function of un-annotated protein can be predicted. Recently, the large interaction networks constructed from high throughput techniques like Yeast2Hybrid experiments are also used in prediction of protein function. Based on the concept that a protein performs similar function like its neighbor in protein Interaction network, two methods are proposed to predict protein function from protein interaction network using neighborhood properties. The first method uses neighborhood approach and second one is an intelligent technique which applies heuristic knowledge to find densely connected regions for better prediction accuracy. The overall match rate achieved in method-I is 95.8% and in method-II, it is 97.8% over 15 functional groups.
Journal of Intelligent Systems, 2013
Cellular and Molecular Biology Letters, 2011
Protein-protein interactions (PPI) control most of the biological processes in a living cell. In ... more Protein-protein interactions (PPI) control most of the biological processes in a living cell. In order to fully understand protein functions, a knowledge of protein-protein interactions is necessary. Prediction of PPI is challenging, especially when the three-dimensional structure of interacting partners is not known. Recently, a novel prediction method was proposed by exploiting physical interactions of constituent domains. We propose here a novel knowledge-based prediction method, namely PPI_SVM, which predicts interactions between two protein sequences by exploiting their domain information. We trained a two-class support vector machine on the benchmarking set of pairs of interacting proteins extracted from the Database of Interacting Proteins (DIP). The method considers all possible combinations of constituent domains between two protein sequences, unlike most of the existing approaches. Moreover, it deals with both single-domain proteins and multi domain proteins; therefore it ...
Page 1. Hierarchical Segmentation of Falsely Touching Characters from Camera Captured Degraded Do... more Page 1. Hierarchical Segmentation of Falsely Touching Characters from Camera Captured Degraded Document Images Satadal Saha 1, Subhadip Basu 2 and Mita Nasipuri 3 1 ECE Department, MCKV Institute of Engg. Liluah, Howrah, India ...
International Journal of Information, 2012
ABSTRACT: Binarization of document images is an extensively studied topic. Among the binarization... more ABSTRACT: Binarization of document images is an extensively studied topic. Among the binarization techniques, locally adaptive ones are most popular and majority of them are convolution-based. Computational requirements of such techniques make them unsuitable for low computing platforms and handheld mobile devices such as cell-phones, Personal Digital Assistants, etc. In this paper, we have presented a novel implementation approach for making convolution-based locally adaptive binarization techniques computationally ...
arXiv preprint arXiv:1003.0645, Mar 2, 2010
Abstract: Business card images are of multiple natures as these often contain graphics, pictures ... more Abstract: Business card images are of multiple natures as these often contain graphics, pictures and texts of various fonts and sizes both in background and foreground. So, the conventional binarization techniques designed for document images can not be directly applied on mobile devices. In this paper, we have presented a fast binarization technique for camera captured business card images. A card image is split into small blocks. Some of these blocks are classified as part of the background based on intensity variance. Then ...