@knit.ac.in
Associate Professor
Kamla Nehru Institute of Technology Sultanpur
Scopus Publications
Scholar Citations
Scholar h-index
Scholar i10-index
Purnima Pal, Manju Nandal, Srishti Dikshit, Aarushi Thusu, and Harsh Vikram Singh
European Alliance for Innovation n.o.
A heart stroke, also known as a myocardial infarction or heart attack, is a critical medical condition that arises when there is an obstruction in the coronary arteries that provide blood to the heart muscles. This blockage results in a diminished flow of blood and oxygen to a specific area of the heart. This abrupt interruption initiates a gradual sequence of heart muscle damage, which can lead to varying degrees of functional impairment. The severity of these impairments is primarily determined by the precise location of the heart muscle affected. Therefore, it is of utmost importance to identify the warning signs and symptoms of a stroke as soon as possible. This is the objective of this paper is to early recognition and prompt action can significantly improve the chances of a healthy and fulfilling life following a stroke. In this research work, the Stroke dataset is pre-processed and on pre-processed dataset machine learning and ensemble machine learning techniques were employed to develop and assess several models aimed at creating a stable framework for predicting the enduring stroke risk. And various matrices like accuracy, F1 score, ROC, precision, and recall are calculated. Among all models, AdaBoost model demonstrated exceptional performance validated through multiple metrics, including Precision, AUC, recall, accuracy, and F1-measure. The results underscored superiority of the AdaBoost classification method, achieving an impressive Accuracy of 99%. AdaBoost model may serve as a stable framework for predicting enduring stroke risk, emphasizing its potential utility in clinical settings for identifying individuals at higher risk of experiencing a stroke.
Nazar Zaki, Harsh Singh, Loo Chu Kiong, Nadeen Zaki, and Salama Alnuaimi
IEEE
Accurately estimating population density is a crucial component of policy-making for the development of any country. Traditionally, population density has been estimated through labor-intensive surveys that can be time-consuming and prone to error. Census data, while useful, is only collected once every 10 years or so and can take a long time to process, depending on the geography and population of the region. This makes it difficult for organizations that require up-to-date population density information for instant policy designing. To address this issue, we propose a novel approach to estimate population density using satellite imagery. Our method leverages the correlation between car density and population density. Specifically, we validate this assumption by counting cars over Dubai city using a Faster RCNN object detector with a ResNeXt-101 ($32\\times 8d$)-FP backbone and calculating the correlation between car density and population density. Our results show a significant value of the Pearson correlation coefficient, demonstrating a strong relationship between population density and car density. This innovative approach allows for the rapid estimation of population density, without the need for time-consuming and labor-intensive surveys.
Kumari Suniti Singh, Pushpanjali Singh, and Harsh Vikram Singh
IEEE
Due to advancement in technology nowadays, it is easy to manipulate and produce illegal multimedia data. As a consequent, different strategy of information security has been created by the researchers and specialists every day. Digital watermarking is one of the various techniques used these days to secure the data in efficient manner. In this data has been embedded using a suitable algorithm and watermark. Again at the receiving end watermark has been extracted for the reconstruction of original data. Watermarking can be done using spatial domain or transform domain. In this paper watermarking has been done using transform domain techniques on digital images. The proposed method includes DWT-SVD hybrid watermarking technique in addition to watermark generation. This method has been performed on medical images for securing the data. The original image is firstly processed by using DWT followed by SVD. The results have been tested by applying various attacks for the robustness of algorithm. The quality of an image can be evaluated by different parameters as PSNR, SNR and SSIM. The proposed method achieved a good value of PSNR and SSIM.
Kumari Suniti Singh and Harsh Vikram Singh
Springer Singapore
Shrish Bajpai, Naimur Rahman Kidwai, Harsh Vikram Singh, and Amit Kumar Singh
Springer Science and Business Media LLC
Sima Sahu, Amit Kumar Singh, Harsh Vikram Singh, and Basant Kumar
Elsevier
S. K. Singh, Harsh Vikram Singh, S. Chakrabarti, and S. N. Singh
Springer Singapore
Kumari Suniti Singh, Yogesh Kumar Mishra, and Harsh Vikram Singh
Springer Singapore
Sima Sahu, Harsh Vikram Singh, Amit Kumar Singh, and Basant Kumar
Springer Singapore
Sima Sahu, Harsh Vikram Singh, Basant Kumar, and Amit Kumar Singh
Walter de Gruyter GmbH
Abstract A Bayesian approach using wavelet coefficient modeling is proposed for de-noising additive white Gaussian noise in medical magnetic resonance imaging (MRI). In a parallel acquisition process, the magnetic resonance image is affected by white Gaussian noise, which is additive in nature. A normal inverse Gaussian probability distribution function is taken for modeling the wavelet coefficients. A Bayesian approach is implemented for filtering the noisy wavelet coefficients. The maximum likelihood estimator and median absolute deviation estimator are used to find the signal parameters, signal variances, and noise variances of the distribution. The minimum mean square error estimator is used for estimating the true wavelet coefficients. The proposed method is simulated on MRI. Performance and image quality parameters show that the proposed method has the capability to reduce the noise more effectively than other state-of-the-art methods. The proposed method provides 8.83%, 2.02%, 6.61%, and 30.74% improvement in peak signal-to-noise ratio, structure similarity index, Pratt’s figure of merit, and Bhattacharyya coefficient, respectively, over existing well-accepted methods. The effectiveness of the proposed method is evaluated by using the mean squared difference (MSD) parameter. MSD shows the degree of dissimilarity and is 0.000324 for the proposed method, which is less than that of the other existing methods and proves the effectiveness of the proposed method. Experimental results show that the proposed method is capable of achieving better signal-to-noise ratio performance than other tested de-noising methods.
Shrish Bajpai, Naimur Rahman Kidwai, Harsh Vikram Singh, and Amit Kumar Singh
Springer Science and Business Media LLC
Shrish Bajpai, Harsh Vikram Singh, and Naimur Rahman Kidwai
Institute of Advanced Engineering and Science
<p><span>A novel wavelet-based efficient hyperspectral image compression scheme for low memory sensors has been proposed. The proposed scheme uses the 3D dyadic wavelet transform to exploit intersubband and intrasubband correlation among the wavelet coefficients. By doing the reconstruction of the transform image cube, taking the difference between the frames, it increases the coding efficiency, reduces the memory requirement and complexity of the hyperspectral compression schemes in comparison with other state-of-the-art compression schemes.</span></p>
Harsh Vikram Singh and Purnima Pal
Springer International Publishing
Harsh Vikram Singh and Sarvesh Kumar Verma
Springer International Publishing
Sima Sahu, Harsh Vikram Singh, Basant Kumar, Amit Kumar Singh, and Prabhat Kumar
Springer International Publishing
Sima Sahu, Harsh Vikram Singh, Basant Kumar, Amit Kumar Singh, and Prabhat Kumar
Springer International Publishing
Sima Sahu, Harsh Vikram Singh, Basant Kumar, and Amit Kumar Singh
Springer Science and Business Media LLC
Ankur Rai and Harsh Vikram Singh
Springer International Publishing
Ankur Rai and Harsh Vikram Singh
Springer International Publishing
Harsh Vikram Singh and Ankur Rai
Springer Singapore
Singh Arun Kumar, Singh Juhi, and Singh Harsh Vikram
Springer Singapore
Purnima Pal, Harsh Vikram Singh, and Sarvesh Kumar Verma
IEEE
This paper comprehend the concept, details, properties, techniques and application of digital watermarking. In today's era, with the increasing use of internet it's become challenging to secure data. In sequence to authenticate or protect digital media from the various attacks techniques like cryptography, steganography and digital watermarking are introduced. Digital watermarking technique hides the secret information (Watermark) in multimedia data for the purpose of protection, copyright and authentication. The secret information is submerged into the cover data with the minimum or negligible distortion of cover data.
Shrish Bajpai, Harsh Vikram Singh, and Naimur Rahman Kidwai
IEEE
This paper presents a new approach for hyperspectral feature extraction & image classification exploiting spectral-spatial information of HSI dataset. Various methods have been developed to improve the classification accuracy. SSA has been applied to the spectral profile of the pixel where the 1-D signal can be composed into the sum of independent components including noise. Removing noisy component in extracting features, it results in the improvement of classification accuracies. Experiment results show that this SSA approach improved results in classification process using multinomial logistic regression classifiers.
Sima Sahu, Harsh Vikram Singh, Basant Kumar, and Amit Kumar Singh
Springer Science and Business Media LLC