A5068266794- Speech and Audio Processing
- Handwritten Text Recognition Techniques
- Advanced Data Compression Techniques
- Vehicle License Plate Recognition
- Speech Recognition and Synthesis
- Image Retrieval and Classification Techniques
- Face and Expression Recognition
- Image Processing and 3D Reconstruction
- Face recognition and analysis
- Recommender Systems and Techniques
- Hand Gesture Recognition Systems
- Wastewater Treatment and Reuse
- Optimization and Mathematical Programming
- CCD and CMOS Imaging Sensors
- Machine Learning and Algorithms
- Sports Performance and Training
- ECG Monitoring and Analysis
- Indoor and Outdoor Localization Technologies
- Music and Audio Processing
- Fractal and DNA sequence analysis
- Wireless Signal Modulation Classification
- Advanced Adaptive Filtering Techniques
- Remote-Sensing Image Classification
- Water Quality Monitoring Technologies
- Digital Filter Design and Implementation
Indian Institute of Information Technology Vadodara
2024
Indian Institute of Technology Guwahati
2019-2024
Jaypee Institute of Information Technology
2023
Indian Institute of Technology Dhanbad
2018-2022
Graphic Era University
2021
Indian Institute of Technology Roorkee
2018
Malaviya National Institute of Technology Jaipur
2010-2016
Punjabi University
2010
Lung cancer is jeopardized disease which easily demolish human body with great impact. Cancer most prominent nowadays. To diagnosis this minacious CT scan one of the methods more often used. Manual detection anomalies from medical images It takes a lot time and prone to aberration. Several computer-aided have been developed for accurate cancers images. With advent deep learning approaches in image processing, researchers moved developing new efficient models classification. In work,...
This paper presents a new approach to off-line handwritten numeral recognition.From the concept of perturbation due writing habits and instruments, we propose recognition method which is able account for variety distortions eccentric handwriting.The numerals challenging task in field image processing pattern recognition.It can be considered as one benchmarks evaluating feature extraction methods performance classifiers.The character system depends heavily on what kind features are being...
In this paper a classification for natural images is proposed using hybrid features.The objective of to develop an image content based classifier, which can perform identity check image.Here we have extracted wavelet and color features from captured classify out three groups.The developed technique able translation rotation invariant matching among feed forward back propagation neural network.The database contains several hundreds groups namely coast, Forest, Mountain found good rate.
This work aims to enhance the quality of narrowband (0–4 kHz) voice signal in terms frequency components, i.e. missing high-frequency components a range 4–8 kHz. The proposed artificial bandwidth extension framework uses H ∞ optimisation. In this context, model is used get better representation wideband (0–8 information signal. optimisation obtain synthesis filter for given model, which synthesise high-band (4–8 discrete Fourier transform addition performed add and estimated removing leaked...
OCR remains a challenging area of research for past few years. Accuracy primarily depends on how correctly characters are segmented. This makes segmentation crucial step. work describes the novel technique character handwritten Hindi text using basic structural properties. The proposed method first uses polygonal approximation to get digitally straight line segment word. On these segments, we apply traversing algorithm which results into characters. success rate is 95.70%, satisfactory...
This work presents a new method for artificial bandwidth extension (ABE) in narrowband telephonic communication. In this regard, we use signal model and H <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">∞</sup> optimization to obtain synthesis filter representing the wideband information of signal. We need estimate high-band Hence, construct which retains filter. Signal models may not be same different speech signals because their non-stationary...
Electrocardiogram (ECG) is a non-invasive technique which used as main diagnostic device for cardiovascular diseases.A cleaned ECG signal offers essential information electrophysiology the heart diseases and ischemic changes that might happen.It gives significant functional aspects of system.In this research, detection cardiac arrhythmias in consists of: pre-processing using DWT, QRS complex signal; feature extraction from detected complexes; classification beats extracted set complexes.To...