Faculty Publications
Permanent URI for this communityhttps://idr.nitk.ac.in/handle/123456789/18736
Publications by NITK Faculty
Browse
63 results
Search Results
Item Performance evaluation of OCR on poor resolution text document images using different pre processing steps(Institute of Electrical and Electronics Engineers Inc., 2015) Naganjaneyulu, G.V.S.S.K.R.; Narasimhadhan, A.V.; Venkatesh, K.The performance of optical character recognition (OCR) algorithm is poor on low resolution scanned text images. The conventional low pass filters in L2 space can slightly improve the performance. The method of enhancement of poor resolution text images using a low pass signal filtering algorithm in the weighted Sobolev space results in high pass correction similar to un sharp masking. This can further improve the performance of OCR on low resolution text images. In this paper, the performance of a typical OCR system on low resolution scanned text images, is studied without using any preprocessing step, with low pass filtering in L2 space, and compared with low pass filtering in weighted Sobolev space as pre processing steps. © 2014 IEEE.Item A Method for QRS Delineation Based on STFT Using Adaptive Threshold(Elsevier, 2015) Shaik, B.S.; Naganjaneyulu, G.V.S.S.K.R.; Chandrasheker, T.; Narasimhadhan, A.V.Electrocardiogram (ECG) is the electrical manifestation of the contractile activity of the heart. In this work, it is proposed to utilize an adaptive threshold technique on spectrogram computed using Short Time Fourier Transform (STFT) for QRS complex detection in electrocardiogram (ECG) signal. The algorithm consists of preprocessing the raw ECG signal to remove the power-line interference, computing the STFT, applying adaptive thresholding technique and followed by identifying QRS peaks. Sensitivity, Specificity and Detection error rate are calculated on MIT-BIH database using the proposed method, which yields a competitive results when compared with the state of art in QRS detection. © 2015 The Authors.Item Adaptive Learning Rate for Visual Tracking Using Correlation Filters(Elsevier B.V., 2016) Asha, C.S.; Narasimhadhan, A.V.Visual tracking is a difficult problem in computer vision due to illumination, pose, scale, appearance variations of object. Most of the trackers use either gray scale/color information or gradient information for image description. However the use of multiple channel features provide more information than single feature alone. Recently correlation filter based video tracking gained popularity due to its efficiency and high frame rate. Existing correlation filters use fixed learning rate to update filter template in every frame. In this paper, a method for adapting learning rate in correlation filter (CF) is presented which depends on the position of target in the present and previous frames (target velocity). This method uses integral channel features in correlation filter framework with adaptive learning rate to efficiently track the object. We experiment this technique on 12 challenging video sequences from visual object tracking (VOT challenges) datasets. Proposed technique can track any object irrespective of illumination variance, occlusion, scale change and outperforms the state-of-the-art trackers. © 2016 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license.Item A novel approach for QRS delineation in ECG signal based on chirplet transform(Institute of Electrical and Electronics Engineers Inc., 2016) Shaik, B.S.; Naganjaneyulu, G.V.S.S.K.R.; Narasimhadhan, A.V.ECG analysis is used significantly in diagnosis, and biometrics. QRS complex detection is an important step in any application involving ECG signal. In this work, a novel approach for QRS complex detection based on chirplet transform is proposed. The QRS detection algorithm proposed in this work mainly consists of four steps. A preprocessing step to remove power line interference, computation of chirplet transform, an adaptive threshold technique for detecting possible QRS complex peaks, and followed by a decision making step. The performance of proposed algorithm for QRS complex detection is evaluated on MIT-BIH database and compared with the results of different algorithms in the state of art. The performance of the algorithm is comparable with the state of art of QRS complex detection. © 2015 IEEE.Item R peak delineation in ECG signal based on polynomial chirplet transform using adaptive threshold(Institute of Electrical and Electronics Engineers Inc., 2016) Naganjaneyulu, G.V.S.S.K.R.; Shaik, B.S.; Narasimhadhan, A.V.R peak delineation is fundamental step in any application implicating electrocardiogram (ECG) signal. ECG is non stationary and non linear. Hence, linear transforms like short time fourier transform, wavelet transform and chirplet transform may be inadequate to represent ECG signal and consequently for R peak delineation. Polynomial chirplet transform (PCT) models the frequency into a higher order polynomial to enhance the representation of non stationary signals whose frequency vary non linearly with time. In this paper, PCT based R peak delineation method using adaptive threshold is proposed. The performance of the proposed algorithm is evaluated on ECG ID data base taken from physionet data bank. This work also presents a comparative study of QRS detection methods employing the uni scale family of time frequency analysis methods, short time fourier transform, chirplet transform, stockwell transform, wigner ville distribution, and pseudo wigner ville distribution out of which stockwell transform, pseudo wigner ville distribution along with adaptive threshold are applied to QRS detection for the first time. The results show that the proposed method outperforms the competitors in terms of sensitivity, specificity and detection error rate. © 2016 IEEE.Item Assessment of speckle denoising in ultrasound carotid images using least square Bayesian estimation approach(Institute of Electrical and Electronics Engineers Inc., 2017) Yamanakkanavar, Y.; Asha, C.S.; Narasimhadhan, A.V.The ultrasound carotid images affected by speckle noise, which highly reduces the image quality and effects the human interpretation. Speckle removal is substantial and critical step for preprocessing of ultrasound carotid images. For robust diagnosis, the carotid images must be free of noise and clear in clinical practices. The carotid ultrasound images have multiplicative noise and is very difficult to remove as compared to additive noise. To address this issue we propose to use Bayesian least square estimation in the logarithmic space. The proposed algorithm is tested on 50 ultrasound B mode carotid images and the performance of the algorithm is compared with the existing algorithms like Median filter, Speckle Reducing Anisotropic Diffusion(SRAD), Non Local Mean (NLM) filter, Total Variation (TV), Detail Preserving Anisotropic Diffusion(DPAD) filter, Lee filter, Frost filter and Wavelet filter. Experimental result shows that proposed algorithm capable of achieving better results as compared to the other methods in terms of signal to noise ratio (SNR), peak signal to noise ratio (PSNR), Correlation of Coefficient (CoC), Structural Similarity Index Map (SSIM) and Image Quality Index(IQI) measures. As per visual inspection concerned the proposed approach is more effective in terms of suppression of noise and image enhancement. © 2016 IEEE.Item A novel method for pitch detection via instantaneous frequency estimation using polynomial chirplet transform(Institute of Electrical and Electronics Engineers Inc., 2017) Naganjaneyulu, G.V.S.S.K.R.; Ramana, M.V.; Narasimhadhan, A.V.Speech processing and synthesis have been the interests of many researchers for the past few decades. One of the primary task in speech processing is the estimation of fundamental frequency of speech (also known as pitch). Speech is a non stationary signal whose frequency varies arbitrarily with time. Linear time frequency analysis tools such as short time fourier transform may not be convenient for estimation of pitch of speech. Polynomial chirplet transform models the frequency of speech signal by a higher order polynomial of time, which makes it suitable for analysis of speech to extract pitch. In this work, a novel algorithm is proposed for pitch detection in speech by estimating instantaneous frequency (IF) using polynomial chirplet transform. The proposed algorithm is applied on a part of TIMIT speech database to find the pitch of speech of different male and female persons. © 2016 IEEE.Item Thermal vision human classification and localization using bag of visual word(Institute of Electrical and Electronics Engineers Inc., 2017) Malpani, S.; Asha, C.S.; Narasimhadhan, A.V.Human detection in thermal images has recently gained a lot of attention in computer vision due to its large number of applications. The characteristics of thermal images are poor illumination, low contrast due to capturing devices and poor environment conditions. Human classification and localization are being done using bag of visual word method. Bag of visual word method has been widely used for visible spectrum. In this work, we have extended it to thermal images. A new human detection scheme is present for thermal image using SURF features with Bag of Word. SURF has been compared with different binary feature descriptors. SURF feature descriptor outperforms BRISK and FREAK feature descriptors in terms of accuracy, F-score. © 2016 IEEE.Item A multi clue heuristic based algorithm for table detection(Institute of Electrical and Electronics Engineers Inc., 2017) Naganjaneyulu, G.V.S.S.K.R.; Sathwik, N.V.; Narasimhadhan, A.V.Research in the field of document analysis and document recognition experienced reverent growth in the past decade as automation of the office document became essential for daily life. Text in documents can take different forms like hand written text, printed text, headings signatures, tables and graphics. Extraction of tables plays a crucial role in layout analysis, and retaining the important information present in tables. In this work, a multi clue heuristic based table detection algorithm using hough lines and corner harris corner is proposed. Hough lines and harris corner points are extracted from the document in two parallel process. The clues extracted from both the process are matched using nearest neighbor framework to yield tables from the documents. The proposed algorithm is a simple paradigm for extraction of tables that are formed by lines. The performance of the proposed algorithm is tested on different types of documents that contain tables to observe an accuracy of 89.7 %. © 2016 IEEE.Item Experimental evaluation of feature channels for object tracking in RGB and thermal imagery using correlation filter(Institute of Electrical and Electronics Engineers Inc., 2017) Asha, C.S.; Narasimhadhan, A.V.Correlation filter based trackers are well studied for object tracking and shown great interest to the research community in recent years. The vast majority of the works make utilization of either color feature channels or Histogram of Gradient feature channels for object tracking in visual spectrum. However, the strength of feature channels varies from RGB videos to thermal infrared videos. Subsequently, an assessment of feature channels in RGB and thermal imagery is needed to select the best features. In this work, we study the performance of various feature channels under kernelized correlation filter framework in RGB recordings, by taking 33 videos from object tracking benchmark (OTB) dataset and thermal infrared recordings, by taking 25 thermal videos from Thermal InfraRed (LTIR) dataset. Performance of each feature channels in both imaging modes are quantified using distance precision score, overlap score, average center location error and speed metrics. The best performance is obtained when HOG and color name features are utilized for RGB videos and gradient and gabor features are used in thermal videos among selected feature sets in kernelized correlation filter framework. © 2017 IEEE.