Amlan Jyoti Das - Academia.edu (original) (raw)

Uploads

Papers by Amlan Jyoti Das

Research paper thumbnail of An Adaptive Sar Image Despeckling Algorithm Using Stationary Wavelet Transform

International Journal of Electronics Signals and Systems, 2013

In this paper, we present a Stationary Wavelet Transform (SWT) based method for the purpose of de... more In this paper, we present a Stationary Wavelet Transform (SWT) based method for the purpose of despeckling the Synthetic Aperture radar (SAR) images by applying a maximum a posteriori probability (MAP) condition to estimate the noise free wavelet coefficients. The solution of the MAP estimator is based on the assumption that the wavelet coefficients have a known distribution. Rayleigh distribution is used for modeling the speckle noise and Laplacian distribution for modeling the statistics of the noise free wavelet coefficients for the purpose of designing the MAP estimator. Rayleigh distribution is used for modeling the speckle noise since speckle noise can be well described by it. The parameters required for MAP estimator is determined by the technique used for parameter estimation after SWT. The experimental results show that the proposed despeckling algorithm efficiently removes speckle noise from the SAR images.

Research paper thumbnail of Cascade-Based Pedestrian Detector Using Edge and Pattern Features

Pedestrian detection plays a significant role in computer vision applications and has been a popu... more Pedestrian detection plays a significant role in computer vision applications and has been a popular research topic. In this paper, a new pedestrian detector is proposed with a combination of the edge-based histogram of oriented gradient features and pattern-based dense local difference binary features. An important requirement in a pedestrian detector is computational speed. Cascade-based classifiers provide a good trade-off between accuracy and speed. This work uses a cascade of boosted classifiers to enhance the detection speed. To further boost the speed, the histogram of oriented gradients is computed using integral images. The proposed system is evaluated in terms of computational speed as well as precision versus recall and miss-rate versus FPPW/FPPI. The performance is also compared with similar existing pedestrian detectors.

Research paper thumbnail of Design of pedestrian detectors using combinations of scale spaces and classifiers

Journal of King Saud University - Computer and Information Sciences, 2019

With the increasing demand for surveillance applications, pedestrian detection has been a topic o... more With the increasing demand for surveillance applications, pedestrian detection has been a topic of interest for many researchers in recent time. The quality of a pedestrian detector is decided in terms of detection accuracy and rate of detection. This paper presents new pedestrian detectors based on two types of classifiers, linear support vector machine and cascade of boosted classifier. These classifiers are trained by using a feature set comprising of the histogram of oriented gradients and dense local difference binary features. Both the image pyramid and non-linear scale space are used to detect pedestrians of various sizes. In order to combine the benefits of the two classifiers, a new two-stage detection scheme is also presented. The detection accuracies of the proposed detectors are studied in terms of miss-rate versus false positive per image and miss-rate versus false positive per window. The performances of the detectors are also compared with the performances of existing detectors of similar type.

Research paper thumbnail of Object Classification and Tracking in Real Time

Emerging Technologies in Intelligent Applications for Image and Video Processing

Research paper thumbnail of Real-time disparity computation and 3D reprojection for hand gesture recognition

2015 International Symposium on Advanced Computing and Communication (ISACC), 2015

Human hand gestures as a natural way of interaction and communicating with computers is becoming ... more Human hand gestures as a natural way of interaction and communicating with computers is becoming an emerging and demanding field of research due to its various applications like sign language recognition, human computer interaction, gaming, virtual reality etc. Hand gesture recognition under 2D environment has some limitations as the information about the other dimension (z-axis) is missed. So, hand gesture recognition under 3D environment is becoming a growing field of research. In this paper, we have implemented the realtime disparity computation and proposed a novel technique to detect gesture of Front and Back along with forward and backward movement towards and from the camera respectively. Our technique is based on stereo-vision and we have used the disparity map-based intensity measure of segmented hand and changing of its intensity as feature to classify and recognize the gesture. Stereo calibration followed by rectification is done to get the rectified images and correspondence gives the depth map. Finally three dimensional reprojection is performed. Our technique works well for the gestures and the results are promising.

Research paper thumbnail of Stereo Vision-based Hand Gesture Recognition under 3D Environment

Procedia Computer Science, 2015

Human hand gestures as a natural way of communication with computers is becoming an emerging fiel... more Human hand gestures as a natural way of communication with computers is becoming an emerging field of research due to its various applications like sign language recognition, human computer interaction, gaming, virtual reality etc. Hand gesture recognition under 2D environment has some limitations as the information about the other dimension (z-axis) is missed out. So, hand gesture recognition under 3D environment is becoming a growing field of research. In this paper, we have proposed a novel technique to detect hand gesture with forward and backward movement towards and away from the camera respectively. Our technique is based on stereo-vision and we have used a disparity map-based centroid movement and changing of its intensity as feature to recognize the gesture with conditional random field (CRF) as classifier. Stereo calibration and rectification is done to get the rectified images and correspondence gives the depth map. We tested our proposed method for Arabic numerals (0-9) and it worked efficiently with an average recognition rate of 88%.

Research paper thumbnail of A Novel Adaptive Stationary Wavelet-based Technique for SAR Image Despeckling

this paper, we present a Stationary Wavelet Transform (SWT) based method for the purpose of despe... more this paper, we present a Stationary Wavelet Transform (SWT) based method for the purpose of despeckling the Synthetic Aperture radar (SAR) images by applying a maximum a posteriori probability (MAP) condition to estimate the noise free wavelet coefficients. A MAP Estimator is designed for this purpose which uses Rayleigh distribution for modeling the speckle noise and Laplacian distribution for modeling

Research paper thumbnail of An Adaptive Rayleigh-Laplacian Based MAP Estimation Technique for Despeckling SAR Images using Stationary Wavelet Transform

International Journal of Applied Evolutionary Computation, 2013

Removal of speckle noise from Synthetic Aperture Radar (SAR) images is an important step before p... more Removal of speckle noise from Synthetic Aperture Radar (SAR) images is an important step before performing any image processing operations on these images. This paper presents a novel Stationary Wavelet Transform (SWT) based technique for the purpose of removing the speckle noise from the SAR returns. Maximum a posteriori probability (MAP) condition which uses a prior knowledge is used to estimate the noise free wavelet coefficients. The proposed MAP estimator is designed for this purpose which uses Rayleigh distribution for modeling the speckle noise and Laplacian distribution for modeling the statistics of the noise free wavelet coefficients. The parameters required for MAP estimator is determined by technique used for parameter estimation after SWT. Moreover an Laplacian – Gaussian based MAP estimator is also applied and the parameter estimation is done using the same method used for the proposed algorithm. For the purpose of enhancing the visual quality and to restore more edge ...

Research paper thumbnail of An Adaptive Stationary Wavelet-Based Technique for Despeckling Sar Image

In this paper, we present a Stationary Wavelet Transform (SWT) based method for the purpose of de... more In this paper, we present a Stationary Wavelet Transform (SWT) based method for the purpose of despeckling the Synthetic Aperture radar (SAR) images by applying a maximum a posteriori probability (MAP) condition to estimate the noise free wavelet coefficients. A MAP Estimator is designed for this purpose which uses Rayleigh distribution for modeling the speckle noise and Laplacian distribution for modeling the statistics of the noise free wavelet coefficients. The parameters required for MAP estimator is determined by the technique used for parameter estimation after SWT. The experimental results show that the proposed despeckling algorithm efficiently removes speckle noise from the SAR images.

Research paper thumbnail of Pedestrian detection using dense LDB descriptor combined with HOG

2016 International Conference on Information Technology (InCITe) - The Next Generation IT Summit on the Theme - Internet of Things: Connect your Worlds, 2016

Pedestrian detection plays a vital role in numerous vision-based safety and security applications... more Pedestrian detection plays a vital role in numerous vision-based safety and security applications in recent days. Given an image, a pedestrian detector computes features from it and works on the features to classify if there is pedestrian. This paper presents a new feature set for pedestrian detection where a modified version of the local difference binary features are combined with the histogram of oriented gradients features. The linear support vector machine is used as the classifier. The performance of the proposed detector is presented in terms of miss-rate versus FPPI and miss-rate versus FPPW, and is compared with available pedestrian detectors of similar type. The computational efficiency of the detector is also studied.

Research paper thumbnail of An Adaptive Sar Image Despeckling Algorithm Using Stationary Wavelet Transform

International Journal of Electronics Signals and Systems, 2013

In this paper, we present a Stationary Wavelet Transform (SWT) based method for the purpose of de... more In this paper, we present a Stationary Wavelet Transform (SWT) based method for the purpose of despeckling the Synthetic Aperture radar (SAR) images by applying a maximum a posteriori probability (MAP) condition to estimate the noise free wavelet coefficients. The solution of the MAP estimator is based on the assumption that the wavelet coefficients have a known distribution. Rayleigh distribution is used for modeling the speckle noise and Laplacian distribution for modeling the statistics of the noise free wavelet coefficients for the purpose of designing the MAP estimator. Rayleigh distribution is used for modeling the speckle noise since speckle noise can be well described by it. The parameters required for MAP estimator is determined by the technique used for parameter estimation after SWT. The experimental results show that the proposed despeckling algorithm efficiently removes speckle noise from the SAR images.

Research paper thumbnail of Cascade-Based Pedestrian Detector Using Edge and Pattern Features

Pedestrian detection plays a significant role in computer vision applications and has been a popu... more Pedestrian detection plays a significant role in computer vision applications and has been a popular research topic. In this paper, a new pedestrian detector is proposed with a combination of the edge-based histogram of oriented gradient features and pattern-based dense local difference binary features. An important requirement in a pedestrian detector is computational speed. Cascade-based classifiers provide a good trade-off between accuracy and speed. This work uses a cascade of boosted classifiers to enhance the detection speed. To further boost the speed, the histogram of oriented gradients is computed using integral images. The proposed system is evaluated in terms of computational speed as well as precision versus recall and miss-rate versus FPPW/FPPI. The performance is also compared with similar existing pedestrian detectors.

Research paper thumbnail of Design of pedestrian detectors using combinations of scale spaces and classifiers

Journal of King Saud University - Computer and Information Sciences, 2019

With the increasing demand for surveillance applications, pedestrian detection has been a topic o... more With the increasing demand for surveillance applications, pedestrian detection has been a topic of interest for many researchers in recent time. The quality of a pedestrian detector is decided in terms of detection accuracy and rate of detection. This paper presents new pedestrian detectors based on two types of classifiers, linear support vector machine and cascade of boosted classifier. These classifiers are trained by using a feature set comprising of the histogram of oriented gradients and dense local difference binary features. Both the image pyramid and non-linear scale space are used to detect pedestrians of various sizes. In order to combine the benefits of the two classifiers, a new two-stage detection scheme is also presented. The detection accuracies of the proposed detectors are studied in terms of miss-rate versus false positive per image and miss-rate versus false positive per window. The performances of the detectors are also compared with the performances of existing detectors of similar type.

Research paper thumbnail of Object Classification and Tracking in Real Time

Emerging Technologies in Intelligent Applications for Image and Video Processing

Research paper thumbnail of Real-time disparity computation and 3D reprojection for hand gesture recognition

2015 International Symposium on Advanced Computing and Communication (ISACC), 2015

Human hand gestures as a natural way of interaction and communicating with computers is becoming ... more Human hand gestures as a natural way of interaction and communicating with computers is becoming an emerging and demanding field of research due to its various applications like sign language recognition, human computer interaction, gaming, virtual reality etc. Hand gesture recognition under 2D environment has some limitations as the information about the other dimension (z-axis) is missed. So, hand gesture recognition under 3D environment is becoming a growing field of research. In this paper, we have implemented the realtime disparity computation and proposed a novel technique to detect gesture of Front and Back along with forward and backward movement towards and from the camera respectively. Our technique is based on stereo-vision and we have used the disparity map-based intensity measure of segmented hand and changing of its intensity as feature to classify and recognize the gesture. Stereo calibration followed by rectification is done to get the rectified images and correspondence gives the depth map. Finally three dimensional reprojection is performed. Our technique works well for the gestures and the results are promising.

Research paper thumbnail of Stereo Vision-based Hand Gesture Recognition under 3D Environment

Procedia Computer Science, 2015

Human hand gestures as a natural way of communication with computers is becoming an emerging fiel... more Human hand gestures as a natural way of communication with computers is becoming an emerging field of research due to its various applications like sign language recognition, human computer interaction, gaming, virtual reality etc. Hand gesture recognition under 2D environment has some limitations as the information about the other dimension (z-axis) is missed out. So, hand gesture recognition under 3D environment is becoming a growing field of research. In this paper, we have proposed a novel technique to detect hand gesture with forward and backward movement towards and away from the camera respectively. Our technique is based on stereo-vision and we have used a disparity map-based centroid movement and changing of its intensity as feature to recognize the gesture with conditional random field (CRF) as classifier. Stereo calibration and rectification is done to get the rectified images and correspondence gives the depth map. We tested our proposed method for Arabic numerals (0-9) and it worked efficiently with an average recognition rate of 88%.

Research paper thumbnail of A Novel Adaptive Stationary Wavelet-based Technique for SAR Image Despeckling

this paper, we present a Stationary Wavelet Transform (SWT) based method for the purpose of despe... more this paper, we present a Stationary Wavelet Transform (SWT) based method for the purpose of despeckling the Synthetic Aperture radar (SAR) images by applying a maximum a posteriori probability (MAP) condition to estimate the noise free wavelet coefficients. A MAP Estimator is designed for this purpose which uses Rayleigh distribution for modeling the speckle noise and Laplacian distribution for modeling

Research paper thumbnail of An Adaptive Rayleigh-Laplacian Based MAP Estimation Technique for Despeckling SAR Images using Stationary Wavelet Transform

International Journal of Applied Evolutionary Computation, 2013

Removal of speckle noise from Synthetic Aperture Radar (SAR) images is an important step before p... more Removal of speckle noise from Synthetic Aperture Radar (SAR) images is an important step before performing any image processing operations on these images. This paper presents a novel Stationary Wavelet Transform (SWT) based technique for the purpose of removing the speckle noise from the SAR returns. Maximum a posteriori probability (MAP) condition which uses a prior knowledge is used to estimate the noise free wavelet coefficients. The proposed MAP estimator is designed for this purpose which uses Rayleigh distribution for modeling the speckle noise and Laplacian distribution for modeling the statistics of the noise free wavelet coefficients. The parameters required for MAP estimator is determined by technique used for parameter estimation after SWT. Moreover an Laplacian – Gaussian based MAP estimator is also applied and the parameter estimation is done using the same method used for the proposed algorithm. For the purpose of enhancing the visual quality and to restore more edge ...

Research paper thumbnail of An Adaptive Stationary Wavelet-Based Technique for Despeckling Sar Image

In this paper, we present a Stationary Wavelet Transform (SWT) based method for the purpose of de... more In this paper, we present a Stationary Wavelet Transform (SWT) based method for the purpose of despeckling the Synthetic Aperture radar (SAR) images by applying a maximum a posteriori probability (MAP) condition to estimate the noise free wavelet coefficients. A MAP Estimator is designed for this purpose which uses Rayleigh distribution for modeling the speckle noise and Laplacian distribution for modeling the statistics of the noise free wavelet coefficients. The parameters required for MAP estimator is determined by the technique used for parameter estimation after SWT. The experimental results show that the proposed despeckling algorithm efficiently removes speckle noise from the SAR images.

Research paper thumbnail of Pedestrian detection using dense LDB descriptor combined with HOG

2016 International Conference on Information Technology (InCITe) - The Next Generation IT Summit on the Theme - Internet of Things: Connect your Worlds, 2016

Pedestrian detection plays a vital role in numerous vision-based safety and security applications... more Pedestrian detection plays a vital role in numerous vision-based safety and security applications in recent days. Given an image, a pedestrian detector computes features from it and works on the features to classify if there is pedestrian. This paper presents a new feature set for pedestrian detection where a modified version of the local difference binary features are combined with the histogram of oriented gradients features. The linear support vector machine is used as the classifier. The performance of the proposed detector is presented in terms of miss-rate versus FPPI and miss-rate versus FPPW, and is compared with available pedestrian detectors of similar type. The computational efficiency of the detector is also studied.