lalit kane | University of petroleum & energy studies (original) (raw)

Papers by lalit kane

Artificial Intelligence Review, Oct 25, 2022

Procedia Computer Science, 2016

This work demonstrates a real time framework to recognize trajectories articulated in the air usi... more This work demonstrates a real time framework to recognize trajectories articulated in the air using bare hand motion. A frontend is established to plot the trajectories as well as to spot the interleaved dynamic gestures. Finger detection based controls and trajectory plotting velocity help to spot the gesture boundaries. Trajectories are described through a unique Equi-Polar Signature (EPS) derived from circular grid normalization of trajectory points. EPS is invariant to translation, scale, rotation and stroke directions. k-Nearest Neighbor (KNN) classification strategy recognizes EPSs of digits 0-9 and operator symbols '+', '-', '×', and '/'. Unlike previous path alignment algorithms, the proposed EPS scheme executes in linear time and fits to real-time constraints. On a customized depth video dataset of 2280 trajectories, 94.1% recognition accuracy is achieved.

This paper presents a framework to recognize the symbols drawn in air using bare hand motion. The... more This paper presents a framework to recognize the symbols drawn in air using bare hand motion. The work marks a step towards development of non-tactile interfaces requiring no physical means for writing or drawing. To overcome the limitations of traditional two dimensional camera based acquisition, a preliminary step in gesture recognition, depth based sensor is used to acquire trajectory signals. In place of DTW (Dynamic Time Warp) and HMM (Hidden Markov Model) a non-time-warping approach is adopted in this work to recognize trajectories. Start and end delimitation of character trajectory drawing is established through finger detection based control gestures. Three simple features are evaluated by rule based and distance based classification, and classifier votes determine the recognition decision. Recognition accuracy up to 96% is achieved.

Expert Systems With Applications, Oct 1, 2022

IEEE Transactions on Human-Machine Systems, Dec 1, 2017

Hand-gesture-based commands may replace touch and electromechanical input panels to make the publ... more Hand-gesture-based commands may replace touch and electromechanical input panels to make the public interactive response systems (IRS) more accessible. This work presents a prototype framework for vision-based mid-air unistroke character input, which can be adapted as an interface for the IRS. At first, we developed an acquisition module which effectively spots the legitimate gesture trajectory by implementing pen-up and pendown actions using depth thresholding and velocity tracking. The extracted trajectory is recognized through a novel, fast, and easy to implement the equipolar signature (EPS) technique. Apart from resistance to rotation, scale, and translation variations, EPS exhibits neutrality to stroking directions as well. On the three selfcollected datasets comprising of digits, alphabets, and symbols, the EPS scheme obtains over 96.5% accurate results with an average of 30-ms running time. The proposed scheme is also validated on an open dataset DAIR (Dataset for AIR Handwriting), where it achieves 95.5% mean accuracy with 24.3-ms recognition time per gesture. The developed approach is compared with benchmark schemes to justify its accuracy and speed.

Computer Vision and Image Understanding, Dec 1, 2015

Automatic recognition of fingerspelling postures in a live environment is a challenging task prim... more Automatic recognition of fingerspelling postures in a live environment is a challenging task primarily due to the complex computation of popular moment-based and spectral descriptors. Shape matrix offers a timeefficient alternative that samples the shape region through the intersection points of adjacent log-polar sections. However, sparse sampling of the region by discrete log-polar intersection points cannot capture salience of the shape. This manuscript proposes modified forms of the shape matrix which can capture salience of the fingerspelling postures by the precise sampling of contours and regions. For effective segmentation and subsequent description, hand postures are acquired through the depth sensor. Proposed shape matrix variants are evaluated for fingerspelling recognition with one-handed and two-handed postures. Experiments are rigorously performed on three datasets including one-handed signs of American Sign Language (ASL), NTU hand digits, and both one-handed and two-handed signs of Indian Sign Language (ISL). Proposed shape matrix variants supersede the benchmark shape context and Gabor features by obtaining 94.15% accuracy on ISL dataset with minimum mean running time of 0.029 s. On ASL and NTU datasets, 91.86% and 95.11% accuracies are obtained with 0.0172 and 0.0483 s mean running times, respectively.

Computer Communications, Jul 1, 2008

Promotion of public transport usage and discouraging the use of cars and other private transports... more Promotion of public transport usage and discouraging the use of cars and other private transports improves the environment (i.e. safety, exhaust emissions, noise) and contributes to reduce the problem of global warming. This paper suggests a new cost-effective idea to locate a vehicle in public transport domain and compares proposed technique to some of the methods advised previously to find the location of a vehicle. This paper further proposes a framework that is inspired by the thematic layers of a GIS (Geographic Information System). Proposed transportation GIS works as a visualization tool and helps to process spatio-temporal queries. Building such a system will first of all require a strategy to keep track of vehicles in specific public transport, a client interface, and a query processing system to handle spatial and temporal queries put up by the users of the transport system.

Pattern Recognition Letters, Apr 1, 2019

A sequence of apparently ad-hoc hand postures can generate meaningful dynamic gestures which can ... more A sequence of apparently ad-hoc hand postures can generate meaningful dynamic gestures which can be utilized in interface controls for computer, television, or games. In order to develop deployable systems with these gestures, selected descriptors should be fast enough to meet the live recognition requirements. This paper proposes framework for a practical system capable of recognizing continuous dynamic gestures characterized by short-duration posture sequences. A depth-based modification to the shape matrix is devised to describe hand silhouettes, which gives a faster alternative to region-based descriptors. Postures are recognized using depth matrix and 1-nearest neighbor strategy. Posture sequence labels are predicted by a dynamic naive Bayes classifier which works in association with an adaptive windowing mechanism. The conducted experiments report up to 96.2% accurate results with mean accuracy of 95.2% on dynamic gesture dataset. Depth matrix computation takes a maximum of 2ms time.

Lecture notes in networks and systems, 2023

Computational Intelligence and Neuroscience, Jul 5, 2022

COVID-19 is an infectious and contagious disease caused by the new coronavirus. e total number of... more COVID-19 is an infectious and contagious disease caused by the new coronavirus. e total number of cases is over 19 million and continues to grow. A common symptom noticed among COVID-19 patients is lung infection that results in breathlessness, and the lack of essential resources such as testing, oxygen, and ventilators enhances its severity. Chest X-ray can be used to design and develop a COVID-19 detection mechanism for a quicker diagnosis using AI and machine learning techniques. Due to this silver lining, various new COVID-19 detection techniques and prediction models have been introduced in recent times based on chest radiography images. However, due to a high level of unpredictability and the absence of essential data, standard models have showcased low e ciency and also su er from overheads and complexities. is paper proposes a model ne tuning transfer learning-coronavirus 19 (Ftl-CoV19) for COVID-19 detection through chest X-rays, which embraces the ideas of transfer learning in pretrained VGG16 model with including combination of convolution, max pooling, and dense layer at di erent stages of model. Ftl-CoV19 reported promising experimental results; it observed training and validation accuracy of 98.82% and 99.27% with precision of 100%, recall of 98%, and F1 score of 99%. ese results outperformed other conventional state of arts such as CNN, ResNet50, InceptionV3, and Xception.

Journal of the Gujarat Research Society, Dec 31, 2019

Journal of the Gujarat Research Society, Dec 5, 2019

Multimedia Tools and Applications, Dec 6, 2016

Skeletons provide landmark points that preserve implicit strokes or medial structure within a sha... more Skeletons provide landmark points that preserve implicit strokes or medial structure within a shape for compact representation. Though, hand posture shapes are usually recognized by region or contour representations, some applications may only be interested in the recognition of medial structures within the postures rather than their exact outlines and regions. Proposed work identifies several unique medial structures formed by a set of both one and two-handed postures and demonstrates their pure skeletal recognition in real-time. Existing skeleton-based recognition schemes apply the complex segmental processing on underlying skeleton and rely on contour information which is not suitable for fast recognition of medial structures. Presented work applies intuitive Eigen-space based Principal Components of Symbolic Structure (PCSS) and geometric Equi-Polar Signature (EPS) features to accomplish the recognition task. Both PCV and EPS process the skeleton globally without sections without associating contour information. Recognition accuracy up to 94% is obtained on a 22 posture dataset comprising of 10,560 depth frames with 480 samples for each posture. Depth sensor based acquisition is employed to meet the real-time requirements.

Expert Systems with Applications

Revue d'Intelligence Artificielle

This work addresses an example for dynamic hand signal acknowledgment by utilizing a Kinect V2. T... more This work addresses an example for dynamic hand signal acknowledgment by utilizing a Kinect V2. The projected plan takes oneself inspired motion (general media stream) as info, separates hand region and processes hand signal highlights, and uses these elements to perceive the motion. We projected free penmanship and our strategy remembers it progressively utilizing the proposed highlight portrayal. This proposed strategy utilizes an efficient fingertip acknowledgment approach and composing with the free hand the utilization of a fingertip. We verify our strategy on Kinect V2. On a dataset gathered from various clients, we accomplish an acknowledgment exactness of 98% for character acknowledgment. We likewise show the way that this framework can be stretched out for word list acknowledgment with solid execution and additionally arranged a dataset containing data frame of the moving video and fetching characters from the database a typical benchmark to manually written character ackno...

Artificial Intelligence Review

Expert Systems with Applications

Design Engineering, Sep 30, 2021

2nd International Conference on Data, Engineering and Applications (IDEA), 2020

HCI makes a evaluation on the use of depth for hand following and gesture recognition through mul... more HCI makes a evaluation on the use of depth for hand following and gesture recognition through multi stroke in free mid-air Trajectory. The survey examines 10 papers describing depth-based gesture trajectory scheme. Various methods used with different algorithm in single stroke of the hand localization and gesture classification methods; the purpose is to make process aqua-rate, faster with various parameters. The review is organized about a new model of extreme importance in designing an intelligent and efficient human-computer interface. In the reviewed literature, the enclosed issues include preprocessing, object tracing and recognition, human activity analysis. Hand Gesture plays a most significant role in HCI for providing interactive framework, movement of the hand gesture trajectory find favorable characteristics. Various inputs should be taken through depth camera of various activities of posture of hand behavior of alphabets and numeric values with single stroke with differe...

Lecture notes in networks and systems, 2023

Artificial Intelligence Review, Oct 25, 2022

Procedia Computer Science, 2016

This work demonstrates a real time framework to recognize trajectories articulated in the air usi... more This work demonstrates a real time framework to recognize trajectories articulated in the air using bare hand motion. A frontend is established to plot the trajectories as well as to spot the interleaved dynamic gestures. Finger detection based controls and trajectory plotting velocity help to spot the gesture boundaries. Trajectories are described through a unique Equi-Polar Signature (EPS) derived from circular grid normalization of trajectory points. EPS is invariant to translation, scale, rotation and stroke directions. k-Nearest Neighbor (KNN) classification strategy recognizes EPSs of digits 0-9 and operator symbols '+', '-', '×', and '/'. Unlike previous path alignment algorithms, the proposed EPS scheme executes in linear time and fits to real-time constraints. On a customized depth video dataset of 2280 trajectories, 94.1% recognition accuracy is achieved.

This paper presents a framework to recognize the symbols drawn in air using bare hand motion. The... more This paper presents a framework to recognize the symbols drawn in air using bare hand motion. The work marks a step towards development of non-tactile interfaces requiring no physical means for writing or drawing. To overcome the limitations of traditional two dimensional camera based acquisition, a preliminary step in gesture recognition, depth based sensor is used to acquire trajectory signals. In place of DTW (Dynamic Time Warp) and HMM (Hidden Markov Model) a non-time-warping approach is adopted in this work to recognize trajectories. Start and end delimitation of character trajectory drawing is established through finger detection based control gestures. Three simple features are evaluated by rule based and distance based classification, and classifier votes determine the recognition decision. Recognition accuracy up to 96% is achieved.

Expert Systems With Applications, Oct 1, 2022

IEEE Transactions on Human-Machine Systems, Dec 1, 2017

Hand-gesture-based commands may replace touch and electromechanical input panels to make the publ... more Hand-gesture-based commands may replace touch and electromechanical input panels to make the public interactive response systems (IRS) more accessible. This work presents a prototype framework for vision-based mid-air unistroke character input, which can be adapted as an interface for the IRS. At first, we developed an acquisition module which effectively spots the legitimate gesture trajectory by implementing pen-up and pendown actions using depth thresholding and velocity tracking. The extracted trajectory is recognized through a novel, fast, and easy to implement the equipolar signature (EPS) technique. Apart from resistance to rotation, scale, and translation variations, EPS exhibits neutrality to stroking directions as well. On the three selfcollected datasets comprising of digits, alphabets, and symbols, the EPS scheme obtains over 96.5% accurate results with an average of 30-ms running time. The proposed scheme is also validated on an open dataset DAIR (Dataset for AIR Handwriting), where it achieves 95.5% mean accuracy with 24.3-ms recognition time per gesture. The developed approach is compared with benchmark schemes to justify its accuracy and speed.

Computer Vision and Image Understanding, Dec 1, 2015

Automatic recognition of fingerspelling postures in a live environment is a challenging task prim... more Automatic recognition of fingerspelling postures in a live environment is a challenging task primarily due to the complex computation of popular moment-based and spectral descriptors. Shape matrix offers a timeefficient alternative that samples the shape region through the intersection points of adjacent log-polar sections. However, sparse sampling of the region by discrete log-polar intersection points cannot capture salience of the shape. This manuscript proposes modified forms of the shape matrix which can capture salience of the fingerspelling postures by the precise sampling of contours and regions. For effective segmentation and subsequent description, hand postures are acquired through the depth sensor. Proposed shape matrix variants are evaluated for fingerspelling recognition with one-handed and two-handed postures. Experiments are rigorously performed on three datasets including one-handed signs of American Sign Language (ASL), NTU hand digits, and both one-handed and two-handed signs of Indian Sign Language (ISL). Proposed shape matrix variants supersede the benchmark shape context and Gabor features by obtaining 94.15% accuracy on ISL dataset with minimum mean running time of 0.029 s. On ASL and NTU datasets, 91.86% and 95.11% accuracies are obtained with 0.0172 and 0.0483 s mean running times, respectively.

Computer Communications, Jul 1, 2008

Promotion of public transport usage and discouraging the use of cars and other private transports... more Promotion of public transport usage and discouraging the use of cars and other private transports improves the environment (i.e. safety, exhaust emissions, noise) and contributes to reduce the problem of global warming. This paper suggests a new cost-effective idea to locate a vehicle in public transport domain and compares proposed technique to some of the methods advised previously to find the location of a vehicle. This paper further proposes a framework that is inspired by the thematic layers of a GIS (Geographic Information System). Proposed transportation GIS works as a visualization tool and helps to process spatio-temporal queries. Building such a system will first of all require a strategy to keep track of vehicles in specific public transport, a client interface, and a query processing system to handle spatial and temporal queries put up by the users of the transport system.

Pattern Recognition Letters, Apr 1, 2019

A sequence of apparently ad-hoc hand postures can generate meaningful dynamic gestures which can ... more A sequence of apparently ad-hoc hand postures can generate meaningful dynamic gestures which can be utilized in interface controls for computer, television, or games. In order to develop deployable systems with these gestures, selected descriptors should be fast enough to meet the live recognition requirements. This paper proposes framework for a practical system capable of recognizing continuous dynamic gestures characterized by short-duration posture sequences. A depth-based modification to the shape matrix is devised to describe hand silhouettes, which gives a faster alternative to region-based descriptors. Postures are recognized using depth matrix and 1-nearest neighbor strategy. Posture sequence labels are predicted by a dynamic naive Bayes classifier which works in association with an adaptive windowing mechanism. The conducted experiments report up to 96.2% accurate results with mean accuracy of 95.2% on dynamic gesture dataset. Depth matrix computation takes a maximum of 2ms time.

Lecture notes in networks and systems, 2023

Computational Intelligence and Neuroscience, Jul 5, 2022

COVID-19 is an infectious and contagious disease caused by the new coronavirus. e total number of... more COVID-19 is an infectious and contagious disease caused by the new coronavirus. e total number of cases is over 19 million and continues to grow. A common symptom noticed among COVID-19 patients is lung infection that results in breathlessness, and the lack of essential resources such as testing, oxygen, and ventilators enhances its severity. Chest X-ray can be used to design and develop a COVID-19 detection mechanism for a quicker diagnosis using AI and machine learning techniques. Due to this silver lining, various new COVID-19 detection techniques and prediction models have been introduced in recent times based on chest radiography images. However, due to a high level of unpredictability and the absence of essential data, standard models have showcased low e ciency and also su er from overheads and complexities. is paper proposes a model ne tuning transfer learning-coronavirus 19 (Ftl-CoV19) for COVID-19 detection through chest X-rays, which embraces the ideas of transfer learning in pretrained VGG16 model with including combination of convolution, max pooling, and dense layer at di erent stages of model. Ftl-CoV19 reported promising experimental results; it observed training and validation accuracy of 98.82% and 99.27% with precision of 100%, recall of 98%, and F1 score of 99%. ese results outperformed other conventional state of arts such as CNN, ResNet50, InceptionV3, and Xception.

Journal of the Gujarat Research Society, Dec 31, 2019

Journal of the Gujarat Research Society, Dec 5, 2019

Multimedia Tools and Applications, Dec 6, 2016

Skeletons provide landmark points that preserve implicit strokes or medial structure within a sha... more Skeletons provide landmark points that preserve implicit strokes or medial structure within a shape for compact representation. Though, hand posture shapes are usually recognized by region or contour representations, some applications may only be interested in the recognition of medial structures within the postures rather than their exact outlines and regions. Proposed work identifies several unique medial structures formed by a set of both one and two-handed postures and demonstrates their pure skeletal recognition in real-time. Existing skeleton-based recognition schemes apply the complex segmental processing on underlying skeleton and rely on contour information which is not suitable for fast recognition of medial structures. Presented work applies intuitive Eigen-space based Principal Components of Symbolic Structure (PCSS) and geometric Equi-Polar Signature (EPS) features to accomplish the recognition task. Both PCV and EPS process the skeleton globally without sections without associating contour information. Recognition accuracy up to 94% is obtained on a 22 posture dataset comprising of 10,560 depth frames with 480 samples for each posture. Depth sensor based acquisition is employed to meet the real-time requirements.

Expert Systems with Applications

Revue d'Intelligence Artificielle

This work addresses an example for dynamic hand signal acknowledgment by utilizing a Kinect V2. T... more This work addresses an example for dynamic hand signal acknowledgment by utilizing a Kinect V2. The projected plan takes oneself inspired motion (general media stream) as info, separates hand region and processes hand signal highlights, and uses these elements to perceive the motion. We projected free penmanship and our strategy remembers it progressively utilizing the proposed highlight portrayal. This proposed strategy utilizes an efficient fingertip acknowledgment approach and composing with the free hand the utilization of a fingertip. We verify our strategy on Kinect V2. On a dataset gathered from various clients, we accomplish an acknowledgment exactness of 98% for character acknowledgment. We likewise show the way that this framework can be stretched out for word list acknowledgment with solid execution and additionally arranged a dataset containing data frame of the moving video and fetching characters from the database a typical benchmark to manually written character ackno...

Artificial Intelligence Review

Expert Systems with Applications

Design Engineering, Sep 30, 2021

2nd International Conference on Data, Engineering and Applications (IDEA), 2020

HCI makes a evaluation on the use of depth for hand following and gesture recognition through mul... more HCI makes a evaluation on the use of depth for hand following and gesture recognition through multi stroke in free mid-air Trajectory. The survey examines 10 papers describing depth-based gesture trajectory scheme. Various methods used with different algorithm in single stroke of the hand localization and gesture classification methods; the purpose is to make process aqua-rate, faster with various parameters. The review is organized about a new model of extreme importance in designing an intelligent and efficient human-computer interface. In the reviewed literature, the enclosed issues include preprocessing, object tracing and recognition, human activity analysis. Hand Gesture plays a most significant role in HCI for providing interactive framework, movement of the hand gesture trajectory find favorable characteristics. Various inputs should be taken through depth camera of various activities of posture of hand behavior of alphabets and numeric values with single stroke with differe...

Lecture notes in networks and systems, 2023