Virtual motion and pose from images and video can be estimated by detecting body joints and their... more Virtual motion and pose from images and video can be estimated by detecting body joints and their interconnection. The human body has diverse and complicated poses in yoga, making its classification challenging. This study estimates yoga poses from the images using a neural network. Five different yoga poses, viz. downdog, tree, plank, warrior2, and goddess in the form of RGB images are used as the target inputs. The BlazePose model was used to localize the body joints of the yoga poses. It detected a maximum of 33 body joints, referred to as keypoints, covering almost all the body parts. Keypoints achieved from the model are considered as predicted joint locations. True keypoints, as the ground truth body joint for individual yoga poses, are identified manually using the open source image annotation tool named Makesense AI. A detailed analysis of the body joint detection accuracy is proposed in the form of percentage of corrected keypoints (PCK) and percentage of detected joints (P...
Interconnected three-element and four-element wideband MIMO antennas have been proposed for milli... more Interconnected three-element and four-element wideband MIMO antennas have been proposed for millimeter-wave 5G applications by performing numerical computations and carrying out experimental measurements. The antenna structure is realized using Rogers 5880 substrate (εr = 2.2, tan δ = 0.0009), where the radiating element has the shape of an inverted L with a partial ground. The unit element is carefully designed and positioned (by orthogonally rotating the elements) to form three-element (case 1) and four-element (case 2) MIMO antennas. The interconnected ground for both cases is ascertained to increase the practical utilization of the resonator. The proposed MIMO antenna size is (0.95λ × 3λ) for case 1 and (2.01λ × 1.95λ) for case 2 (at the lowest functional frequency). Both the designs give an impedance bandwidth of approximately 26–40 GHz (43%). Moreover, they achieve greater than 15 dB isolation and more than 6 dBi gain with an ECC value lower than 0.02, which meets the MIMO div...
For autonomous navigation, three main functions are essential: finding the location, creating the... more For autonomous navigation, three main functions are essential: finding the location, creating the mapping, and getting the optimum path. Since Human operators study the map and correlate it to aerial pictures to locate target locations, consistent localization and mapping concurrently are challenging tasks. Light detection and ranging (LiDAR) can create a 2-dimensional (2D) map and generate positional data for the indoor area, but it fails in the presence of a dynamic object. The global positioning systems (GPS) data offering precise location tracking in outdoor spaces can tackle the weakness of LiDAR. Therefore, we design a robot operating system (ROS) based vehicular system integrating GPS and LiDAR data. The Inertial Measurement Unit (IMU) is used to make an educated approximation for LiDAR registration. A Rao-Blackwellized particle filter (RBPF) based Gmapping algorithm has been retreated in the proposed system using sensors data for navigation and mapping, where each particle h...
Purpose This paper aims to present, design and implement a novel half-mode substrate integrated w... more Purpose This paper aims to present, design and implement a novel half-mode substrate integrated waveguide (HMSIW)-based narrow bandpass filter, which offers advantages like low insertion loss, compact size and high selectivity. Proposed filter will be used in the K-band automotive radar application. Design/methodology/approach The filtering response in the proposed design is achieved by inserting inductive posts in the HMSIW cavity. Ansoft high frequency structure Simulator (HFSS) is used for the simulation of the proposed structure, which is a three-dimensional full-wave solver using the finite element method (FEM). The proposed filter is fabricated on the dielectric material RT duroid 5,880 with the dielectric constant ɛr = 2.2, dissipation factor t and = 4 × 10–4 and height h = 0.508 mm. Findings Frequency tuning is also carried out by changing the lateral distance between two inductive posts. Moreover, a comparison of the proposed structure with the previously published work is ...
Modeling and Optimization in Science and Technologies, 2015
This chapter presents the application of a new meta-heuristic optimization algorithm called cucko... more This chapter presents the application of a new meta-heuristic optimization algorithm called cuckoo search algorithm (CSA) to solve optimal reactive power dispatch problem (ORPD) of the power system in the presence of wind power plants (WPP). Due to the inclusion of WPP, the ORPD problem becomes a complex combinatorial optimization problem and it has a nonlinear objective function with many local minima, and discontinuous and nonlinear constraint functions. CSA is based on the obligate brood parasitic behavior of some cuckoo species in combination with the Lѐvy flight behavior of some birds and fruit flies. The effectiveness and feasibility of CSA have been tested on a 41-bus WPP test system and the obtained results that have been compared with particle swarm optimization (PSO). Simulation results yield that the CSA converges to better optimal solutions faster than PSO.
Image segmentation is the fundamental step in various video processing applications like object t... more Image segmentation is the fundamental step in various video processing applications like object tracking, pattern recognition, industrial automation, object classification etc. Earlier, low-level curve gradient methods were used for the real time image segmentation where images were processed pixel by pixel. However, the gradient models provide discontinuous edges and are highly sensitive to the noise in the image. High level Mumford-Shah based Active Contour Model (ACM) provides continuous edges and is less sensitive to noise but the complexity in the algorithm and its iterative implementation makes it challenging for the real time applications. This paper investigates the computational complexity and proposes optimum approximation of segmentation with removal of contour initialization. The proposed algorithm is implemented on Xilinx's Zynq ZC702 FPGA board. The removal of contour initialization increased the speed by 10% in software and fixed point implementation on Xilinx provided 80% segmentation accuracy with very low latency of 7.3270 ns per iteration.
Purpose The digital down converter (DDC) is a principal component in modern communication systems... more Purpose The digital down converter (DDC) is a principal component in modern communication systems. The DDC process traditionally entails quadrature down conversion, bandwidth reducing filters and commensurate sample rate reduction. To avoid group delay, distortion linear phase FIR filters are used in the DDC. The filter performance specifications related to deep stopband attenuation, small in-band ripple and narrow transition bandwidth lead to filters with a large number of coefficients. To reduce the computational workload of the filtering process, filtering is often performed as a two-stage process, the first stage being a down sampling Hoegenauer (or cascade-integrated comb) filter and a reduced sample rate FIR filter. An alternative option is an M-Path polyphase partition of a band cantered FIR filter. Even though IIR filters offer reduced workload to implement a specific filtering task, the authors avoid using them because of their poor group delay characteristics. This paper a...
International Journal of Online Engineering (iJOE), 2017
This paper proposes the design and implementation of prototype model to control embroidery machin... more This paper proposes the design and implementation of prototype model to control embroidery machine using single board ARM processor. Wilcom is the software largely used for embroidery design. Designs prepared using Wilcom software are printed and supplied to the machine for further process. The proposed prototype model extracts the design data from this software by avoiding the printing and/or scanning of the user design and the design data are supplied to ARM based prototype embroidery machine for printing. In proposed model, the pencil is used as printing head and A4 size paper is used for the printing. Further to control the head, Use of web server design is also proposed using ARM processor.
This paper embodies the design and development of a compact Coplanar Waveguide (CPW) fed connecte... more This paper embodies the design and development of a compact Coplanar Waveguide (CPW) fed connected ground Multiple-Input-Multiple-Output (MIMO) antennas operating in the sub-millimeter-wave 5G New Radio (NR) n257/n258/n261 bands. The planar geometry leads to a small and compact structure while achieving a wide operating bandwidth, high gain, and better radiation efficiency. The top surface of the antenna comprises a modified CPW in the form of two circular structures that feeds the centrally slotted circular patch. The single antenna structure is arranged in a rotational orthogonal manner forming a 4-port structure. The ground plane on the bottom of a 4-port structure is connected using a circular ring which is carefully optimized for achieving isolation levels>20 dB across the band of interest. The sub-mm-wave resonating 4port antenna achieves a compact size of 24 Â 24 mm 2 , a wide bandwidth of 24.8-44.45 GHz (79.35%), the maximum gain of 8.6dBi, and minimum efficiency of 85% across the bands of interest. The proposed antenna element is fabricated over Rogers 5880 substrate and experimental tests are carried out, where a good correlation between the scattering parameters, transmission
Journal of Automation, Mobile Robotics and Intelligent Systems, 2019
Presently the interaction of robots with human plays an important role in various social applicat... more Presently the interaction of robots with human plays an important role in various social applications. Reliable tracking is an important aspect for the social robots where robots need to follow the moving person. This paper proposes the implementation of automated tracking and real time following algorithm for robotic automation. Occlusion and identity retention are the major challenges in the tracking process. Hence, a feature set based identity retention algorithm is used and integrated with robot operating system. The tracking algorithm is implemented using robot operating system in Linux and using OpenCV. The tracking algorithm achieved 85% accuracy and 72.30% precision. Further analysis of tracking algorithm corresponds to the integration of ROS and OpenCV is presented. The analysis of tracking algorithm concludes that ROS linking required 0.64% more time in comparison with simple OpenCV code based tracking algorithm.
Palmprints can be characterized by their texture and the patterns of that texture dominate in a v... more Palmprints can be characterized by their texture and the patterns of that texture dominate in a vertical direction. Therefore, the energy of the coefficients in the transform domain is more concentrated in the vertical sideband. Using this idea, this paper proposes the characterization of the texture features of the palmprint using zero-crossing signatures based on the dyadic discrete wavelet transform (DWT) to effectively identify an individual. A zero-crossing signature of 4 x 256 was generated from the lower four resolution levels of dyadic DWT in the enrolment process and stored in the database to identify the person in recognition mode. Euclidean distance was determined to find the best fit for query palmprints zero-crossing signature from the dataset. The proposed algorithm was tested on the PolyU dataset containing 6000 multi-spectral images. The proposed algorithm achieved 96.27% accuracy with a lower recognition time of 0.76 seconds. ABSTRAK: Pengesan Tapak Tangan boleh dik...
Leaf a significant part of the plant, produces food using the process called photosynthesis. Leaf... more Leaf a significant part of the plant, produces food using the process called photosynthesis. Leaf disease can cause damage to the entire plant and eventually lowers crop production. Machine learning algorithm for classifying five types of diseases, such as Alternaria leaf diseases, Bacterial Blight, Gray Mildew, Leaf Curl and Myrothecium leaf diseases, is proposed in the proposed study. The classification of diseases needs front face of leafs. This paper proposes an automated image acquisition process using a USB camera interfaced with Raspberry PI SoC. The image is transmitted to host PC for classification of diseases using online web server. Pre-processing of the acquired image by host PC to obtain full leaf, and later classification model based on SVM is used to detect type diseases. Results were checked with a 97% accuracy for the collection of acquired images. Keywords—Plant Disease classification; Support Vector Machine; Graph Cut; Gray-level Co-occurance Matrix
Object identification and segmentation application requires extraction of object in foreground fr... more Object identification and segmentation application requires extraction of object in foreground from the background. In this paper the Bhattacharya distance based probabilistic approach is utilized with an active contour model (ACM) to segment an object from the background. In the proposed approach, the Bhattacharya histogram is calculated on non-linear structure tensor space. Based on the histogram, new formulation of active contour model is proposed to segment images. The results are tested on both color and gray images from the Berkeley image database. The experimental results show that the proposed model is applicable to both color and gray images as well as both texture images and natural images. Again in comparing to the Bhattacharya based ACM in ICA space, the proposed model is able to segment multiple object too.
Virtual motion and pose from images and video can be estimated by detecting body joints and their... more Virtual motion and pose from images and video can be estimated by detecting body joints and their interconnection. The human body has diverse and complicated poses in yoga, making its classification challenging. This study estimates yoga poses from the images using a neural network. Five different yoga poses, viz. downdog, tree, plank, warrior2, and goddess in the form of RGB images are used as the target inputs. The BlazePose model was used to localize the body joints of the yoga poses. It detected a maximum of 33 body joints, referred to as keypoints, covering almost all the body parts. Keypoints achieved from the model are considered as predicted joint locations. True keypoints, as the ground truth body joint for individual yoga poses, are identified manually using the open source image annotation tool named Makesense AI. A detailed analysis of the body joint detection accuracy is proposed in the form of percentage of corrected keypoints (PCK) and percentage of detected joints (P...
Interconnected three-element and four-element wideband MIMO antennas have been proposed for milli... more Interconnected three-element and four-element wideband MIMO antennas have been proposed for millimeter-wave 5G applications by performing numerical computations and carrying out experimental measurements. The antenna structure is realized using Rogers 5880 substrate (εr = 2.2, tan δ = 0.0009), where the radiating element has the shape of an inverted L with a partial ground. The unit element is carefully designed and positioned (by orthogonally rotating the elements) to form three-element (case 1) and four-element (case 2) MIMO antennas. The interconnected ground for both cases is ascertained to increase the practical utilization of the resonator. The proposed MIMO antenna size is (0.95λ × 3λ) for case 1 and (2.01λ × 1.95λ) for case 2 (at the lowest functional frequency). Both the designs give an impedance bandwidth of approximately 26–40 GHz (43%). Moreover, they achieve greater than 15 dB isolation and more than 6 dBi gain with an ECC value lower than 0.02, which meets the MIMO div...
For autonomous navigation, three main functions are essential: finding the location, creating the... more For autonomous navigation, three main functions are essential: finding the location, creating the mapping, and getting the optimum path. Since Human operators study the map and correlate it to aerial pictures to locate target locations, consistent localization and mapping concurrently are challenging tasks. Light detection and ranging (LiDAR) can create a 2-dimensional (2D) map and generate positional data for the indoor area, but it fails in the presence of a dynamic object. The global positioning systems (GPS) data offering precise location tracking in outdoor spaces can tackle the weakness of LiDAR. Therefore, we design a robot operating system (ROS) based vehicular system integrating GPS and LiDAR data. The Inertial Measurement Unit (IMU) is used to make an educated approximation for LiDAR registration. A Rao-Blackwellized particle filter (RBPF) based Gmapping algorithm has been retreated in the proposed system using sensors data for navigation and mapping, where each particle h...
Purpose This paper aims to present, design and implement a novel half-mode substrate integrated w... more Purpose This paper aims to present, design and implement a novel half-mode substrate integrated waveguide (HMSIW)-based narrow bandpass filter, which offers advantages like low insertion loss, compact size and high selectivity. Proposed filter will be used in the K-band automotive radar application. Design/methodology/approach The filtering response in the proposed design is achieved by inserting inductive posts in the HMSIW cavity. Ansoft high frequency structure Simulator (HFSS) is used for the simulation of the proposed structure, which is a three-dimensional full-wave solver using the finite element method (FEM). The proposed filter is fabricated on the dielectric material RT duroid 5,880 with the dielectric constant ɛr = 2.2, dissipation factor t and = 4 × 10–4 and height h = 0.508 mm. Findings Frequency tuning is also carried out by changing the lateral distance between two inductive posts. Moreover, a comparison of the proposed structure with the previously published work is ...
Modeling and Optimization in Science and Technologies, 2015
This chapter presents the application of a new meta-heuristic optimization algorithm called cucko... more This chapter presents the application of a new meta-heuristic optimization algorithm called cuckoo search algorithm (CSA) to solve optimal reactive power dispatch problem (ORPD) of the power system in the presence of wind power plants (WPP). Due to the inclusion of WPP, the ORPD problem becomes a complex combinatorial optimization problem and it has a nonlinear objective function with many local minima, and discontinuous and nonlinear constraint functions. CSA is based on the obligate brood parasitic behavior of some cuckoo species in combination with the Lѐvy flight behavior of some birds and fruit flies. The effectiveness and feasibility of CSA have been tested on a 41-bus WPP test system and the obtained results that have been compared with particle swarm optimization (PSO). Simulation results yield that the CSA converges to better optimal solutions faster than PSO.
Image segmentation is the fundamental step in various video processing applications like object t... more Image segmentation is the fundamental step in various video processing applications like object tracking, pattern recognition, industrial automation, object classification etc. Earlier, low-level curve gradient methods were used for the real time image segmentation where images were processed pixel by pixel. However, the gradient models provide discontinuous edges and are highly sensitive to the noise in the image. High level Mumford-Shah based Active Contour Model (ACM) provides continuous edges and is less sensitive to noise but the complexity in the algorithm and its iterative implementation makes it challenging for the real time applications. This paper investigates the computational complexity and proposes optimum approximation of segmentation with removal of contour initialization. The proposed algorithm is implemented on Xilinx's Zynq ZC702 FPGA board. The removal of contour initialization increased the speed by 10% in software and fixed point implementation on Xilinx provided 80% segmentation accuracy with very low latency of 7.3270 ns per iteration.
Purpose The digital down converter (DDC) is a principal component in modern communication systems... more Purpose The digital down converter (DDC) is a principal component in modern communication systems. The DDC process traditionally entails quadrature down conversion, bandwidth reducing filters and commensurate sample rate reduction. To avoid group delay, distortion linear phase FIR filters are used in the DDC. The filter performance specifications related to deep stopband attenuation, small in-band ripple and narrow transition bandwidth lead to filters with a large number of coefficients. To reduce the computational workload of the filtering process, filtering is often performed as a two-stage process, the first stage being a down sampling Hoegenauer (or cascade-integrated comb) filter and a reduced sample rate FIR filter. An alternative option is an M-Path polyphase partition of a band cantered FIR filter. Even though IIR filters offer reduced workload to implement a specific filtering task, the authors avoid using them because of their poor group delay characteristics. This paper a...
International Journal of Online Engineering (iJOE), 2017
This paper proposes the design and implementation of prototype model to control embroidery machin... more This paper proposes the design and implementation of prototype model to control embroidery machine using single board ARM processor. Wilcom is the software largely used for embroidery design. Designs prepared using Wilcom software are printed and supplied to the machine for further process. The proposed prototype model extracts the design data from this software by avoiding the printing and/or scanning of the user design and the design data are supplied to ARM based prototype embroidery machine for printing. In proposed model, the pencil is used as printing head and A4 size paper is used for the printing. Further to control the head, Use of web server design is also proposed using ARM processor.
This paper embodies the design and development of a compact Coplanar Waveguide (CPW) fed connecte... more This paper embodies the design and development of a compact Coplanar Waveguide (CPW) fed connected ground Multiple-Input-Multiple-Output (MIMO) antennas operating in the sub-millimeter-wave 5G New Radio (NR) n257/n258/n261 bands. The planar geometry leads to a small and compact structure while achieving a wide operating bandwidth, high gain, and better radiation efficiency. The top surface of the antenna comprises a modified CPW in the form of two circular structures that feeds the centrally slotted circular patch. The single antenna structure is arranged in a rotational orthogonal manner forming a 4-port structure. The ground plane on the bottom of a 4-port structure is connected using a circular ring which is carefully optimized for achieving isolation levels>20 dB across the band of interest. The sub-mm-wave resonating 4port antenna achieves a compact size of 24 Â 24 mm 2 , a wide bandwidth of 24.8-44.45 GHz (79.35%), the maximum gain of 8.6dBi, and minimum efficiency of 85% across the bands of interest. The proposed antenna element is fabricated over Rogers 5880 substrate and experimental tests are carried out, where a good correlation between the scattering parameters, transmission
Journal of Automation, Mobile Robotics and Intelligent Systems, 2019
Presently the interaction of robots with human plays an important role in various social applicat... more Presently the interaction of robots with human plays an important role in various social applications. Reliable tracking is an important aspect for the social robots where robots need to follow the moving person. This paper proposes the implementation of automated tracking and real time following algorithm for robotic automation. Occlusion and identity retention are the major challenges in the tracking process. Hence, a feature set based identity retention algorithm is used and integrated with robot operating system. The tracking algorithm is implemented using robot operating system in Linux and using OpenCV. The tracking algorithm achieved 85% accuracy and 72.30% precision. Further analysis of tracking algorithm corresponds to the integration of ROS and OpenCV is presented. The analysis of tracking algorithm concludes that ROS linking required 0.64% more time in comparison with simple OpenCV code based tracking algorithm.
Palmprints can be characterized by their texture and the patterns of that texture dominate in a v... more Palmprints can be characterized by their texture and the patterns of that texture dominate in a vertical direction. Therefore, the energy of the coefficients in the transform domain is more concentrated in the vertical sideband. Using this idea, this paper proposes the characterization of the texture features of the palmprint using zero-crossing signatures based on the dyadic discrete wavelet transform (DWT) to effectively identify an individual. A zero-crossing signature of 4 x 256 was generated from the lower four resolution levels of dyadic DWT in the enrolment process and stored in the database to identify the person in recognition mode. Euclidean distance was determined to find the best fit for query palmprints zero-crossing signature from the dataset. The proposed algorithm was tested on the PolyU dataset containing 6000 multi-spectral images. The proposed algorithm achieved 96.27% accuracy with a lower recognition time of 0.76 seconds. ABSTRAK: Pengesan Tapak Tangan boleh dik...
Leaf a significant part of the plant, produces food using the process called photosynthesis. Leaf... more Leaf a significant part of the plant, produces food using the process called photosynthesis. Leaf disease can cause damage to the entire plant and eventually lowers crop production. Machine learning algorithm for classifying five types of diseases, such as Alternaria leaf diseases, Bacterial Blight, Gray Mildew, Leaf Curl and Myrothecium leaf diseases, is proposed in the proposed study. The classification of diseases needs front face of leafs. This paper proposes an automated image acquisition process using a USB camera interfaced with Raspberry PI SoC. The image is transmitted to host PC for classification of diseases using online web server. Pre-processing of the acquired image by host PC to obtain full leaf, and later classification model based on SVM is used to detect type diseases. Results were checked with a 97% accuracy for the collection of acquired images. Keywords—Plant Disease classification; Support Vector Machine; Graph Cut; Gray-level Co-occurance Matrix
Object identification and segmentation application requires extraction of object in foreground fr... more Object identification and segmentation application requires extraction of object in foreground from the background. In this paper the Bhattacharya distance based probabilistic approach is utilized with an active contour model (ACM) to segment an object from the background. In the proposed approach, the Bhattacharya histogram is calculated on non-linear structure tensor space. Based on the histogram, new formulation of active contour model is proposed to segment images. The results are tested on both color and gray images from the Berkeley image database. The experimental results show that the proposed model is applicable to both color and gray images as well as both texture images and natural images. Again in comparing to the Bhattacharya based ACM in ICA space, the proposed model is able to segment multiple object too.
Uploads
Papers by Hiren Mewada