Sangit Saha - Academia.edu (original) (raw)
Papers by Sangit Saha
2020 4th International Conference on Electronics, Communication and Aerospace Technology (ICECA), 2020
This paper presents the design and development of an autonomous intelligent robot on the Robotic ... more This paper presents the design and development of an autonomous intelligent robot on the Robotic Operating System (ROS) with automated navigation to monitor the grass health in turfs by spraying fertilizers or nutrients for an early detection of stress in the leaves through chlorophyll index estimation and further the disease detection is performed by processing the images snapped by the bot. The accuracy in regards to automated navigation of the bot is in the order of twenty centimeters in outdoor operations, while the accuracy of stress detection in leaves has been about 90% on field trials, which makes the system very promising for a commercial deployment.
2021 2nd International Conference on Secure Cyber Computing and Communications (ICSCCC), 2021
In the field of agriculture, there is need of recognizing as well as classifying diseases from le... more In the field of agriculture, there is need of recognizing as well as classifying diseases from leaf images that are taken from plant. Finding the diseases of paddy leaf by making use of image processing mechanism would reduce the reliance on farmers in order to save the product related to agricultural activity. This research paper is finding and categorizing the disease in paddy leaf with the help of CNN with integration of edge detection mechanism. The proposed work is focusing on the classification of paddy leaf on the basis of the brown spot, bacterial blight, blast disease and sheath rot after spot detection. However, there have been some previous researches to fulfill such objectives but the proposed work used edge detection mechanism to reduce the time consumption as well as space consumption.Leaves of the rice plant have been captured from the field for the normal, brown spot, bacterial blight, blast diseases and sheath rot. During the pre-processing operation RGB images have been converted in HSV images. Image processing is made on the basis of hue and saturation. The portions of binary graphical contents have been captured to split the infected and non-infected portion. A clustering mechanism has been used for segmentation of the infected portion. However, there are several existing researches that have classified diseases with the support of optimized DNN_JOA (Deep Neural Network with Jaya Optimization Algorithm). Time and space consumption are the major issue in those researches. There is need to provide the solution to improve the performance and space consumption thus proposed work is making use of deep neural network with integration of canny edge detection. In this paper, pattern detection using edge based CNN (convolution neural network) algorithm is proposed.
Plug-in hybrid electric vehicles (PHEVs) have a conventional internal combustion (IC) engine runn... more Plug-in hybrid electric vehicles (PHEVs) have a conventional internal combustion (IC) engine running on fossil fuels as well as an electric motor that gets supply from a battery which can be charged using external supply like electric vehicles (EVs). Using renewable energy sources like fuel cells or solar photovoltaics (PVs) for charging the battery, EVs and PHEVs can become even cleaner as far as the environment is concerned. These energy sources generate a low voltage which is limited to around 40–50 V due to practical constraints, while batteries that power EVs and PHEVs need a charging voltage of around 350 V. This paper presents a novel naturally clamped, isolated, DC–DC converter which has a voltage doubler rectifier at the output end which converts 12 V supply from renewable sources to 350 V to enable battery charging.
Information and Communication Technology for Competitive Strategies (ICTCS 2020), 2021
Grains are generally stored in warehouses after harvesting often for months. These stored grains ... more Grains are generally stored in warehouses after harvesting often for months. These stored grains are distributed to consumers through distribution channels as and when required. During such long storage, grains are affected by pests/insects and there can be considerable amount of storage losses. Pest infestation is highly affected by the temperature and relative humidity of the warehouse. In order to control such loss, a process called fumigation is performed to kill the pests by means of pesticides. The operation is hazardous and is carried out in buildings, soil and grain warehouses. Fumigation is also done during the process of export and import of goods to prevent the transfer of unusual organisms (Saeung et al., 2018 5th International conference on industrial engineering and applications (ICIEA), Singapore, 2018, pp 179–183 ( Saeung, P., Santalunai, S., Thosdeekoraphat, T., Thongsopa, C.: Improved efficiency of insect pest control system by SSPA. In: 2018 5th International Conference on Industrial Engineering and Applications (ICIEA), Singapore, pp. 179–183 (2018). 10.1109/IEA.2018.8387092)). Phosphine (PH3) gas is mostly used presently to carry out fumigation in warehouses. This is important to know the performance of the fumigation process. But phosphine being highly toxic in nature, the warehouse is kept sealed during the entire fumigation period. Hence, the efficacy of fumigation during the process cannot be known. This paper describes an indirect way of checking the effectiveness of fumigation by measuring the concentration of fumigant in real time. A device named FUMON (fumigation monitoring system) is designed and developed for the said purpose. FUMON device automatically sniffs fumigant from the grain stack under fumigation at certain predefined interval of time. A PC/laptop connected to the FUMON device continuously plots the concentration (C) against time (T) and calculates the CT product value at every instance of time. The CT product value indicates the effectiveness of ongoing fumigation and seeing the value, the warehouse manager may take decision to lengthen or shorten the fumigation. Validation of the developed FUMON device is also done at Indian Institute of Food Processing Technology (IIFPT), Thanjavur.
2021 12th International Conference on Computing Communication and Networking Technologies (ICCCNT), 2021
The agriculture industry faces huge economic losses due to bacterial, viral or fungal infections ... more The agriculture industry faces huge economic losses due to bacterial, viral or fungal infections in the crops due to which farmers lose 15 to 20% of their total profit every year. India is the second largest producer of rice and a leading exporter of the same in the global market. Thus, early detection of diseases in essential crops is a significant area of research in order to prevent further damage to them. The widespread development of Deep Learning makes it possible to achieve the goal of disease detection in crops. The novelty of this work is early detection of Brown spot disease in rice paddy using Convolution Neural Networks. The area of the disease affected was also found to optimize the usage of fertilizers. This work makes use of Image recognition and preprocessing algorithm based on real time data. Data preprocessing and feature extraction has been done using a self-designed image-processing tool. Tensor flow and Keras framework has been implemented on both training and testing data which was collected manually from rice fields. The proposed model achieved an accuracy of 97.32%.
2017 International Conference on Energy, Communication, Data Analytics and Soft Computing (ICECDS), 2017
Electric vehicles and plug-in hybrid electric vehicles are two concepts that are gaining populari... more Electric vehicles and plug-in hybrid electric vehicles are two concepts that are gaining popularity in order to reduce our dependence on fossil fuels. In this paper, a full bridge, current fed, interleaved, isolated DC-DC converter is proposed for battery charging of electric vehicles from solar PV or fuel cells. It boosts voltage of around 36 V to 270 V. Interleaving leads to lesser input current ripple, a factor crucial for sources like solar PVs and fuel cells and lesser stress on switches. The concept of soft switching is used to reduce switching losses. Simulation results are presented for a power level of 3.45 kW. A scaled down, 200 W hardware model of the circuit was built and tested. A closed loop feedback system was also designed using k factor method to provide constant output voltage.
International Journal of Vehicle Structures and Systems, 2016
In this a paper, a current fed, interleaved, high gain, DC-DC converter is proposed for fuel cell... more In this a paper, a current fed, interleaved, high gain, DC-DC converter is proposed for fuel cell applications. The converter also provides electrical isolation between the load and the source by using a transformer. The input features two current fed, full bridge inverters in parallel while the output features two full bridge diode rectifiers in series. By using this topology, the high input current is shared between the two inverters. These enables the use of lower current rating semiconductor devices, reduces switching stresses and reduces the size of magnetic components. It also results in reducing the input current ripple and the output voltage ripple.
2017 International Conference on Energy, Communication, Data Analytics and Soft Computing (ICECDS), 2017
This project entails the design of a low voltage DC microgrid system for rural electrification in... more This project entails the design of a low voltage DC microgrid system for rural electrification in South Africa. Solar energy is freely available, environmental friendly and it is considered as a promising power generating source due to its availability and topological advantages for local power generation. Off-grid solar systems are perceived to be a viable means of power delivery to households in rural outlying areas in South Africa as solar panels can be used almost anywhere in the country. The design presented in this paper is based on the power demand estimation, photovoltaic panel selection, battery sizing and wire selection for the distribution system.
2020 4th International Conference on Electronics, Communication and Aerospace Technology (ICECA), 2020
This paper presents the design and development of an autonomous intelligent robot on the Robotic ... more This paper presents the design and development of an autonomous intelligent robot on the Robotic Operating System (ROS) with automated navigation to monitor the grass health in turfs by spraying fertilizers or nutrients for an early detection of stress in the leaves through chlorophyll index estimation and further the disease detection is performed by processing the images snapped by the bot. The accuracy in regards to automated navigation of the bot is in the order of twenty centimeters in outdoor operations, while the accuracy of stress detection in leaves has been about 90% on field trials, which makes the system very promising for a commercial deployment.
2021 2nd International Conference on Secure Cyber Computing and Communications (ICSCCC), 2021
In the field of agriculture, there is need of recognizing as well as classifying diseases from le... more In the field of agriculture, there is need of recognizing as well as classifying diseases from leaf images that are taken from plant. Finding the diseases of paddy leaf by making use of image processing mechanism would reduce the reliance on farmers in order to save the product related to agricultural activity. This research paper is finding and categorizing the disease in paddy leaf with the help of CNN with integration of edge detection mechanism. The proposed work is focusing on the classification of paddy leaf on the basis of the brown spot, bacterial blight, blast disease and sheath rot after spot detection. However, there have been some previous researches to fulfill such objectives but the proposed work used edge detection mechanism to reduce the time consumption as well as space consumption.Leaves of the rice plant have been captured from the field for the normal, brown spot, bacterial blight, blast diseases and sheath rot. During the pre-processing operation RGB images have been converted in HSV images. Image processing is made on the basis of hue and saturation. The portions of binary graphical contents have been captured to split the infected and non-infected portion. A clustering mechanism has been used for segmentation of the infected portion. However, there are several existing researches that have classified diseases with the support of optimized DNN_JOA (Deep Neural Network with Jaya Optimization Algorithm). Time and space consumption are the major issue in those researches. There is need to provide the solution to improve the performance and space consumption thus proposed work is making use of deep neural network with integration of canny edge detection. In this paper, pattern detection using edge based CNN (convolution neural network) algorithm is proposed.
Plug-in hybrid electric vehicles (PHEVs) have a conventional internal combustion (IC) engine runn... more Plug-in hybrid electric vehicles (PHEVs) have a conventional internal combustion (IC) engine running on fossil fuels as well as an electric motor that gets supply from a battery which can be charged using external supply like electric vehicles (EVs). Using renewable energy sources like fuel cells or solar photovoltaics (PVs) for charging the battery, EVs and PHEVs can become even cleaner as far as the environment is concerned. These energy sources generate a low voltage which is limited to around 40–50 V due to practical constraints, while batteries that power EVs and PHEVs need a charging voltage of around 350 V. This paper presents a novel naturally clamped, isolated, DC–DC converter which has a voltage doubler rectifier at the output end which converts 12 V supply from renewable sources to 350 V to enable battery charging.
Information and Communication Technology for Competitive Strategies (ICTCS 2020), 2021
Grains are generally stored in warehouses after harvesting often for months. These stored grains ... more Grains are generally stored in warehouses after harvesting often for months. These stored grains are distributed to consumers through distribution channels as and when required. During such long storage, grains are affected by pests/insects and there can be considerable amount of storage losses. Pest infestation is highly affected by the temperature and relative humidity of the warehouse. In order to control such loss, a process called fumigation is performed to kill the pests by means of pesticides. The operation is hazardous and is carried out in buildings, soil and grain warehouses. Fumigation is also done during the process of export and import of goods to prevent the transfer of unusual organisms (Saeung et al., 2018 5th International conference on industrial engineering and applications (ICIEA), Singapore, 2018, pp 179–183 ( Saeung, P., Santalunai, S., Thosdeekoraphat, T., Thongsopa, C.: Improved efficiency of insect pest control system by SSPA. In: 2018 5th International Conference on Industrial Engineering and Applications (ICIEA), Singapore, pp. 179–183 (2018). 10.1109/IEA.2018.8387092)). Phosphine (PH3) gas is mostly used presently to carry out fumigation in warehouses. This is important to know the performance of the fumigation process. But phosphine being highly toxic in nature, the warehouse is kept sealed during the entire fumigation period. Hence, the efficacy of fumigation during the process cannot be known. This paper describes an indirect way of checking the effectiveness of fumigation by measuring the concentration of fumigant in real time. A device named FUMON (fumigation monitoring system) is designed and developed for the said purpose. FUMON device automatically sniffs fumigant from the grain stack under fumigation at certain predefined interval of time. A PC/laptop connected to the FUMON device continuously plots the concentration (C) against time (T) and calculates the CT product value at every instance of time. The CT product value indicates the effectiveness of ongoing fumigation and seeing the value, the warehouse manager may take decision to lengthen or shorten the fumigation. Validation of the developed FUMON device is also done at Indian Institute of Food Processing Technology (IIFPT), Thanjavur.
2021 12th International Conference on Computing Communication and Networking Technologies (ICCCNT), 2021
The agriculture industry faces huge economic losses due to bacterial, viral or fungal infections ... more The agriculture industry faces huge economic losses due to bacterial, viral or fungal infections in the crops due to which farmers lose 15 to 20% of their total profit every year. India is the second largest producer of rice and a leading exporter of the same in the global market. Thus, early detection of diseases in essential crops is a significant area of research in order to prevent further damage to them. The widespread development of Deep Learning makes it possible to achieve the goal of disease detection in crops. The novelty of this work is early detection of Brown spot disease in rice paddy using Convolution Neural Networks. The area of the disease affected was also found to optimize the usage of fertilizers. This work makes use of Image recognition and preprocessing algorithm based on real time data. Data preprocessing and feature extraction has been done using a self-designed image-processing tool. Tensor flow and Keras framework has been implemented on both training and testing data which was collected manually from rice fields. The proposed model achieved an accuracy of 97.32%.
2017 International Conference on Energy, Communication, Data Analytics and Soft Computing (ICECDS), 2017
Electric vehicles and plug-in hybrid electric vehicles are two concepts that are gaining populari... more Electric vehicles and plug-in hybrid electric vehicles are two concepts that are gaining popularity in order to reduce our dependence on fossil fuels. In this paper, a full bridge, current fed, interleaved, isolated DC-DC converter is proposed for battery charging of electric vehicles from solar PV or fuel cells. It boosts voltage of around 36 V to 270 V. Interleaving leads to lesser input current ripple, a factor crucial for sources like solar PVs and fuel cells and lesser stress on switches. The concept of soft switching is used to reduce switching losses. Simulation results are presented for a power level of 3.45 kW. A scaled down, 200 W hardware model of the circuit was built and tested. A closed loop feedback system was also designed using k factor method to provide constant output voltage.
International Journal of Vehicle Structures and Systems, 2016
In this a paper, a current fed, interleaved, high gain, DC-DC converter is proposed for fuel cell... more In this a paper, a current fed, interleaved, high gain, DC-DC converter is proposed for fuel cell applications. The converter also provides electrical isolation between the load and the source by using a transformer. The input features two current fed, full bridge inverters in parallel while the output features two full bridge diode rectifiers in series. By using this topology, the high input current is shared between the two inverters. These enables the use of lower current rating semiconductor devices, reduces switching stresses and reduces the size of magnetic components. It also results in reducing the input current ripple and the output voltage ripple.
2017 International Conference on Energy, Communication, Data Analytics and Soft Computing (ICECDS), 2017
This project entails the design of a low voltage DC microgrid system for rural electrification in... more This project entails the design of a low voltage DC microgrid system for rural electrification in South Africa. Solar energy is freely available, environmental friendly and it is considered as a promising power generating source due to its availability and topological advantages for local power generation. Off-grid solar systems are perceived to be a viable means of power delivery to households in rural outlying areas in South Africa as solar panels can be used almost anywhere in the country. The design presented in this paper is based on the power demand estimation, photovoltaic panel selection, battery sizing and wire selection for the distribution system.