Pub Date : 2022-12-08DOI: 10.1109/IBSSC56953.2022.10037285
Umar Mohammad, M. Y. Yasin, Fang Tang
Filter circuits from the very beginning have played a vital role in the broader areas of signal processing and communications systems. In the paper, a new paradigm of bilateral filtering has been presented. The scheme presents a new class of filters that have the ability to work on the either side of the device. The working of the presented circuits in this study commensurates to the working of the transceivers. The initial verification of the circuits was done using nodal analysis and then put for simulation verifications. The Active device used to realise the proposed circuits is the translinear circuit scheme of second generation current conveyor-CCCII. Basic filter signals like low-pass, high-pass, band-pass and band-stop responses were perceived and authenticate the theory proposed. The simulation work was done on Hspice tool using the 45nm PTM CMOS technology node. The detailed analysis of the proposed circuits is open for the readers in this study. The power consumption of the proposed circuits is in the range of $100-200mu mathrm{W}$ with $pm$ 1 V rail to rail voltages. The Proposed circuit is expected to get applicable as the major analog front-end circuit in numerous biomedical devices.
{"title":"A New Class of Bilateral SISO-(Single Input Single Output) Active Filters Targeted for Health-Care Applications","authors":"Umar Mohammad, M. Y. Yasin, Fang Tang","doi":"10.1109/IBSSC56953.2022.10037285","DOIUrl":"https://doi.org/10.1109/IBSSC56953.2022.10037285","url":null,"abstract":"Filter circuits from the very beginning have played a vital role in the broader areas of signal processing and communications systems. In the paper, a new paradigm of bilateral filtering has been presented. The scheme presents a new class of filters that have the ability to work on the either side of the device. The working of the presented circuits in this study commensurates to the working of the transceivers. The initial verification of the circuits was done using nodal analysis and then put for simulation verifications. The Active device used to realise the proposed circuits is the translinear circuit scheme of second generation current conveyor-CCCII. Basic filter signals like low-pass, high-pass, band-pass and band-stop responses were perceived and authenticate the theory proposed. The simulation work was done on Hspice tool using the 45nm PTM CMOS technology node. The detailed analysis of the proposed circuits is open for the readers in this study. The power consumption of the proposed circuits is in the range of $100-200mu mathrm{W}$ with $pm$ 1 V rail to rail voltages. The Proposed circuit is expected to get applicable as the major analog front-end circuit in numerous biomedical devices.","PeriodicalId":426897,"journal":{"name":"2022 IEEE Bombay Section Signature Conference (IBSSC)","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115503770","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Robotic arms have a wide variety of applications in various industries. One of the most important industries where robots can be deployed is agriculture. In this paper, the LearnByResearch(LBR) team has attempted to design a robotic arm to pick and place strawberries in a strawberry plantation. The design is based on the typical dimensions of a strawberry plant. The design and structure of the robot have been justified through kinematic analysis and the Finite Element Method. The robot was simulated in MATLAB to generate the workspace. The research also involves the design of a dedicated control system to control the robot.
{"title":"Design and Analysis of Strawberry-Picking Industrial Robotic Arm","authors":"Adnan Amir, Akshat Verma, Abhinav Goswami, Ashish Kabra, Shrikar Nakhye, Sanket Chaudhary","doi":"10.1109/IBSSC56953.2022.10037337","DOIUrl":"https://doi.org/10.1109/IBSSC56953.2022.10037337","url":null,"abstract":"Robotic arms have a wide variety of applications in various industries. One of the most important industries where robots can be deployed is agriculture. In this paper, the LearnByResearch(LBR) team has attempted to design a robotic arm to pick and place strawberries in a strawberry plantation. The design is based on the typical dimensions of a strawberry plant. The design and structure of the robot have been justified through kinematic analysis and the Finite Element Method. The robot was simulated in MATLAB to generate the workspace. The research also involves the design of a dedicated control system to control the robot.","PeriodicalId":426897,"journal":{"name":"2022 IEEE Bombay Section Signature Conference (IBSSC)","volume":"85 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132557743","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-12-08DOI: 10.1109/IBSSC56953.2022.10037437
Pritesh Shah, R. Aravind Sekhar, Margi Shah
Automatic voltage regulator (AVR) systems are critically required in many systems such as hydro, gas, steam turbines and engines. Effective control of an AVR system is complex and challenging due to its unstable open loop response characteristics. The AVR system is required to sustain appropriate voltage levels in spite of fluctuations in the main supply. Hence, the AVR system needs to be primarily designed for stability. This paper presents design implementation of the first and second generations of CRONE controllers (Commande Robuste d'Ordre Non Entier) for an AVR system. CRONE is a non-integer order robust controller which is far more effective than the standard PID controllers as it offers better tuning and flexibility in controlling the AVR process. In the present study, various time-domain performance criteria with and without controller were simulated to validate the accomplishment of AVR CRONE controllers. The proposed control procedure for enhancing AVR systems' efficiency has been shown to be effective and better in terms of improved phase and gain margins.
{"title":"Design and Implementation of CRONE Controller for Automatic Voltage Regulator (AVR) System","authors":"Pritesh Shah, R. Aravind Sekhar, Margi Shah","doi":"10.1109/IBSSC56953.2022.10037437","DOIUrl":"https://doi.org/10.1109/IBSSC56953.2022.10037437","url":null,"abstract":"Automatic voltage regulator (AVR) systems are critically required in many systems such as hydro, gas, steam turbines and engines. Effective control of an AVR system is complex and challenging due to its unstable open loop response characteristics. The AVR system is required to sustain appropriate voltage levels in spite of fluctuations in the main supply. Hence, the AVR system needs to be primarily designed for stability. This paper presents design implementation of the first and second generations of CRONE controllers (Commande Robuste d'Ordre Non Entier) for an AVR system. CRONE is a non-integer order robust controller which is far more effective than the standard PID controllers as it offers better tuning and flexibility in controlling the AVR process. In the present study, various time-domain performance criteria with and without controller were simulated to validate the accomplishment of AVR CRONE controllers. The proposed control procedure for enhancing AVR systems' efficiency has been shown to be effective and better in terms of improved phase and gain margins.","PeriodicalId":426897,"journal":{"name":"2022 IEEE Bombay Section Signature Conference (IBSSC)","volume":"113 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125251762","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-12-08DOI: 10.1109/IBSSC56953.2022.10037545
Ravinder M, Vikram Kulkarni
Advancements in sensors and actuator technology incorporated with data communication are enhancing the standards of control and automation areas for different applications. The Home area network (HAN) is based on IEEE 802.15.4 is considered for applications like Advanced metering infrastructure in Smart grid applications. In this paper, Wireless Sensor Network is considered for HAN. The sensor data is communicated to the sink via Cluster heads (CH). The work proposed in this paper is identifying a proper CH, among the available cluster nodes. Based on the above we are proposing a novel algorithm Energy-Efficient Algorithm for Cluster formation and Cluster Head selection (EEA-CFCHS) for heterogeneous WSN. The proposed methodology takes into account the energy degeneracy threshold value for all different kinds of nodes in a WSN network. Each CH and cluster nodes proceed to the following round based on the threshold value. Each round ends with a calculation of the CH residual energy. The network begins building a new cluster and electing a new CH, if the amount of energy left is less than the threshold value. The energy consumption is lowered as a result, and the stability period is greatly increased. In this algorithm, the simulation in MATLAB shows how to improve the network's lifetime. When compared to the ESRA and P-SEP protocols, the EEA-CFCHS for WSN improves the stability period by 42 %, 72% the network lifespan by 62%, and the network lifetime by 73.16%.
{"title":"Energy-Efficient Algorithm for Cluster Formation and Cluster Head Selection for WSN","authors":"Ravinder M, Vikram Kulkarni","doi":"10.1109/IBSSC56953.2022.10037545","DOIUrl":"https://doi.org/10.1109/IBSSC56953.2022.10037545","url":null,"abstract":"Advancements in sensors and actuator technology incorporated with data communication are enhancing the standards of control and automation areas for different applications. The Home area network (HAN) is based on IEEE 802.15.4 is considered for applications like Advanced metering infrastructure in Smart grid applications. In this paper, Wireless Sensor Network is considered for HAN. The sensor data is communicated to the sink via Cluster heads (CH). The work proposed in this paper is identifying a proper CH, among the available cluster nodes. Based on the above we are proposing a novel algorithm Energy-Efficient Algorithm for Cluster formation and Cluster Head selection (EEA-CFCHS) for heterogeneous WSN. The proposed methodology takes into account the energy degeneracy threshold value for all different kinds of nodes in a WSN network. Each CH and cluster nodes proceed to the following round based on the threshold value. Each round ends with a calculation of the CH residual energy. The network begins building a new cluster and electing a new CH, if the amount of energy left is less than the threshold value. The energy consumption is lowered as a result, and the stability period is greatly increased. In this algorithm, the simulation in MATLAB shows how to improve the network's lifetime. When compared to the ESRA and P-SEP protocols, the EEA-CFCHS for WSN improves the stability period by 42 %, 72% the network lifespan by 62%, and the network lifetime by 73.16%.","PeriodicalId":426897,"journal":{"name":"2022 IEEE Bombay Section Signature Conference (IBSSC)","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116860627","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-12-08DOI: 10.1109/IBSSC56953.2022.10037264
Ajita A. Patil, B. S. Agarkar
Sketch to face recognition automation can play important role in forensic operations. The forensic departments can generate sketches with the help of drawing artists. The resulting sketch images may have difference compared to actual faces in terms of facial parts and expressions. The convolutional neural network (CNN) based method proposed in this paper shows augmentation based sketch and facial expression dataset generation by modifying the public dataset. The generated dataset is thus used to train the VGGFaceNet CNN model and performance is evaluated. The performance of VGGFaceNet model is tested with reference to parameters like accuracy, specificity and sensitivity. The proposed system indicates accuracy of 88% over to other conventional methods such as Local Binary Pattern, Support Vector Machine.
{"title":"VGG FaceNet Based Sketch to Face Recognition with Morphable Model","authors":"Ajita A. Patil, B. S. Agarkar","doi":"10.1109/IBSSC56953.2022.10037264","DOIUrl":"https://doi.org/10.1109/IBSSC56953.2022.10037264","url":null,"abstract":"Sketch to face recognition automation can play important role in forensic operations. The forensic departments can generate sketches with the help of drawing artists. The resulting sketch images may have difference compared to actual faces in terms of facial parts and expressions. The convolutional neural network (CNN) based method proposed in this paper shows augmentation based sketch and facial expression dataset generation by modifying the public dataset. The generated dataset is thus used to train the VGGFaceNet CNN model and performance is evaluated. The performance of VGGFaceNet model is tested with reference to parameters like accuracy, specificity and sensitivity. The proposed system indicates accuracy of 88% over to other conventional methods such as Local Binary Pattern, Support Vector Machine.","PeriodicalId":426897,"journal":{"name":"2022 IEEE Bombay Section Signature Conference (IBSSC)","volume":"292 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115603106","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-12-08DOI: 10.1109/IBSSC56953.2022.10037566
Lie Xu, Yangyang Zhai, Daxiong Ji, Zhangying Ye, Songming Zhu, Yao Zhang, Guan-nan Li
In this paper, we propose a new guidance law based on Line-of-Sight (LOS) guidance, i.e., Line-of-Sight-With-Sideway (LOSWS) for autonomous underwater vehicle (AUV) with sideway motion in small water area, such as fish pond. It is able to maximize the use of kinetic characteristic in sideway and realize a more precise path following. In addition, to overcome the demerit of course angle variation which is required to “through” π-axis or -π-axis, we also propose a reference point course angle updating algorithm to implement the optimal steering path. The simulation results in MATLAB validate that the proposed methods can achieve a better performance compare to conventional LOS and PID-LOS guidance.
{"title":"A new path following method of AUV using line-of-sight with sideway","authors":"Lie Xu, Yangyang Zhai, Daxiong Ji, Zhangying Ye, Songming Zhu, Yao Zhang, Guan-nan Li","doi":"10.1109/IBSSC56953.2022.10037566","DOIUrl":"https://doi.org/10.1109/IBSSC56953.2022.10037566","url":null,"abstract":"In this paper, we propose a new guidance law based on Line-of-Sight (LOS) guidance, i.e., Line-of-Sight-With-Sideway (LOSWS) for autonomous underwater vehicle (AUV) with sideway motion in small water area, such as fish pond. It is able to maximize the use of kinetic characteristic in sideway and realize a more precise path following. In addition, to overcome the demerit of course angle variation which is required to “through” π-axis or -π-axis, we also propose a reference point course angle updating algorithm to implement the optimal steering path. The simulation results in MATLAB validate that the proposed methods can achieve a better performance compare to conventional LOS and PID-LOS guidance.","PeriodicalId":426897,"journal":{"name":"2022 IEEE Bombay Section Signature Conference (IBSSC)","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125520335","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-12-08DOI: 10.1109/IBSSC56953.2022.10037503
Suchitra A. Patil, Chandrakant J. Gaikwad
The problem of depth estimation using monocular images is very challenging contrasted to methods for estimating depth that involve several pictures, like stereo depth perception. Previous studies in this field have usually focused on utilizing geometrical priors or relied on other data collection techniques. Various machine learning methods, notably deep convolutional neural networks (CNN) integrated with artificial intelligence (AI) approaches, have recently produced new marks for a variety of visual applications. In this paper, a convolution neural network is used for estimating high-resolution depth image. We have used a pretrained model DenseNet-169 which is trained on ImageNet. The encoder-decoder model used is a simple Unet model, used in biomedical image analysis. The proposed model is more accurate and efficient with reduced complexity in terms of the fewer parameters used for training. This model is also noteworthy when comparing a state of art and qualitatively it performs well and captures better edges and corners of the depth map, which is the most important factor in the depth estimation.
{"title":"Depth Estimation of Monocular Images using Transfer Learning based Unet Model","authors":"Suchitra A. Patil, Chandrakant J. Gaikwad","doi":"10.1109/IBSSC56953.2022.10037503","DOIUrl":"https://doi.org/10.1109/IBSSC56953.2022.10037503","url":null,"abstract":"The problem of depth estimation using monocular images is very challenging contrasted to methods for estimating depth that involve several pictures, like stereo depth perception. Previous studies in this field have usually focused on utilizing geometrical priors or relied on other data collection techniques. Various machine learning methods, notably deep convolutional neural networks (CNN) integrated with artificial intelligence (AI) approaches, have recently produced new marks for a variety of visual applications. In this paper, a convolution neural network is used for estimating high-resolution depth image. We have used a pretrained model DenseNet-169 which is trained on ImageNet. The encoder-decoder model used is a simple Unet model, used in biomedical image analysis. The proposed model is more accurate and efficient with reduced complexity in terms of the fewer parameters used for training. This model is also noteworthy when comparing a state of art and qualitatively it performs well and captures better edges and corners of the depth map, which is the most important factor in the depth estimation.","PeriodicalId":426897,"journal":{"name":"2022 IEEE Bombay Section Signature Conference (IBSSC)","volume":"220 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127528043","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In this paper, a design was proposed and tested for the implementation of an auxiliary security system that is used along with the standard CCTV setup that is present in most commercial settings. The design mainly focuses on providing a layer of automation in a cost-effective manner. The bot, as a single unit or a team of up to 4 units, makes use of technologies like perception, path following, IoT-based Twitter alerts, and camouflage to patrol a typical commercial setting such as a retail store at night.
{"title":"Design and Implementation of an IoT Based Patrol Robot","authors":"Adnan Amir, Aaryaman Chandgothia, Moksh Goel, Dattatray Sawant, Nirmal Thakur","doi":"10.1109/IBSSC56953.2022.10037325","DOIUrl":"https://doi.org/10.1109/IBSSC56953.2022.10037325","url":null,"abstract":"In this paper, a design was proposed and tested for the implementation of an auxiliary security system that is used along with the standard CCTV setup that is present in most commercial settings. The design mainly focuses on providing a layer of automation in a cost-effective manner. The bot, as a single unit or a team of up to 4 units, makes use of technologies like perception, path following, IoT-based Twitter alerts, and camouflage to patrol a typical commercial setting such as a retail store at night.","PeriodicalId":426897,"journal":{"name":"2022 IEEE Bombay Section Signature Conference (IBSSC)","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124136029","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-12-08DOI: 10.1109/IBSSC56953.2022.10037294
Devarsh Patel
The globalization of Logistics and Supply Chain has escalated the demand for efficiency in its Information Systems and Security. With supply chain systems now crossing the boundaries, it has become increasingly important to manage the vast amount of data it generates. Conventional supply chain system has a lot of manual tasks to be done which is time consuming and makes it prone to more errors. Supply chain systems may leverage blockchain technology (BT) to streamline supply chain operations in a trustworthy and secure way. The integration of the blockchain into supply chain systems looks promising but has a lot of challenges right now. The challenges are primarily because of the relatively new technology with lack of standards and guidelines, risk of adoption, and other risks like cost, time and information sharing. This paper investigates the challenges of logistics and supply chain and its solutions using BT followed by analysis of their pros and cons.
{"title":"Blockchain for Challenges of Logistics and Supply Chain Information System","authors":"Devarsh Patel","doi":"10.1109/IBSSC56953.2022.10037294","DOIUrl":"https://doi.org/10.1109/IBSSC56953.2022.10037294","url":null,"abstract":"The globalization of Logistics and Supply Chain has escalated the demand for efficiency in its Information Systems and Security. With supply chain systems now crossing the boundaries, it has become increasingly important to manage the vast amount of data it generates. Conventional supply chain system has a lot of manual tasks to be done which is time consuming and makes it prone to more errors. Supply chain systems may leverage blockchain technology (BT) to streamline supply chain operations in a trustworthy and secure way. The integration of the blockchain into supply chain systems looks promising but has a lot of challenges right now. The challenges are primarily because of the relatively new technology with lack of standards and guidelines, risk of adoption, and other risks like cost, time and information sharing. This paper investigates the challenges of logistics and supply chain and its solutions using BT followed by analysis of their pros and cons.","PeriodicalId":426897,"journal":{"name":"2022 IEEE Bombay Section Signature Conference (IBSSC)","volume":"102 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126342105","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-12-08DOI: 10.1109/IBSSC56953.2022.10037429
R. Kale, S. Shitole
Quality and safety are important factors in the food industry. In recent years automatic visual inspection technology has become more potential and important for fruit grading applications. This is because quality is an important factor for consumers and so essential for the market. This paper focuses on a comparative study of deep learning optimizers for pomegranate fruit quality grading. It plays an important role in maximizing the efficiency of the neural network model. Optimizers are mathematical functions or algorithms which are dependent on various parameters of the model i.e., weights and biases. This paper presents the performances of the various deep learning optimizers for pomegranate fruit quality grading. The dataset used for this study is named as Pomegranate Fruit dataset from the Kaggle dataset. Dataset has three grades G1, G2, and G3. Each grade is having four internal quality labels and has 90 images in it. Training is done using SGD, Adadelta, Adagrad, RMSprop, and Adam optimizers. This study helped in analyzing better optimizer and identifying the need for overall improvement in performance of the optimization.
{"title":"Deep learning optimizer performance analysis for pomegranate fruit quality gradation","authors":"R. Kale, S. Shitole","doi":"10.1109/IBSSC56953.2022.10037429","DOIUrl":"https://doi.org/10.1109/IBSSC56953.2022.10037429","url":null,"abstract":"Quality and safety are important factors in the food industry. In recent years automatic visual inspection technology has become more potential and important for fruit grading applications. This is because quality is an important factor for consumers and so essential for the market. This paper focuses on a comparative study of deep learning optimizers for pomegranate fruit quality grading. It plays an important role in maximizing the efficiency of the neural network model. Optimizers are mathematical functions or algorithms which are dependent on various parameters of the model i.e., weights and biases. This paper presents the performances of the various deep learning optimizers for pomegranate fruit quality grading. The dataset used for this study is named as Pomegranate Fruit dataset from the Kaggle dataset. Dataset has three grades G1, G2, and G3. Each grade is having four internal quality labels and has 90 images in it. Training is done using SGD, Adadelta, Adagrad, RMSprop, and Adam optimizers. This study helped in analyzing better optimizer and identifying the need for overall improvement in performance of the optimization.","PeriodicalId":426897,"journal":{"name":"2022 IEEE Bombay Section Signature Conference (IBSSC)","volume":"32 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125710458","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：