Pub Date : 2023-05-04DOI: 10.1109/ICAAIC56838.2023.10140728
Ashish Rajanand, Pradeep Singh
Recent developments in computing technology have resulted in the continuous accumulation of enormous volumes of stock data and information. Due to the market's volatile, unpredictable, and non-stationary nature, analyzing stock market movements and price behavior is extremely difficult. In this study, stock prediction model is proposed using a recurrent neural network with depthwise separable convolution for smoothing the prediction. Depthwise separable convolution is used in the proposed model to improve feature extraction. Extracted features are provided in sequential LSTM to forecast future price of the stock. The proposed model., Depth-wise Separable CNN with Sequential LSTM (DWCNN-SLSTM) is evaluated on S&P 500, HSI, CSI300, and Nikkei 225 datasets. The proposed model outperforms the existing and achieved MAPEof 0.4734, 0.5051, 0.4865, and 0.4776 on S&P500, HSI, CSI300, and Nikkei 225 respectively.
{"title":"Stock Price Prediction using Depthwise Pointwise CNN with Sequential LSTM","authors":"Ashish Rajanand, Pradeep Singh","doi":"10.1109/ICAAIC56838.2023.10140728","DOIUrl":"https://doi.org/10.1109/ICAAIC56838.2023.10140728","url":null,"abstract":"Recent developments in computing technology have resulted in the continuous accumulation of enormous volumes of stock data and information. Due to the market's volatile, unpredictable, and non-stationary nature, analyzing stock market movements and price behavior is extremely difficult. In this study, stock prediction model is proposed using a recurrent neural network with depthwise separable convolution for smoothing the prediction. Depthwise separable convolution is used in the proposed model to improve feature extraction. Extracted features are provided in sequential LSTM to forecast future price of the stock. The proposed model., Depth-wise Separable CNN with Sequential LSTM (DWCNN-SLSTM) is evaluated on S&P 500, HSI, CSI300, and Nikkei 225 datasets. The proposed model outperforms the existing and achieved MAPEof 0.4734, 0.5051, 0.4865, and 0.4776 on S&P500, HSI, CSI300, and Nikkei 225 respectively.","PeriodicalId":267906,"journal":{"name":"2023 2nd International Conference on Applied Artificial Intelligence and Computing (ICAAIC)","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117020100","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 : 2023-05-04DOI: 10.1109/ICAAIC56838.2023.10141290
V. Kanagaraj, G. Nareshbabu, DN. Chandni, Jaswant Kumar, Sankar R. Krithik
The proposed system is a process which could double or even triple the rate of crop growth and cultivation if done efficiently. It involves basic water flow motors, soil nutrient sensors (will vary based on the crop cultivated), soil pH sensor and an IoT based controller (Arduino nano) integrating them together. The system will allow for the best delivery of water and nutrients to the crop with minimal losses, hence significantly improving the rate of growth and significantly reducing the area of cultivation. Water streams will be passed directly through the plant roots which only grow to the required lengths instead of water streams or sprinklers in the soil which may cause severe water and nutrient losses as mentioned. Hydroponics farms are estimated to have an increase in crop cultivation and production by around 110 tons (160 tons to 270 tons). This is paired with a 90% improvement in water saving when compared to present systems. An estimated area of $10 mathrm{x} 10 mathrm{x} 10$ meters would be able to produce a crop yield equivalent to 1 acre of conventional agriculture when using hydroponics.
{"title":"Design and Development of an Automated Hydroponics System based on IoT with Data Logging","authors":"V. Kanagaraj, G. Nareshbabu, DN. Chandni, Jaswant Kumar, Sankar R. Krithik","doi":"10.1109/ICAAIC56838.2023.10141290","DOIUrl":"https://doi.org/10.1109/ICAAIC56838.2023.10141290","url":null,"abstract":"The proposed system is a process which could double or even triple the rate of crop growth and cultivation if done efficiently. It involves basic water flow motors, soil nutrient sensors (will vary based on the crop cultivated), soil pH sensor and an IoT based controller (Arduino nano) integrating them together. The system will allow for the best delivery of water and nutrients to the crop with minimal losses, hence significantly improving the rate of growth and significantly reducing the area of cultivation. Water streams will be passed directly through the plant roots which only grow to the required lengths instead of water streams or sprinklers in the soil which may cause severe water and nutrient losses as mentioned. Hydroponics farms are estimated to have an increase in crop cultivation and production by around 110 tons (160 tons to 270 tons). This is paired with a 90% improvement in water saving when compared to present systems. An estimated area of $10 mathrm{x} 10 mathrm{x} 10$ meters would be able to produce a crop yield equivalent to 1 acre of conventional agriculture when using hydroponics.","PeriodicalId":267906,"journal":{"name":"2023 2nd International Conference on Applied Artificial Intelligence and Computing (ICAAIC)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128751247","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 : 2023-05-04DOI: 10.1109/ICAAIC56838.2023.10140400
S. Vijayashaarathi, V. Tamilselvam, K. Saranya, J. Harirajkumar, L. Satheeskumar
The modern world, Digital electronics systems are compact and faster. But, the major problem of these systems are power dissipation. The Power dissipation have different variants such as a static power, dynamic power, short circuit and leakage current dissipation. In VLSI Design, the power consumption plays an important role. In order to minimize the power dissipation there are many different low power methodologies are used such as a multi-Vth method, clock gating and reversible logic gate method. The major advantages of a circuit designing using a reversible logic gates will be compatible with an obtainable resources and the reversible Gates have a zero heat dissipation. The Arithmetic and Logical Unit is fundamental part of a computing systems. This paper, presents a Design of low garbage Reversible Arithmetic and logical unit design for computing system and the design includes Adder, subtractor and Multiplier blocks. The functionality of a design performance, trash outputs, Quantum cost are analysed. The proposed design has a 11 trash outputs and 57 quantum costs. The design is coded on Verilog HDL and synthesized, simulated by a Xilinx software.
{"title":"Optimized Arithmetic and Logical Unit Design using Reversible Logic Gates","authors":"S. Vijayashaarathi, V. Tamilselvam, K. Saranya, J. Harirajkumar, L. Satheeskumar","doi":"10.1109/ICAAIC56838.2023.10140400","DOIUrl":"https://doi.org/10.1109/ICAAIC56838.2023.10140400","url":null,"abstract":"The modern world, Digital electronics systems are compact and faster. But, the major problem of these systems are power dissipation. The Power dissipation have different variants such as a static power, dynamic power, short circuit and leakage current dissipation. In VLSI Design, the power consumption plays an important role. In order to minimize the power dissipation there are many different low power methodologies are used such as a multi-Vth method, clock gating and reversible logic gate method. The major advantages of a circuit designing using a reversible logic gates will be compatible with an obtainable resources and the reversible Gates have a zero heat dissipation. The Arithmetic and Logical Unit is fundamental part of a computing systems. This paper, presents a Design of low garbage Reversible Arithmetic and logical unit design for computing system and the design includes Adder, subtractor and Multiplier blocks. The functionality of a design performance, trash outputs, Quantum cost are analysed. The proposed design has a 11 trash outputs and 57 quantum costs. The design is coded on Verilog HDL and synthesized, simulated by a Xilinx software.","PeriodicalId":267906,"journal":{"name":"2023 2nd International Conference on Applied Artificial Intelligence and Computing (ICAAIC)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129919396","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 : 2023-05-04DOI: 10.1109/ICAAIC56838.2023.10140516
Anjali Sagar Jangde, Dilip Singh Sisodia
Epilepsy affects 1% of the population across all age groups, making it the fourth most dangerous brain disorder diagnosed worldwide. The seizures, limited to a specific area of the brain and affecting up to 60% of epileptic patients, can be diagnosed using an intracranial electroencephalogram (iEEG). However, identifying the epileptic focal channel using iEEG is time-taking and labor-intensive. An automated approach is required to classify both focal and non-focal iEEG signals. Although various machine learning models have been developed using multiple wavelets to address this issue, they have increased model complexity. Deep learning models, which automatically extract features and produce accurate classification, were therefore developed. However, previous attempts using deep learning models were computationally intensive and had unsatisfactory results. To address this issue, in this research, a one-dimensional convolutional neural network (1D-CNN) is proposed, which can directly extract features from the raw iEEG signals of focal and non-focal seizures. Compared to other deep-learning methods, the proposed model significantly reduces the number of parameters. With a classification accuracy of 94%, the model successfully differentiated between the focal and non-focal epileptic iEEG signals.
{"title":"Automated Classification of Focal and Non-focal Epileptic iEEG Signals using 1D-Convolutional Neural Network","authors":"Anjali Sagar Jangde, Dilip Singh Sisodia","doi":"10.1109/ICAAIC56838.2023.10140516","DOIUrl":"https://doi.org/10.1109/ICAAIC56838.2023.10140516","url":null,"abstract":"Epilepsy affects 1% of the population across all age groups, making it the fourth most dangerous brain disorder diagnosed worldwide. The seizures, limited to a specific area of the brain and affecting up to 60% of epileptic patients, can be diagnosed using an intracranial electroencephalogram (iEEG). However, identifying the epileptic focal channel using iEEG is time-taking and labor-intensive. An automated approach is required to classify both focal and non-focal iEEG signals. Although various machine learning models have been developed using multiple wavelets to address this issue, they have increased model complexity. Deep learning models, which automatically extract features and produce accurate classification, were therefore developed. However, previous attempts using deep learning models were computationally intensive and had unsatisfactory results. To address this issue, in this research, a one-dimensional convolutional neural network (1D-CNN) is proposed, which can directly extract features from the raw iEEG signals of focal and non-focal seizures. Compared to other deep-learning methods, the proposed model significantly reduces the number of parameters. With a classification accuracy of 94%, the model successfully differentiated between the focal and non-focal epileptic iEEG signals.","PeriodicalId":267906,"journal":{"name":"2023 2nd International Conference on Applied Artificial Intelligence and Computing (ICAAIC)","volume":"88 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126812652","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 : 2023-05-04DOI: 10.1109/ICAAIC56838.2023.10140866
W. Auccahuasi, Oscar Linares, K. Urbano, K. Rojas, Gabriel Aiquipa, Tamara Pando-Ezcurra
Currently, digital technologies are widely used to record any event by using different types of cameras, for different uses. There are cameras that use an optical sensor, which generates images in the RGB model. There are some cameras that use thermal sensors, which perform the registration of the temperature of the object being recorded and presents them under a color model using a color map. This research study uses a novel method to fuse optical and thermal images with the intention of being able to recognize certain parts of the human body. This study evaluates the presence of veins that have a higher temperature when performing rehabilitation exercises. The results allow to evaluate different combinations of color bands. In order to demonstrate the method, the use and application will depend on the analysis of the image and the interpretation by health personnel. At the time of merging the images, the optical image provides the structural part of the image, and the thermal image provides the functionality characterized by the body temperature. As a conclusion, this research study indicates that the proposed method can be applied to other applications in order to look for applications where the method can help in the diagnosis and evaluation of the treatment.
{"title":"Method for Fusing Optical and Thermal Images Applied to Muscle Analysis","authors":"W. Auccahuasi, Oscar Linares, K. Urbano, K. Rojas, Gabriel Aiquipa, Tamara Pando-Ezcurra","doi":"10.1109/ICAAIC56838.2023.10140866","DOIUrl":"https://doi.org/10.1109/ICAAIC56838.2023.10140866","url":null,"abstract":"Currently, digital technologies are widely used to record any event by using different types of cameras, for different uses. There are cameras that use an optical sensor, which generates images in the RGB model. There are some cameras that use thermal sensors, which perform the registration of the temperature of the object being recorded and presents them under a color model using a color map. This research study uses a novel method to fuse optical and thermal images with the intention of being able to recognize certain parts of the human body. This study evaluates the presence of veins that have a higher temperature when performing rehabilitation exercises. The results allow to evaluate different combinations of color bands. In order to demonstrate the method, the use and application will depend on the analysis of the image and the interpretation by health personnel. At the time of merging the images, the optical image provides the structural part of the image, and the thermal image provides the functionality characterized by the body temperature. As a conclusion, this research study indicates that the proposed method can be applied to other applications in order to look for applications where the method can help in the diagnosis and evaluation of the treatment.","PeriodicalId":267906,"journal":{"name":"2023 2nd International Conference on Applied Artificial Intelligence and Computing (ICAAIC)","volume":"62 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123848684","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 : 2023-05-04DOI: 10.1109/ICAAIC56838.2023.10140669
E. Sivajothi, J. Jayaudhaya, S. Santhiya, Naresh Kumar Thapa, S. Kamatchi, N. Ganapathy
The research work provides a brand-new technique for distributing bandwidth in wireless networks that is also energy-efficient. To begin, look into how channel allocations, which ultimately determine the transmission rate for mobile terminals, affect the fundamental relationship that exists between energy use and transmission rates. The Fatality Selection Algorithm (FSA) and the Receiver Selection Algorithm (RSA) are two methodologies that control connection admittance to reduce the amount of energy used by each individual terminal. Furthermore, provide a corrective strategy for statistically satisfying the quality of service (QoS) criteria throughout the resource allocation process. Throughput, call blockage probability, and the energy consumption rate of each successfully sent bit are used to evaluate the efficiency of the recommended solutions. An extensive investigation into analysis and simulation is carried out in the case of Poisson and self-similar.
{"title":"A Novel Framework in Scheduling Packets for Energy-Efficient Bandwidth Allocation in Wireless Networks","authors":"E. Sivajothi, J. Jayaudhaya, S. Santhiya, Naresh Kumar Thapa, S. Kamatchi, N. Ganapathy","doi":"10.1109/ICAAIC56838.2023.10140669","DOIUrl":"https://doi.org/10.1109/ICAAIC56838.2023.10140669","url":null,"abstract":"The research work provides a brand-new technique for distributing bandwidth in wireless networks that is also energy-efficient. To begin, look into how channel allocations, which ultimately determine the transmission rate for mobile terminals, affect the fundamental relationship that exists between energy use and transmission rates. The Fatality Selection Algorithm (FSA) and the Receiver Selection Algorithm (RSA) are two methodologies that control connection admittance to reduce the amount of energy used by each individual terminal. Furthermore, provide a corrective strategy for statistically satisfying the quality of service (QoS) criteria throughout the resource allocation process. Throughput, call blockage probability, and the energy consumption rate of each successfully sent bit are used to evaluate the efficiency of the recommended solutions. An extensive investigation into analysis and simulation is carried out in the case of Poisson and self-similar.","PeriodicalId":267906,"journal":{"name":"2023 2nd International Conference on Applied Artificial Intelligence and Computing (ICAAIC)","volume":"7 5","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114115097","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 : 2023-05-04DOI: 10.1109/ICAAIC56838.2023.10141524
K. Alice, A. Thillaivanan, Ganga Rama Koteswara Rao, R. S, Kamlesh Singh, Ravi Rastogi
Automated Forest Fire Detection (AFFD) contains the technology used to recognize and alert authorities on latent wildfires in a forested region. AFFD methods are latent to enhance response times and decrease the damage led by wildfires. But, these systems are utilized in conjunction with typical fire management practices like fire prevention and suppression measures, to provide the best achievable outcomes. There are several algorithms to AFFD, comprising computer vision (CV), remote sensing, and machine learning (ML). This article develops an Automated Forest Fire Detection using Atom Search Optimizer with Deep Transfer Learning (AFFD-ASODTL) model. The goal of the AFFD-ASODTL technique lies in the effectual recognition of forest fires accurately and promptly. In the presented AFFD-ASODTL technique, residual network (ResNet50) model is applied for feature vector generation. Besides, the ASO technique is exploited for the optimal hyperparameter tuning of the ResNet model. Meanwhile, Quasi-Recurrent Neural Network (QRNN) model is used for forest fire classification. To exhibit the optimum resultant of the AFFD-AS ODTL system, a comprehensive set of simulations is carried out. The comparative study highlighted the improvised results of the AFFD-ASODTL method over other models.
{"title":"Automated Forest Fire Detection using Atom Search Optimizer with Deep Transfer Learning Model","authors":"K. Alice, A. Thillaivanan, Ganga Rama Koteswara Rao, R. S, Kamlesh Singh, Ravi Rastogi","doi":"10.1109/ICAAIC56838.2023.10141524","DOIUrl":"https://doi.org/10.1109/ICAAIC56838.2023.10141524","url":null,"abstract":"Automated Forest Fire Detection (AFFD) contains the technology used to recognize and alert authorities on latent wildfires in a forested region. AFFD methods are latent to enhance response times and decrease the damage led by wildfires. But, these systems are utilized in conjunction with typical fire management practices like fire prevention and suppression measures, to provide the best achievable outcomes. There are several algorithms to AFFD, comprising computer vision (CV), remote sensing, and machine learning (ML). This article develops an Automated Forest Fire Detection using Atom Search Optimizer with Deep Transfer Learning (AFFD-ASODTL) model. The goal of the AFFD-ASODTL technique lies in the effectual recognition of forest fires accurately and promptly. In the presented AFFD-ASODTL technique, residual network (ResNet50) model is applied for feature vector generation. Besides, the ASO technique is exploited for the optimal hyperparameter tuning of the ResNet model. Meanwhile, Quasi-Recurrent Neural Network (QRNN) model is used for forest fire classification. To exhibit the optimum resultant of the AFFD-AS ODTL system, a comprehensive set of simulations is carried out. The comparative study highlighted the improvised results of the AFFD-ASODTL method over other models.","PeriodicalId":267906,"journal":{"name":"2023 2nd International Conference on Applied Artificial Intelligence and Computing (ICAAIC)","volume":"120 8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126309490","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 : 2023-05-04DOI: 10.1109/ICAAIC56838.2023.10140417
Madiri Divya Sumitra, P. Swetha, Modugumudi Natesh Venkata Babu, Yanamala Raj Kumar, M. Lakshmi
Sleep apnea occurs when breathing stops for more than 10 seconds at a time during the night. These occurrences must be correctly diagnosed. The recordings began with preliminary processing and segmentation of electrocardiogram (ECG) data. Deep learning and machine learning were used to make the diagnosis of sleep apnea. Each network was modified in the same way to be suitable for biosignal processing. The training, validation, and test sets were used to optimize model parameters and hyperparameters, while the test set was used to evaluate the model's performance on new data. Each recording was tested several times using a technique known as 5-fold cross-validation. Deep learning models had the highest detection accuracy rate of 88.13%. Sleep apnea and other sleep disorders can be difficult to diagnose, but this study demonstrates the effectiveness of various machine learning and deep learning algorithms.
{"title":"Deep Learning Model for ECG-based Sleep Apnea Detection","authors":"Madiri Divya Sumitra, P. Swetha, Modugumudi Natesh Venkata Babu, Yanamala Raj Kumar, M. Lakshmi","doi":"10.1109/ICAAIC56838.2023.10140417","DOIUrl":"https://doi.org/10.1109/ICAAIC56838.2023.10140417","url":null,"abstract":"Sleep apnea occurs when breathing stops for more than 10 seconds at a time during the night. These occurrences must be correctly diagnosed. The recordings began with preliminary processing and segmentation of electrocardiogram (ECG) data. Deep learning and machine learning were used to make the diagnosis of sleep apnea. Each network was modified in the same way to be suitable for biosignal processing. The training, validation, and test sets were used to optimize model parameters and hyperparameters, while the test set was used to evaluate the model's performance on new data. Each recording was tested several times using a technique known as 5-fold cross-validation. Deep learning models had the highest detection accuracy rate of 88.13%. Sleep apnea and other sleep disorders can be difficult to diagnose, but this study demonstrates the effectiveness of various machine learning and deep learning algorithms.","PeriodicalId":267906,"journal":{"name":"2023 2nd International Conference on Applied Artificial Intelligence and Computing (ICAAIC)","volume":"4303 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126358591","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 : 2023-05-04DOI: 10.1109/ICAAIC56838.2023.10140337
K. Santoshi, G. Saranya, Ch.Rama Reddy, Ch. Jathin Reddy, K. Gyananandu, G. N. Tej
One of the major health problems that modern humans encounter is malaria, which affects people of all ages. Malaria is a fatal disease caused by parasites carried by the infected mosquitoes. One way for diagnosing malaria is to examine a sample of the person's blood underneath a microscope for the presence of parasites. The project involves the creation of a web app that employs deep learning to recognize malaria parasites in images from blood smears. This can be accomplished by collecting and labeling a dataset of blood smear images utilizing convolutional neural network (CNN) models such as ResNet50, VGG19, and Customized CNN to discover patterns and features in the images. A Convolutional Neural Network (CNN) model is customized by including convolutional layers, max-pooling layers, totally connected layers, and a SoftMax layer. This approach has the power to increase the detection speed, precision of parasite diagnosis and assist in lowering the disease's global health impact.
{"title":"Deep Learning based Web App for Malaria Parasite Detection in Granular Blood Samples","authors":"K. Santoshi, G. Saranya, Ch.Rama Reddy, Ch. Jathin Reddy, K. Gyananandu, G. N. Tej","doi":"10.1109/ICAAIC56838.2023.10140337","DOIUrl":"https://doi.org/10.1109/ICAAIC56838.2023.10140337","url":null,"abstract":"One of the major health problems that modern humans encounter is malaria, which affects people of all ages. Malaria is a fatal disease caused by parasites carried by the infected mosquitoes. One way for diagnosing malaria is to examine a sample of the person's blood underneath a microscope for the presence of parasites. The project involves the creation of a web app that employs deep learning to recognize malaria parasites in images from blood smears. This can be accomplished by collecting and labeling a dataset of blood smear images utilizing convolutional neural network (CNN) models such as ResNet50, VGG19, and Customized CNN to discover patterns and features in the images. A Convolutional Neural Network (CNN) model is customized by including convolutional layers, max-pooling layers, totally connected layers, and a SoftMax layer. This approach has the power to increase the detection speed, precision of parasite diagnosis and assist in lowering the disease's global health impact.","PeriodicalId":267906,"journal":{"name":"2023 2nd International Conference on Applied Artificial Intelligence and Computing (ICAAIC)","volume":"34 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126460269","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 : 2023-05-04DOI: 10.1109/ICAAIC56838.2023.10141095
Madhuri Pagale, Richa Purohit, Pallavi Dhade, A. Thakare, Santwana S. Gudadhe, Pradnya Narkhede
This Traffic Sign Recognition (TSR) plays a vital role in disciplining drivers and managing traffic on the road, which helps to prevent road accidents, damage, fatalities and property injury. Traffic sign recognition and management with automatic detection are critical components of any Smart Transportation System (STS). Throughout this era of autonomous vehicles, automated detection as well as identification of traffic signs are a must. This research discusses a self-directed traffic sign identification system in India that is based on deep learning. Automatic traffic sign identification as well as recognition was created utilizing Convolutional Neural Network (CNN) learning from the ground up. Deep Convolutional Neural Networks are now used to an increasing number of object recognition applications. Convolutional neural networks(CNN) have improved both current and new computer vision tasks due to their high detection rate and superior performance. This study proposes a strategy for identifying traffic signals that makes use of deep convolution neural network. This research study compares many CNN designs against one another. TensorFlow, a prominent machine learning framework is built by utilizing the massively parallel multithreaded programming of CUDA architecture for deep neural network training. The trial findings validated the effectiveness of the created computer vision system. The proposed model attained an accuracy of 97.08%, which is superior to the present approach of traffic sign detection.
{"title":"Insights of Deep Convolutional Neural Network for Traffic Sign Detection in Autonomous Vehicle","authors":"Madhuri Pagale, Richa Purohit, Pallavi Dhade, A. Thakare, Santwana S. Gudadhe, Pradnya Narkhede","doi":"10.1109/ICAAIC56838.2023.10141095","DOIUrl":"https://doi.org/10.1109/ICAAIC56838.2023.10141095","url":null,"abstract":"This Traffic Sign Recognition (TSR) plays a vital role in disciplining drivers and managing traffic on the road, which helps to prevent road accidents, damage, fatalities and property injury. Traffic sign recognition and management with automatic detection are critical components of any Smart Transportation System (STS). Throughout this era of autonomous vehicles, automated detection as well as identification of traffic signs are a must. This research discusses a self-directed traffic sign identification system in India that is based on deep learning. Automatic traffic sign identification as well as recognition was created utilizing Convolutional Neural Network (CNN) learning from the ground up. Deep Convolutional Neural Networks are now used to an increasing number of object recognition applications. Convolutional neural networks(CNN) have improved both current and new computer vision tasks due to their high detection rate and superior performance. This study proposes a strategy for identifying traffic signals that makes use of deep convolution neural network. This research study compares many CNN designs against one another. TensorFlow, a prominent machine learning framework is built by utilizing the massively parallel multithreaded programming of CUDA architecture for deep neural network training. The trial findings validated the effectiveness of the created computer vision system. The proposed model attained an accuracy of 97.08%, which is superior to the present approach of traffic sign detection.","PeriodicalId":267906,"journal":{"name":"2023 2nd International Conference on Applied Artificial Intelligence and Computing (ICAAIC)","volume":"2008 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125622320","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：