Pub Date : 2023-05-04DOI: 10.1109/ESDC56251.2023.10149851
Dinu Thomas, David Pratap, B. Sudha
Volume of video content surpass all other content types in internet. As per the reports from different sources, video traffic had acquired 82% of internet usage in 2022. Video is going to be more important in the years to come for user engagement, advertisement & marketing, news, education etc. Video information retrieval becomes an important problem to solve in this context. An accurate and fast video tagging system can aid a good content recommendation to the end users. It helps to audit the content automatically thereby platforms can control the contents which are politically and morally harmful. There are not many faster or cost-effective mechanisms to tag user generated videos at this moment. Manual tagging is a costly and highly time taking task. A delay in indexing the videos like news, sports etc., shall reduce its freshness and relevancy. Deep learning techniques have reached its maturity in the contents like text and images, but it is not the case with videos. Deep learning models need more resources to deal with videos due to its multi-modality nature, and temporal behavior. Apart from that, there are not many large-scale video datasets available at this moment. Youtube-8M is the largest dataset which is publicly available as of now. Much research works happened over Youtube-8M dataset. From our study, all these have a potential limitation. For example, in Youtube-8M, Video labels are only around 3.8K which are not covering all real-world tags. It is not covering the new domains which are created along with the surge in the content traffic. This study aims to handle this problem of tag creation through different methods available thereby enhancing the labels to a much wider set. This work also aims to produce a scalable tagging pipeline which uses multiple retrieval mechanisms, combine their results. The work aims to standardize the retrieved tokens across languages. This work creates a dataset as an outcome from ‘WikiData’, which can be used for any NLP based standardization use cases. An attempt has been made to do disambiguation through WikiId embedding. A new WikiData embedding is created in this work, which can be used for eliminating the tags which are noisy.
{"title":"Video Label Enhancing and Standardization through Transcription and WikiId Mapping Techniques","authors":"Dinu Thomas, David Pratap, B. Sudha","doi":"10.1109/ESDC56251.2023.10149851","DOIUrl":"https://doi.org/10.1109/ESDC56251.2023.10149851","url":null,"abstract":"Volume of video content surpass all other content types in internet. As per the reports from different sources, video traffic had acquired 82% of internet usage in 2022. Video is going to be more important in the years to come for user engagement, advertisement & marketing, news, education etc. Video information retrieval becomes an important problem to solve in this context. An accurate and fast video tagging system can aid a good content recommendation to the end users. It helps to audit the content automatically thereby platforms can control the contents which are politically and morally harmful. There are not many faster or cost-effective mechanisms to tag user generated videos at this moment. Manual tagging is a costly and highly time taking task. A delay in indexing the videos like news, sports etc., shall reduce its freshness and relevancy. Deep learning techniques have reached its maturity in the contents like text and images, but it is not the case with videos. Deep learning models need more resources to deal with videos due to its multi-modality nature, and temporal behavior. Apart from that, there are not many large-scale video datasets available at this moment. Youtube-8M is the largest dataset which is publicly available as of now. Much research works happened over Youtube-8M dataset. From our study, all these have a potential limitation. For example, in Youtube-8M, Video labels are only around 3.8K which are not covering all real-world tags. It is not covering the new domains which are created along with the surge in the content traffic. This study aims to handle this problem of tag creation through different methods available thereby enhancing the labels to a much wider set. This work also aims to produce a scalable tagging pipeline which uses multiple retrieval mechanisms, combine their results. The work aims to standardize the retrieved tokens across languages. This work creates a dataset as an outcome from ‘WikiData’, which can be used for any NLP based standardization use cases. An attempt has been made to do disambiguation through WikiId embedding. A new WikiData embedding is created in this work, which can be used for eliminating the tags which are noisy.","PeriodicalId":354855,"journal":{"name":"2023 11th International Symposium on Electronic Systems Devices and Computing (ESDC)","volume":"240 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":"114329586","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/ESDC56251.2023.10149869
Narayana Darapaneni, A. Paduri, Dinu Thomas, Jisha C U, Abhinao Shrivastava, Seema Biradar
In today’s world, the UGC (User Generated Contents) videos have increased exponentially. Billions of videos are uploaded, played and exchanged between different actors. In this context, automatic video content classification has become a critical and challenging problem, especially in areas like video-based search, recommendation etc. In this work we try to extract frame-level visual and audio features, pre-extracted features are then converted into a compact video level representation effectively and efficiently. We aim to classify the video into a set of categories with high accuracy. From the literature survey, we identified that, the tagging of videos has been a problem which has not reached its maturity yet, and there are many researches happening in this area. It is observed that, the clustering based video description methodologies show a better result compared to the temporal algorithms. We also have identified that, majority of the SOTA techniques use the VLAD (Vector of Locally Aggregated Descriptors) technique to extract the video features and make the codebook learnable through some adjustments introduced in the NetVLAD. The key descriptors would be mostly noisy, and many of them are insignificant. In this work we aim to cascade a Self-Attention Block on the NetVLAD which can extract the significant descriptors and filter out the Noise. The YouTube 8M dataset shall be used for training the model and performance will be compared with other SOTA techniques. Like other similar works, model performance will be measured by GAP Metric (Global Average Precision) for all the videos predicted labels. We aim to achieve a GAP score close to 85% for this work.
{"title":"Video understanding : Tagging of videos through self attentive learnable key descriptors","authors":"Narayana Darapaneni, A. Paduri, Dinu Thomas, Jisha C U, Abhinao Shrivastava, Seema Biradar","doi":"10.1109/ESDC56251.2023.10149869","DOIUrl":"https://doi.org/10.1109/ESDC56251.2023.10149869","url":null,"abstract":"In today’s world, the UGC (User Generated Contents) videos have increased exponentially. Billions of videos are uploaded, played and exchanged between different actors. In this context, automatic video content classification has become a critical and challenging problem, especially in areas like video-based search, recommendation etc. In this work we try to extract frame-level visual and audio features, pre-extracted features are then converted into a compact video level representation effectively and efficiently. We aim to classify the video into a set of categories with high accuracy. From the literature survey, we identified that, the tagging of videos has been a problem which has not reached its maturity yet, and there are many researches happening in this area. It is observed that, the clustering based video description methodologies show a better result compared to the temporal algorithms. We also have identified that, majority of the SOTA techniques use the VLAD (Vector of Locally Aggregated Descriptors) technique to extract the video features and make the codebook learnable through some adjustments introduced in the NetVLAD. The key descriptors would be mostly noisy, and many of them are insignificant. In this work we aim to cascade a Self-Attention Block on the NetVLAD which can extract the significant descriptors and filter out the Noise. The YouTube 8M dataset shall be used for training the model and performance will be compared with other SOTA techniques. Like other similar works, model performance will be measured by GAP Metric (Global Average Precision) for all the videos predicted labels. We aim to achieve a GAP score close to 85% for this work.","PeriodicalId":354855,"journal":{"name":"2023 11th International Symposium on Electronic Systems Devices and Computing (ESDC)","volume":"65 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":"126513519","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/ESDC56251.2023.10149854
Rahul Deo Verma, Mahesh Chandra Govil, Pankaj Kumar Keserwani
The Border Gateway Protocol (BGP) is an essential element of the Internet infrastructure, playing a crucial role in ensuring global connectivity and stability for smooth communication. Numerous techniques are created by the researchers to detect anomalies in the network for enhancing stability of the BGP network (BGP based environment). On the other hand, the dynamic nature and intricate structure of the network poses challenges in identifying attacks environment. As a result, applying a single classifier to classify them is a challenging task. In this paper a method based on ensemble learning is proposed where Extreme Learning Machine (ELM), K-Nearest Neighbor (KNN), and Naive Bayes (NB), classifiers are ensembled to detect the attacks in BGP based environment. The proposed approach is evaluated on Reseaux IP Europeens (RIPE) and British Columbia’s Advanced Network (BCNET) datasets and compared with other recent approaches. On investigating the results, it is found that the proposed approach is providing better performance on both datasets.
{"title":"ELM based Ensemble of Classifiers for BGP Security against Network Anomalies","authors":"Rahul Deo Verma, Mahesh Chandra Govil, Pankaj Kumar Keserwani","doi":"10.1109/ESDC56251.2023.10149854","DOIUrl":"https://doi.org/10.1109/ESDC56251.2023.10149854","url":null,"abstract":"The Border Gateway Protocol (BGP) is an essential element of the Internet infrastructure, playing a crucial role in ensuring global connectivity and stability for smooth communication. Numerous techniques are created by the researchers to detect anomalies in the network for enhancing stability of the BGP network (BGP based environment). On the other hand, the dynamic nature and intricate structure of the network poses challenges in identifying attacks environment. As a result, applying a single classifier to classify them is a challenging task. In this paper a method based on ensemble learning is proposed where Extreme Learning Machine (ELM), K-Nearest Neighbor (KNN), and Naive Bayes (NB), classifiers are ensembled to detect the attacks in BGP based environment. The proposed approach is evaluated on Reseaux IP Europeens (RIPE) and British Columbia’s Advanced Network (BCNET) datasets and compared with other recent approaches. On investigating the results, it is found that the proposed approach is providing better performance on both datasets.","PeriodicalId":354855,"journal":{"name":"2023 11th International Symposium on Electronic Systems Devices and Computing (ESDC)","volume":"149 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":"133249412","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/ESDC56251.2023.10149871
Venkata Maha Lakshmi N, R. Rout
The prediction of Heart attack is one of the burning problems in the medical field. There are various attributes tresultults in the stress and health of the human being. Existing researchers concentrated on the attributes based on the tests related to heart attacks like restECG, echo and others but along with these metrics like weight, gender, working hours and others plays a vital role. The proposed model studies the importance of features by varying the layers of neural networks with different possibilities of activation functions because out of the different estimators available for the neural network, these functions are the one that transforms the behavior of the network rapidly with fewer resources utilization. The model has considered 9 activation functions and designed a 4-layered neural network and the hidden layers are customized with a grid search selection of activation functions. The main advantage of grid search optimization is it constructs a complete problem search space by considering every minute detail. The input and output layers are static with standard ReLu for input layer and sigmoid for the output layer because the dataset is a binary classification problem. The model compared the proposed model with static layers of network on the same 61 records has got training accuracy of 95.67% but the validation accuracy is 79% which is less when compared to the validation accuracy of the proposed is 81.9%.
{"title":"A Novel Grid Ann for Prediction of Heart Disease","authors":"Venkata Maha Lakshmi N, R. Rout","doi":"10.1109/ESDC56251.2023.10149871","DOIUrl":"https://doi.org/10.1109/ESDC56251.2023.10149871","url":null,"abstract":"The prediction of Heart attack is one of the burning problems in the medical field. There are various attributes tresultults in the stress and health of the human being. Existing researchers concentrated on the attributes based on the tests related to heart attacks like restECG, echo and others but along with these metrics like weight, gender, working hours and others plays a vital role. The proposed model studies the importance of features by varying the layers of neural networks with different possibilities of activation functions because out of the different estimators available for the neural network, these functions are the one that transforms the behavior of the network rapidly with fewer resources utilization. The model has considered 9 activation functions and designed a 4-layered neural network and the hidden layers are customized with a grid search selection of activation functions. The main advantage of grid search optimization is it constructs a complete problem search space by considering every minute detail. The input and output layers are static with standard ReLu for input layer and sigmoid for the output layer because the dataset is a binary classification problem. The model compared the proposed model with static layers of network on the same 61 records has got training accuracy of 95.67% but the validation accuracy is 79% which is less when compared to the validation accuracy of the proposed is 81.9%.","PeriodicalId":354855,"journal":{"name":"2023 11th International Symposium on Electronic Systems Devices and Computing (ESDC)","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":"130860876","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/ESDC56251.2023.10149876
P. Hrishikesh, Densen Puthussery, K. A. Akhil, C. Jiji
Generative adversarial networks (GAN) are proved to be extremely useful to solve the Single image super resolution(SISR) problem as they can recover the finer texture details even with large upsamplng factors. In this paper, we propose a deep network architecture using a relativistic generative adversarial network (V-SRGAN) with receptive field block (RFB) for image super-resolution having good perceptual quality. Our generator network uses multi-scale RFBs which are capable of extracting the coarse and finer features from the input low resolution image to recover the super resolved image with finer details and textures. It is initially trained on mean absolute error (MAE) followed with relativistic average GAN (RaGAN) loss for both discriminator and generator. Training based on RaGAN loss enables the network to map the low resolution images to more realistic high-resolution counterparts. The proposed network was able to attain better results in terms of PSNR and learned perceptual image patch similarity (LPIPS) metric in comparison with the other GAN based methods. https://github.com/hrishikeshps94/RGAN-with-Receptive-Field-Block-for-SISR
{"title":"Relativistic GAN using Receptive Field Block for Single Image Super-Resolution with improved Perceptual Quality","authors":"P. Hrishikesh, Densen Puthussery, K. A. Akhil, C. Jiji","doi":"10.1109/ESDC56251.2023.10149876","DOIUrl":"https://doi.org/10.1109/ESDC56251.2023.10149876","url":null,"abstract":"Generative adversarial networks (GAN) are proved to be extremely useful to solve the Single image super resolution(SISR) problem as they can recover the finer texture details even with large upsamplng factors. In this paper, we propose a deep network architecture using a relativistic generative adversarial network (V-SRGAN) with receptive field block (RFB) for image super-resolution having good perceptual quality. Our generator network uses multi-scale RFBs which are capable of extracting the coarse and finer features from the input low resolution image to recover the super resolved image with finer details and textures. It is initially trained on mean absolute error (MAE) followed with relativistic average GAN (RaGAN) loss for both discriminator and generator. Training based on RaGAN loss enables the network to map the low resolution images to more realistic high-resolution counterparts. The proposed network was able to attain better results in terms of PSNR and learned perceptual image patch similarity (LPIPS) metric in comparison with the other GAN based methods. https://github.com/hrishikeshps94/RGAN-with-Receptive-Field-Block-for-SISR","PeriodicalId":354855,"journal":{"name":"2023 11th International Symposium on Electronic Systems Devices and Computing (ESDC)","volume":"17 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":"116607395","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/ESDC56251.2023.10149875
Amal S Namboodiri, Rakesh Kumar Sanodiya, PV Arun
High Resolution satellite images are of at most importance in the field of remote sensing. However, these images require quite a bit of preprocessing to ensure that the underlying landscape is not obstructed by any kind of unwanted noise. This paper addresses the problem of obstruction of remote sensing satellite data by clouds using a unique Generative Adversarial Network (GAN) model. Our proposed model Spatial Attention + Edges Generative Adverserial Network(SpA+Edges GAN) uses the spatial attention feature to focus on the regions of importance, namely the cloudy region during the reconstruction process. We combine this with the use of an edge filter that is used by the discriminator to compare the edges of the generated non-cloudy image and the cloud-free image. We also introduce a new loss function that forces the model to focus more on the cloudy region during the reconstruction process. We compare our model with other existing models on popular remote sensing datasets and also on a new dataset of our own using Peak signal to noise ratio (PSNR) and Structural Similarity index (SSIM). Through our experiments we show that combining the spatial attentive feature along with the edge filter provide promising results in removing clouds from remote sensing data.
{"title":"Remote Sensing Cloud Removal using a Combination of Spatial Attention and Edge Detection","authors":"Amal S Namboodiri, Rakesh Kumar Sanodiya, PV Arun","doi":"10.1109/ESDC56251.2023.10149875","DOIUrl":"https://doi.org/10.1109/ESDC56251.2023.10149875","url":null,"abstract":"High Resolution satellite images are of at most importance in the field of remote sensing. However, these images require quite a bit of preprocessing to ensure that the underlying landscape is not obstructed by any kind of unwanted noise. This paper addresses the problem of obstruction of remote sensing satellite data by clouds using a unique Generative Adversarial Network (GAN) model. Our proposed model Spatial Attention + Edges Generative Adverserial Network(SpA+Edges GAN) uses the spatial attention feature to focus on the regions of importance, namely the cloudy region during the reconstruction process. We combine this with the use of an edge filter that is used by the discriminator to compare the edges of the generated non-cloudy image and the cloud-free image. We also introduce a new loss function that forces the model to focus more on the cloudy region during the reconstruction process. We compare our model with other existing models on popular remote sensing datasets and also on a new dataset of our own using Peak signal to noise ratio (PSNR) and Structural Similarity index (SSIM). Through our experiments we show that combining the spatial attentive feature along with the edge filter provide promising results in removing clouds from remote sensing data.","PeriodicalId":354855,"journal":{"name":"2023 11th International Symposium on Electronic Systems Devices and Computing (ESDC)","volume":"31 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":"114536483","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/ESDC56251.2023.10149881
Manchala Shivamani, K.C. Meghavardhan Reddy, Shaik Shakeera, H. Venkataraman
Autonomous Vehicles (AV) are the future of the smart digital world. Auto Navigation is the heart of AVs. However, researchers have been working on precise auto navigation control of AVs for many years. Hence, In this paper, a GPS-aided auto navigation system is proposed with position control and heading control of AV. Further, to test the proposed algorithm an AV is designed and developed with low-cost sensors and actuators. The real-time testing of the proposed adaptive control auto navigation mechanism has been performed with three test cases such as straight line, L-shaped and Semi-circular. The results shown in this paper are the deviation between the travelled path and the defined path of the vehicle in different trajectories by assuming no obstacles in the path. The deviation of the path travelled by AV with respect to the defined path using the proposed algorithm is less than 0.5m. Finally, the telemetry of AV has been monitored by the developed Graphical User Interface (GUI). This work is very useful in auto navigation of vehicles where humans can not be sustained.
{"title":"GPS-Aided Auto Navigation System for Autonomous Vehicles","authors":"Manchala Shivamani, K.C. Meghavardhan Reddy, Shaik Shakeera, H. Venkataraman","doi":"10.1109/ESDC56251.2023.10149881","DOIUrl":"https://doi.org/10.1109/ESDC56251.2023.10149881","url":null,"abstract":"Autonomous Vehicles (AV) are the future of the smart digital world. Auto Navigation is the heart of AVs. However, researchers have been working on precise auto navigation control of AVs for many years. Hence, In this paper, a GPS-aided auto navigation system is proposed with position control and heading control of AV. Further, to test the proposed algorithm an AV is designed and developed with low-cost sensors and actuators. The real-time testing of the proposed adaptive control auto navigation mechanism has been performed with three test cases such as straight line, L-shaped and Semi-circular. The results shown in this paper are the deviation between the travelled path and the defined path of the vehicle in different trajectories by assuming no obstacles in the path. The deviation of the path travelled by AV with respect to the defined path using the proposed algorithm is less than 0.5m. Finally, the telemetry of AV has been monitored by the developed Graphical User Interface (GUI). This work is very useful in auto navigation of vehicles where humans can not be sustained.","PeriodicalId":354855,"journal":{"name":"2023 11th International Symposium on Electronic Systems Devices and Computing (ESDC)","volume":"48 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":"116856221","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/ESDC56251.2023.10149864
Umasankar Thogaram, Pradeep Kumar Asthana
Algorithmic Trading (AT) is the most popular trading pattern, to identify trends and movements in a particular direction. Algo trading is widely adopted across all global markets. It is a computerized rule-based process, eliminating human intervention and large orders are placed in the market, as per set rules. The algorithms scan the entire spectrum of the chosen area and compute the decision based on multiple indicators, with high speed and precision. Multiple Algorithms are available, based on vendor/ broker preference for specific goals. As the trading activity is getting complex with the fast pace of market moves and escalating volumes, with greater volatility, data coupled with artificial intelligence is providing cutting edge to Algo Trading, for spotting and predicting the trends with matching speed, to tilt trading activity in favour of the trader.
{"title":"Algo Trading – A New Paradigm in The Stock Trading","authors":"Umasankar Thogaram, Pradeep Kumar Asthana","doi":"10.1109/ESDC56251.2023.10149864","DOIUrl":"https://doi.org/10.1109/ESDC56251.2023.10149864","url":null,"abstract":"Algorithmic Trading (AT) is the most popular trading pattern, to identify trends and movements in a particular direction. Algo trading is widely adopted across all global markets. It is a computerized rule-based process, eliminating human intervention and large orders are placed in the market, as per set rules. The algorithms scan the entire spectrum of the chosen area and compute the decision based on multiple indicators, with high speed and precision. Multiple Algorithms are available, based on vendor/ broker preference for specific goals. As the trading activity is getting complex with the fast pace of market moves and escalating volumes, with greater volatility, data coupled with artificial intelligence is providing cutting edge to Algo Trading, for spotting and predicting the trends with matching speed, to tilt trading activity in favour of the trader.","PeriodicalId":354855,"journal":{"name":"2023 11th International Symposium on Electronic Systems Devices and Computing (ESDC)","volume":"59 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":"115855173","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/ESDC56251.2023.10149856
Jagadheesh Samala, S. J
With the advancements in VLSI technology, the possibility of multiple cores is now a reality. This has brought on some new challenges like communication and scalability to the field. Network-on-Chips (NoCs) are proposed to address the challenges of multi-core systems and have become a prominent solution for the multi-processor system-on-chip (MPSoC). Mesh topology is one of the simple and efficient topologies of NoC. The routing algorithm used in mesh topology is static. All the packets move in fixed paths, which create more load on some routers. A dynamic routing algorithm is proposed in this paper to address traffic distribution and prevent the routers from failure due to the excess load. The proposed routing algorithm uses a univariate linear regression model to predict the router utilization and, based on the prediction, distributes the traffic uniformly over all the routers. Experimentations are performed by implementing the proposed algorithm in NoC simulator. The results show that the proposed dynamic routing algorithm significantly improves traffic distribution over the XY-routing algorithm, algorithms proposed in [9] and [10].
{"title":"Dynamic routing algorithm to normalize the routers utilization in mesh based NoC","authors":"Jagadheesh Samala, S. J","doi":"10.1109/ESDC56251.2023.10149856","DOIUrl":"https://doi.org/10.1109/ESDC56251.2023.10149856","url":null,"abstract":"With the advancements in VLSI technology, the possibility of multiple cores is now a reality. This has brought on some new challenges like communication and scalability to the field. Network-on-Chips (NoCs) are proposed to address the challenges of multi-core systems and have become a prominent solution for the multi-processor system-on-chip (MPSoC). Mesh topology is one of the simple and efficient topologies of NoC. The routing algorithm used in mesh topology is static. All the packets move in fixed paths, which create more load on some routers. A dynamic routing algorithm is proposed in this paper to address traffic distribution and prevent the routers from failure due to the excess load. The proposed routing algorithm uses a univariate linear regression model to predict the router utilization and, based on the prediction, distributes the traffic uniformly over all the routers. Experimentations are performed by implementing the proposed algorithm in NoC simulator. The results show that the proposed dynamic routing algorithm significantly improves traffic distribution over the XY-routing algorithm, algorithms proposed in [9] and [10].","PeriodicalId":354855,"journal":{"name":"2023 11th International Symposium on Electronic Systems Devices and Computing (ESDC)","volume":"76 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":"115680859","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}
Aerial surveillance is one of the widely used modern days surveillance methodologies, finding applications in many important fields including military and civilian. This article presents a comprehensive study of Deep Neural Network (DNN) based solutions for real-time object tracking from Unmanned Aerial Vehicle (UAV) using a modified version of the state-of-the-art object detection algorithm YOLOv5 model. The modified YOLOv5 architecture is achieved by changing the activation function to Rectified Linear Unit (ReLU) and fine-tuning the network’s hyperparameter. A comparative analysis was then done on a subset of the AU-AIR dataset by comparing the different YOLOv5 models based on the network depth to determine the improvements in training speed and accuracy. The modified network was also compared in terms of mean average precision (mAP) to the original paper, a performance gain of almost 2.9 times was achieved in the best-case scenario.
{"title":"Deep Neural Network Based Multi-Object Detection for Real-time Aerial Surveillance","authors":"Rebanta Dey, Binit Kumar Pandit, Anirban Ganguly, Anirban Chakraborty, Ayan Banerjee","doi":"10.1109/ESDC56251.2023.10149866","DOIUrl":"https://doi.org/10.1109/ESDC56251.2023.10149866","url":null,"abstract":"Aerial surveillance is one of the widely used modern days surveillance methodologies, finding applications in many important fields including military and civilian. This article presents a comprehensive study of Deep Neural Network (DNN) based solutions for real-time object tracking from Unmanned Aerial Vehicle (UAV) using a modified version of the state-of-the-art object detection algorithm YOLOv5 model. The modified YOLOv5 architecture is achieved by changing the activation function to Rectified Linear Unit (ReLU) and fine-tuning the network’s hyperparameter. A comparative analysis was then done on a subset of the AU-AIR dataset by comparing the different YOLOv5 models based on the network depth to determine the improvements in training speed and accuracy. The modified network was also compared in terms of mean average precision (mAP) to the original paper, a performance gain of almost 2.9 times was achieved in the best-case scenario.","PeriodicalId":354855,"journal":{"name":"2023 11th International Symposium on Electronic Systems Devices and Computing (ESDC)","volume":"46 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":"134442857","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：