Pub Date : 2023-02-22DOI: 10.1109/ICECCT56650.2023.10179671
C. Marimuthu, B. Priyanka
Currently, cybersecurity plays a crucial role in various fields to ensure secure data communication. True Random Number Generators (TRNGs) are crucial components for many critical security applications. While analog-based entropy sources are often used in digital-based solutions, there is a high demand for digital-based solutions, especially for Field Programmable Gate Array (FPGA)-based digital systems. A unique technique has been developed to simplify the design of TRNGs on FPGA devices. This technique utilizes the runtime capabilities of the hardware primitives provided by the Digital Clock Manager (DCM) to adjust the phase shift between two clock signals. An auto-tuning approach automatically adjusts the phase difference between the clock signals to force one or more flip-flops (FFs) to enter the metastability zone, which is used as a source of unpredictability in the system. Additionally, the fast carry-chain hardware primitive is uniquely used to further enhance the randomness of the generated bits. Lastly, a powerful on-chip post-processing strategy is employed to prevent any interference with the TRNG output. This work was implemented in verilog HDL, with 32 and 64 data width, and synthesized in Xilinx Zynq FPGA. The characteristics of the TRNG design were evaluated based on area, delay, and power consumption.
{"title":"FPGA Implementation of High Speed 64-Bit Data Width True Random Number Generator using Clock Managers With Metastability","authors":"C. Marimuthu, B. Priyanka","doi":"10.1109/ICECCT56650.2023.10179671","DOIUrl":"https://doi.org/10.1109/ICECCT56650.2023.10179671","url":null,"abstract":"Currently, cybersecurity plays a crucial role in various fields to ensure secure data communication. True Random Number Generators (TRNGs) are crucial components for many critical security applications. While analog-based entropy sources are often used in digital-based solutions, there is a high demand for digital-based solutions, especially for Field Programmable Gate Array (FPGA)-based digital systems. A unique technique has been developed to simplify the design of TRNGs on FPGA devices. This technique utilizes the runtime capabilities of the hardware primitives provided by the Digital Clock Manager (DCM) to adjust the phase shift between two clock signals. An auto-tuning approach automatically adjusts the phase difference between the clock signals to force one or more flip-flops (FFs) to enter the metastability zone, which is used as a source of unpredictability in the system. Additionally, the fast carry-chain hardware primitive is uniquely used to further enhance the randomness of the generated bits. Lastly, a powerful on-chip post-processing strategy is employed to prevent any interference with the TRNG output. This work was implemented in verilog HDL, with 32 and 64 data width, and synthesized in Xilinx Zynq FPGA. The characteristics of the TRNG design were evaluated based on area, delay, and power consumption.","PeriodicalId":180790,"journal":{"name":"2023 Fifth International Conference on Electrical, Computer and Communication Technologies (ICECCT)","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131283552","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 : 2023-02-22DOI: 10.1109/ICECCT56650.2023.10179834
V. Venkateshwarlu, Durgunala Ranjith, Amireddy Raju
In cloud computing environment DDoS attacks are continually evolving with intelligent strategies. Low-rate DDoS attack is one such strategy that make it difficult to detect attack. At the same time, cloud infrastructure is also evolving rapidly. Container based technology enables cloud computing to have lightweight approaches in resource utilization and flexibility in scaling services. The existing DDoS attack detection methods used in cloud computing are not adequate when adversaries employ the modality known low-rate DDoS attack. There is need for an approach that not only detects the attack but also defeat the attack as much as possible. Towards this end, in this paper, we proposed a framework named Low-Rate DDoS Attack Detection Framework (LRDADF). Since low-rate DDoS attacks are difficult to be defeated, we proposed a mathematical model to realize mitigation strategy besides employing deep learning methods to have effective means of detecting such attacks. We proposed an algorithm named Hybrid Approach for Low-Rate DDoS Detection (HA-LRDD). The algorithm uses an Artificial Intelligence (AI) enabled methods comprising of deep Convolutional Neural Network (CNN) and deep autoencoder. Another algorithm known as Dynamic Low-Rate DDoS Mitigation (DLDM) is used to minimize the effect of the attack after detection or even defeat the attack by ensuring the smooth functioning of cloud infrastructure which is under attack. Extensive simulation study revealed that the proposed framework is able to detect low-rate DDoS attacks and also mitigate the attacks to ensure there is acceptable quality of service in cloud computing environments.
{"title":"LRDADF: An AI Enabled Framework for Detecting Low-Rate DDoS Attacks in Cloud Computing Environments","authors":"V. Venkateshwarlu, Durgunala Ranjith, Amireddy Raju","doi":"10.1109/ICECCT56650.2023.10179834","DOIUrl":"https://doi.org/10.1109/ICECCT56650.2023.10179834","url":null,"abstract":"In cloud computing environment DDoS attacks are continually evolving with intelligent strategies. Low-rate DDoS attack is one such strategy that make it difficult to detect attack. At the same time, cloud infrastructure is also evolving rapidly. Container based technology enables cloud computing to have lightweight approaches in resource utilization and flexibility in scaling services. The existing DDoS attack detection methods used in cloud computing are not adequate when adversaries employ the modality known low-rate DDoS attack. There is need for an approach that not only detects the attack but also defeat the attack as much as possible. Towards this end, in this paper, we proposed a framework named Low-Rate DDoS Attack Detection Framework (LRDADF). Since low-rate DDoS attacks are difficult to be defeated, we proposed a mathematical model to realize mitigation strategy besides employing deep learning methods to have effective means of detecting such attacks. We proposed an algorithm named Hybrid Approach for Low-Rate DDoS Detection (HA-LRDD). The algorithm uses an Artificial Intelligence (AI) enabled methods comprising of deep Convolutional Neural Network (CNN) and deep autoencoder. Another algorithm known as Dynamic Low-Rate DDoS Mitigation (DLDM) is used to minimize the effect of the attack after detection or even defeat the attack by ensuring the smooth functioning of cloud infrastructure which is under attack. Extensive simulation study revealed that the proposed framework is able to detect low-rate DDoS attacks and also mitigate the attacks to ensure there is acceptable quality of service in cloud computing environments.","PeriodicalId":180790,"journal":{"name":"2023 Fifth International Conference on Electrical, Computer and Communication Technologies (ICECCT)","volume":"277 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124230828","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 : 2023-02-22DOI: 10.1109/ICECCT56650.2023.10179703
Thirunavukkuarasu G, Nithin R, N. S, Rajalingam J
The access points used for indoor applications with WLAN and WiFi uses 2.45 GHz band nowadays. In the next generation wireless communication, it is proposed to use sub-6 GHz band also for indoor radio coverage. It requires upgraded hardware and coverage. This project work proposes a design of dual band microstrip patch antenna to use this band and also the existing band for radio communication at indoors. The objective of this project is to design a dual band antenna for indoor access point (AP) receiver. The proposed antenna design works in the frequency range of 1–3 GHz and 5–7 GHz with one narrow band frequency and another a wide band frequency. The antenna is designed and the simulation results shows that it is resonating at 2.6 GHz, 3.24 GHz, 5.42 GHz and 6.7 GHz.
{"title":"Dual-band Patch Antenna for Next Generation Indoor Access Point Receiver","authors":"Thirunavukkuarasu G, Nithin R, N. S, Rajalingam J","doi":"10.1109/ICECCT56650.2023.10179703","DOIUrl":"https://doi.org/10.1109/ICECCT56650.2023.10179703","url":null,"abstract":"The access points used for indoor applications with WLAN and WiFi uses 2.45 GHz band nowadays. In the next generation wireless communication, it is proposed to use sub-6 GHz band also for indoor radio coverage. It requires upgraded hardware and coverage. This project work proposes a design of dual band microstrip patch antenna to use this band and also the existing band for radio communication at indoors. The objective of this project is to design a dual band antenna for indoor access point (AP) receiver. The proposed antenna design works in the frequency range of 1–3 GHz and 5–7 GHz with one narrow band frequency and another a wide band frequency. The antenna is designed and the simulation results shows that it is resonating at 2.6 GHz, 3.24 GHz, 5.42 GHz and 6.7 GHz.","PeriodicalId":180790,"journal":{"name":"2023 Fifth International Conference on Electrical, Computer and Communication Technologies (ICECCT)","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114498948","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 : 2023-02-22DOI: 10.1109/ICECCT56650.2023.10179798
Qasim Mohsin Luhaib, R. K. Ajeena
In this work, a group ring (GR), which is defined on an elliptic curve over a prime field, is used to create new versions of public key cryptosystems. First version is elliptic curve group ring ElGamal public key (ECGR-EPKC). On the ECGR- EPKC, a public key, ciphertext and decrypting a plaintext are computing in according to the scalar multiplication that defined on the ECGR over a prime field. Two cases have been considered on the proposed ECGR-EPKC. First case, with an order of ring is less than or equal to the elliptic curve group (ECG). Whereas, on second case, an order of a ring is greater than the order of ECG. Also, the hybrid algorithm of RSA and EPKC based on the ECGR is proposed. The security of the proposed versions, ECGR-EPKE and RSA-EPKC based on ECGR are determined. The proposed RSA-EPKC based on ECGR considers as a more secure for communication schemes in compare with the proposed ECGR-EPKC in two cases and with original algorithms of EPKC and RSA alone.
{"title":"New versions Elliptic Curve Group Ring Public Key Cryptosystems","authors":"Qasim Mohsin Luhaib, R. K. Ajeena","doi":"10.1109/ICECCT56650.2023.10179798","DOIUrl":"https://doi.org/10.1109/ICECCT56650.2023.10179798","url":null,"abstract":"In this work, a group ring (GR), which is defined on an elliptic curve over a prime field, is used to create new versions of public key cryptosystems. First version is elliptic curve group ring ElGamal public key (ECGR-EPKC). On the ECGR- EPKC, a public key, ciphertext and decrypting a plaintext are computing in according to the scalar multiplication that defined on the ECGR over a prime field. Two cases have been considered on the proposed ECGR-EPKC. First case, with an order of ring is less than or equal to the elliptic curve group (ECG). Whereas, on second case, an order of a ring is greater than the order of ECG. Also, the hybrid algorithm of RSA and EPKC based on the ECGR is proposed. The security of the proposed versions, ECGR-EPKE and RSA-EPKC based on ECGR are determined. The proposed RSA-EPKC based on ECGR considers as a more secure for communication schemes in compare with the proposed ECGR-EPKC in two cases and with original algorithms of EPKC and RSA alone.","PeriodicalId":180790,"journal":{"name":"2023 Fifth International Conference on Electrical, Computer and Communication Technologies (ICECCT)","volume":"86 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114905387","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 : 2023-02-22DOI: 10.1109/ICECCT56650.2023.10179796
Hafsa Ouchra, A. Belangour, Allae Erraissi
Geo-spatial artificial intelligence (GeoAI) is an interdisciplinary field that combines techniques and methods from engineering, computer science, statistics, and space science to analyze and model spatial and temporal phenomena using artificial intelligence (AI) methods. This subject has a large influence on society and the economy since it focuses on real-world challenges. Artificial intelligence is fast developing in the automatic recognition of characteristics in geospatial data such as satellite photos, aerial photographs, etc. The computer community is very interested in satellite imagery because it wants to help machines understand their surroundings by analyzing satellite data. This type of treatment has the potential to test large areas at low cost. Remote sensing and geographic data enable collection, analysis, and processing of global observation data for civilian and military applications. This document provides an overview of GeoAI methods in urban planning, including the definition of GeoSpatial Artificial Intelligence and the distinctions between GeoAI and traditional AI. Different types of geospatial data satellites are also depicted. Integrating Geographic Information System (GIS) with Artificial Intelligence and using GeoAI tools and techniques are critical steps in successful geographic data analysis.
{"title":"An overview of GeoSpatial Artificial Intelligence technologies for city planning and development","authors":"Hafsa Ouchra, A. Belangour, Allae Erraissi","doi":"10.1109/ICECCT56650.2023.10179796","DOIUrl":"https://doi.org/10.1109/ICECCT56650.2023.10179796","url":null,"abstract":"Geo-spatial artificial intelligence (GeoAI) is an interdisciplinary field that combines techniques and methods from engineering, computer science, statistics, and space science to analyze and model spatial and temporal phenomena using artificial intelligence (AI) methods. This subject has a large influence on society and the economy since it focuses on real-world challenges. Artificial intelligence is fast developing in the automatic recognition of characteristics in geospatial data such as satellite photos, aerial photographs, etc. The computer community is very interested in satellite imagery because it wants to help machines understand their surroundings by analyzing satellite data. This type of treatment has the potential to test large areas at low cost. Remote sensing and geographic data enable collection, analysis, and processing of global observation data for civilian and military applications. This document provides an overview of GeoAI methods in urban planning, including the definition of GeoSpatial Artificial Intelligence and the distinctions between GeoAI and traditional AI. Different types of geospatial data satellites are also depicted. Integrating Geographic Information System (GIS) with Artificial Intelligence and using GeoAI tools and techniques are critical steps in successful geographic data analysis.","PeriodicalId":180790,"journal":{"name":"2023 Fifth International Conference on Electrical, Computer and Communication Technologies (ICECCT)","volume":"48 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123596601","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 : 2023-02-22DOI: 10.1109/ICECCT56650.2023.10179614
Parmeshwar Birajadar, Abhijit Shete
In the literature various algorithms for content based image retrieval (CBIR) were proposed on different variety of databases such as texture, face, and medical images. It is observed that the CBIR performance strongly depends on effective extraction of image features such as color, texture, and shape. In the proposed research work, contourlet transform based CBIR system is proposed for texture images. Similar to wavelets, texture images can be characterized via marginal distributions of subband coefficients of recently introduced contourlets. In addition, contourlets have two new properties namely directionality and anisotropy which makes them powerful in efficient representation of natural texture images. Hence, in the proposed research work, we have analyzed and compared the performance of both wavelet and contourlet schemes for content based texture image retrieval. It is observed that contourlet based method outperforms the wavelet method for highly texture images in terms of retrieval efficiency.
{"title":"A Comparative Performance Analysis of CBIR using Wavelet and Contourlet Transform","authors":"Parmeshwar Birajadar, Abhijit Shete","doi":"10.1109/ICECCT56650.2023.10179614","DOIUrl":"https://doi.org/10.1109/ICECCT56650.2023.10179614","url":null,"abstract":"In the literature various algorithms for content based image retrieval (CBIR) were proposed on different variety of databases such as texture, face, and medical images. It is observed that the CBIR performance strongly depends on effective extraction of image features such as color, texture, and shape. In the proposed research work, contourlet transform based CBIR system is proposed for texture images. Similar to wavelets, texture images can be characterized via marginal distributions of subband coefficients of recently introduced contourlets. In addition, contourlets have two new properties namely directionality and anisotropy which makes them powerful in efficient representation of natural texture images. Hence, in the proposed research work, we have analyzed and compared the performance of both wavelet and contourlet schemes for content based texture image retrieval. It is observed that contourlet based method outperforms the wavelet method for highly texture images in terms of retrieval efficiency.","PeriodicalId":180790,"journal":{"name":"2023 Fifth International Conference on Electrical, Computer and Communication Technologies (ICECCT)","volume":"144 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125777156","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 : 2023-02-22DOI: 10.1109/ICECCT56650.2023.10179825
N. Pradhan, B. Chaudhari, P. Khandekar
Recent years show the rapid adoption of Industrial Internet of Things (IIoT) technologies in automation and manufacturing. The IEEE 802.15.4e has evolved the industry in many different ways. Its time synchronization and channel hopping techniques are crucial for reliable operations in resource-constrained networks. With the advent of the 6TiSCH, it became feasible to connect industrial operational technology with an IPv6 network. It enabled determinism and quality of service in IIoT. Several 6TiSCH scheduling functions are being developed, optimizing different performance metrics. Network traffic and topology are the challenges in designing an efficient schedule. In this paper, the performance of the traffic-aware schedule has been analysed under different network scenarios. This technique designs a slotframe by considering the node's location and the traffic condition in the network. The result shows that the end-to-end latency in this technique is reduced up to 14% with similar throughput compared to the existing techniques.
{"title":"Enhanced Time Slotted Channel Scheduling for Constrained Networks","authors":"N. Pradhan, B. Chaudhari, P. Khandekar","doi":"10.1109/ICECCT56650.2023.10179825","DOIUrl":"https://doi.org/10.1109/ICECCT56650.2023.10179825","url":null,"abstract":"Recent years show the rapid adoption of Industrial Internet of Things (IIoT) technologies in automation and manufacturing. The IEEE 802.15.4e has evolved the industry in many different ways. Its time synchronization and channel hopping techniques are crucial for reliable operations in resource-constrained networks. With the advent of the 6TiSCH, it became feasible to connect industrial operational technology with an IPv6 network. It enabled determinism and quality of service in IIoT. Several 6TiSCH scheduling functions are being developed, optimizing different performance metrics. Network traffic and topology are the challenges in designing an efficient schedule. In this paper, the performance of the traffic-aware schedule has been analysed under different network scenarios. This technique designs a slotframe by considering the node's location and the traffic condition in the network. The result shows that the end-to-end latency in this technique is reduced up to 14% with similar throughput compared to the existing techniques.","PeriodicalId":180790,"journal":{"name":"2023 Fifth International Conference on Electrical, Computer and Communication Technologies (ICECCT)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129698698","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 : 2023-02-22DOI: 10.1109/ICECCT56650.2023.10179733
V. Sreevalli, M. Rathaiah, R. Kiranmayi, P. Bharath Kumar, K. Nagabhushanam, Ahamadia. K Firdouse Kowsar
Proportional-Integral-Derivative (PID) controller is the most popular technique used in several industrial applications and has done more research work throughout many years. Being popular, the PID controller exhibits undesirable output response for some open loop unstable process such as integrating ones. This paper projects a design of smith predictor with an outer loop Proportional-Integral (PI) controller and interior loop Proportional-Derivative (PD) controller. The smith predictor uses plant process as a predictive model and it eliminates the delay part of the closed loop characteristic equation. Three various examples of integrating plants with time delay are carried out in simulation study to evaluate the obtained performance of projected method. The results from the simulation study are compared with the optimal PI-PD method. To illustrate the robustness of projected method, parameter uncertainties are included to the process and the following results are exhibited. The projected method is more robust to such dissimilarities. Disturbance of magnitude 0.1 and a white noise of 0.0001 magnitude are added to the system and the improved performance are shown in the simulation results. The proposed control algorithm will be designed and implemented in MATLAB/SIMULINK environment.
{"title":"Enhanced Design of PI-PD Controller based on Smith Predictor for Integrating Processes with Time Delay","authors":"V. Sreevalli, M. Rathaiah, R. Kiranmayi, P. Bharath Kumar, K. Nagabhushanam, Ahamadia. K Firdouse Kowsar","doi":"10.1109/ICECCT56650.2023.10179733","DOIUrl":"https://doi.org/10.1109/ICECCT56650.2023.10179733","url":null,"abstract":"Proportional-Integral-Derivative (PID) controller is the most popular technique used in several industrial applications and has done more research work throughout many years. Being popular, the PID controller exhibits undesirable output response for some open loop unstable process such as integrating ones. This paper projects a design of smith predictor with an outer loop Proportional-Integral (PI) controller and interior loop Proportional-Derivative (PD) controller. The smith predictor uses plant process as a predictive model and it eliminates the delay part of the closed loop characteristic equation. Three various examples of integrating plants with time delay are carried out in simulation study to evaluate the obtained performance of projected method. The results from the simulation study are compared with the optimal PI-PD method. To illustrate the robustness of projected method, parameter uncertainties are included to the process and the following results are exhibited. The projected method is more robust to such dissimilarities. Disturbance of magnitude 0.1 and a white noise of 0.0001 magnitude are added to the system and the improved performance are shown in the simulation results. The proposed control algorithm will be designed and implemented in MATLAB/SIMULINK environment.","PeriodicalId":180790,"journal":{"name":"2023 Fifth International Conference on Electrical, Computer and Communication Technologies (ICECCT)","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128540429","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 : 2023-02-22DOI: 10.1109/ICECCT56650.2023.10179689
M. Devi, J. Pandian, Aparna Joshi, Yeluri Praveen
The assessment of how two distinct images are equal are indeed called image similarity and consistency. In other words, it measures how much the intensity patterns in two images are comparable to one another. In order to achieve this, researchers examine the image descriptors recursively in order to identify descriptor pairs that are comparable. The two images are deemed comparable if the number of related descriptors exceeds a predetermined threshold and both images exhibit the very same entity. The computation of image similarity is used for various applications which graves to be the mandatory process for production of the application. With this intent, the Fashion MNIST dataset from KAGGLE is used for implementing the image similarity estimation. This paper proposes Batch Normalized Siamese Network (BNSN) deep learning based model for computing the image similarity. The BNSN model is designed with two subnetworks that generates feature vectors of two input images. The lambda batch normalization is performed with single dense layer to predict the image similarity with label 0 indicating the identical images and label 1 denoting the different images. The 30,000 training images were fitted with BNSN and tested with 30,000 images. Python was implemented on a Geforce Tesla V100 NVidia Graphics card webserver with a batch size of 64 and 30 training epochs. The training images are also tested with traditional image similarity method and implementation of proposed BNSN shows the accuracy of 91.91%, Precision of 92.93%, Recall of 90.72% and FScore of 91.81%.
对两幅不同的图像如何相等的评估确实被称为图像相似性和一致性。换句话说，它测量的是两幅图像的强度模式之间的可比性。为了实现这一点，研究人员递归地检查图像描述符，以识别具有可比性的描述符对。如果相关描述符的数量超过预定的阈值并且两个图像显示非常相同的实体，则认为这两个图像具有可比性。图像相似度的计算用于各种应用程序，这是应用程序生产的强制性过程。出于这个目的，我们使用KAGGLE的Fashion MNIST数据集来实现图像相似性估计。提出了一种基于Batch Normalized Siamese Network (BNSN)深度学习的图像相似度计算模型。BNSN模型设计了两个子网络，分别生成两个输入图像的特征向量。用单致密层进行lambda批处理归一化来预测图像的相似度，标签0表示相同图像，标签1表示不同图像。对3万张训练图像进行BNSN拟合，用3万张图像进行测试。Python在Geforce Tesla V100 NVidia显卡web服务器上实现，批处理大小为64个，训练epoch为30个。采用传统的图像相似度方法对训练图像进行测试，结果表明，BNSN的准确率为91.91%，精密度为92.93%，查全率为90.72%，FScore为91.81%。
{"title":"Batch Normalized Siamese Network Deep Learning Based Image Similarity Estimation","authors":"M. Devi, J. Pandian, Aparna Joshi, Yeluri Praveen","doi":"10.1109/ICECCT56650.2023.10179689","DOIUrl":"https://doi.org/10.1109/ICECCT56650.2023.10179689","url":null,"abstract":"The assessment of how two distinct images are equal are indeed called image similarity and consistency. In other words, it measures how much the intensity patterns in two images are comparable to one another. In order to achieve this, researchers examine the image descriptors recursively in order to identify descriptor pairs that are comparable. The two images are deemed comparable if the number of related descriptors exceeds a predetermined threshold and both images exhibit the very same entity. The computation of image similarity is used for various applications which graves to be the mandatory process for production of the application. With this intent, the Fashion MNIST dataset from KAGGLE is used for implementing the image similarity estimation. This paper proposes Batch Normalized Siamese Network (BNSN) deep learning based model for computing the image similarity. The BNSN model is designed with two subnetworks that generates feature vectors of two input images. The lambda batch normalization is performed with single dense layer to predict the image similarity with label 0 indicating the identical images and label 1 denoting the different images. The 30,000 training images were fitted with BNSN and tested with 30,000 images. Python was implemented on a Geforce Tesla V100 NVidia Graphics card webserver with a batch size of 64 and 30 training epochs. The training images are also tested with traditional image similarity method and implementation of proposed BNSN shows the accuracy of 91.91%, Precision of 92.93%, Recall of 90.72% and FScore of 91.81%.","PeriodicalId":180790,"journal":{"name":"2023 Fifth International Conference on Electrical, Computer and Communication Technologies (ICECCT)","volume":"65 2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129817669","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 : 2023-02-22DOI: 10.1109/ICECCT56650.2023.10179616
Akash Bagchi, Akshaya Nadathur, Pranjal Padakannaya, Siddiq Khan, Siddiq Khan
Young children, especially those on the autistic spectrum, often do not learn educational subjects and life skills in the same manner as neurotypicals. Recreational activities like games, puzzles and co-curriculars are known to provide additional opportunities for learning while also engaging the child. The proposed project is aimed to design and implement an interactive game that will help shape the behavioral, cognitive and/or life skills of children with Autism Spectrum Disorders (ASD). Such an implementation will be carried out using the Unity engine, which utilizes the language C# for the source code. The final result is expected to be a working solution of a game that will aid healthy development of autistic patients in early childhood, and will be deployed for public use.
{"title":"Building Accessible Video Games for Children with an Autism Spectrum Disorder","authors":"Akash Bagchi, Akshaya Nadathur, Pranjal Padakannaya, Siddiq Khan, Siddiq Khan","doi":"10.1109/ICECCT56650.2023.10179616","DOIUrl":"https://doi.org/10.1109/ICECCT56650.2023.10179616","url":null,"abstract":"Young children, especially those on the autistic spectrum, often do not learn educational subjects and life skills in the same manner as neurotypicals. Recreational activities like games, puzzles and co-curriculars are known to provide additional opportunities for learning while also engaging the child. The proposed project is aimed to design and implement an interactive game that will help shape the behavioral, cognitive and/or life skills of children with Autism Spectrum Disorders (ASD). Such an implementation will be carried out using the Unity engine, which utilizes the language C# for the source code. The final result is expected to be a working solution of a game that will aid healthy development of autistic patients in early childhood, and will be deployed for public use.","PeriodicalId":180790,"journal":{"name":"2023 Fifth International Conference on Electrical, Computer and Communication Technologies (ICECCT)","volume":"99 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127545441","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：