Pub Date : 2023-03-17DOI: 10.1109/iCoMET57998.2023.10099223
Wajahat Akbar, A. Soomro, M. Ullah, Muhammad Inam Ul Haq, Sana Ullah Khan, Tahir Ali Shah
Early detection of plant diseases is crucial before plant growth is affected. Plant diseases have been detected and classified using a variety of machine learning (ML) models in the past. Deep Learning (DL) appears to have great potential in terms of increased accuracy; however, in agricultural applications of Convolutional Neural Networks (CNN) has widely been utilised by researchers. CNNs are so effective at identifying plant species, managing yields, detecting weeds, managing soil, and water, counting fruits, detecting diseases and pests, and evaluating plant nutrient status. A farmer can diagnose plant diseases quickly and accurately with an automated disease detection system. To speed up crop diagnosis, plant leaf disease detection systems must be automated. In this paper, we evaluated twelve different models on a new plant diseases dataset and demonstrated that the most accurate model was Densenet169. In training and validation, the accuracy was 97.2% and 97.8%, respectively.
{"title":"Performance Evaluation of Deep Learning Models for Leaf Disease Detection: A Comparative Study","authors":"Wajahat Akbar, A. Soomro, M. Ullah, Muhammad Inam Ul Haq, Sana Ullah Khan, Tahir Ali Shah","doi":"10.1109/iCoMET57998.2023.10099223","DOIUrl":"https://doi.org/10.1109/iCoMET57998.2023.10099223","url":null,"abstract":"Early detection of plant diseases is crucial before plant growth is affected. Plant diseases have been detected and classified using a variety of machine learning (ML) models in the past. Deep Learning (DL) appears to have great potential in terms of increased accuracy; however, in agricultural applications of Convolutional Neural Networks (CNN) has widely been utilised by researchers. CNNs are so effective at identifying plant species, managing yields, detecting weeds, managing soil, and water, counting fruits, detecting diseases and pests, and evaluating plant nutrient status. A farmer can diagnose plant diseases quickly and accurately with an automated disease detection system. To speed up crop diagnosis, plant leaf disease detection systems must be automated. In this paper, we evaluated twelve different models on a new plant diseases dataset and demonstrated that the most accurate model was Densenet169. In training and validation, the accuracy was 97.2% and 97.8%, respectively.","PeriodicalId":369792,"journal":{"name":"2023 4th International Conference on Computing, Mathematics and Engineering Technologies (iCoMET)","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125903798","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 : 2023-03-17DOI: 10.1109/iCoMET57998.2023.10099200
Hina Magsi, M. Shah, Syed Hadi Hussain Shah, Faiza, A. Hussain, Arsalan Muhammad Soomar, F. Chachar
Positioning, Navigation, and Timing (PNT) information play a vital role in everyday life of common persons. People greatly rely on Global Navigation Satellite System (GNSS)-enabled applications for navigation to reach their desired destination. However, GNSS navigation performance is highly degraded in urban environments due to the high probability of signal interruption, multipath (MP), and/or non-line-of-sight (NLOS) signal reception. Multipath and NLOS being the major causes of disrupted positioning performance for GNSS in urban environments. The navigation signals encountered various environmental factors they are reflected, refracted, diffracted, and completely blocked by high-roof buildings, bridges, and trees, thus leading to severe uncertainties in position estimation. GNSS system has gained notable advancements in terms of number of satellites, satellite geometry and signal quality. In this paper newly established constellation BeiDou Navigation Satellite System (BDS-3) performance is quantified with respect to environmental changes and compared in terms of positional accuracy. The paper also discussed the innovative current developments and status of BDS-3 in 2023. For this reason, series of field experiments were carried out at clear open and urban environment with BDS-3 and GPS mode during the observation time of 6 hours. The BDS-3 system is configured for data logging and used for the first time at Pakistan region. The positioning and navigation performance of BDS-3 is evaluated by utilizing key performance indicators e.g satellite availability, geometric distribution in terms of PDOP, and statistical accuracy measures (i.e., Circular Error Probable (CEP) and Distance Root Mean Square (DRMS)). The experimental results shows that BDS-3 provides more number of satellites, favorable satellite geometry and reduced position error compared to GPS constellation in clear open sky environment. In urban environment it is observed that BDS-3 performance is reduced/dropped due to obstructions that leads to increase the positioning inaccuracies. It is comprehended that BDS-3 system performance is less affected in urban site in terms of satellite availability, PDOP and position error as compared to GPS system. The statistical positional accuracy for BDS-3 and GPS found to be similar at clear open sky environment. BDS is more resilient to environmental factors.
{"title":"Performance Analysis and Quantification of BeiDou Navigation Satellite System (BDS-3)","authors":"Hina Magsi, M. Shah, Syed Hadi Hussain Shah, Faiza, A. Hussain, Arsalan Muhammad Soomar, F. Chachar","doi":"10.1109/iCoMET57998.2023.10099200","DOIUrl":"https://doi.org/10.1109/iCoMET57998.2023.10099200","url":null,"abstract":"Positioning, Navigation, and Timing (PNT) information play a vital role in everyday life of common persons. People greatly rely on Global Navigation Satellite System (GNSS)-enabled applications for navigation to reach their desired destination. However, GNSS navigation performance is highly degraded in urban environments due to the high probability of signal interruption, multipath (MP), and/or non-line-of-sight (NLOS) signal reception. Multipath and NLOS being the major causes of disrupted positioning performance for GNSS in urban environments. The navigation signals encountered various environmental factors they are reflected, refracted, diffracted, and completely blocked by high-roof buildings, bridges, and trees, thus leading to severe uncertainties in position estimation. GNSS system has gained notable advancements in terms of number of satellites, satellite geometry and signal quality. In this paper newly established constellation BeiDou Navigation Satellite System (BDS-3) performance is quantified with respect to environmental changes and compared in terms of positional accuracy. The paper also discussed the innovative current developments and status of BDS-3 in 2023. For this reason, series of field experiments were carried out at clear open and urban environment with BDS-3 and GPS mode during the observation time of 6 hours. The BDS-3 system is configured for data logging and used for the first time at Pakistan region. The positioning and navigation performance of BDS-3 is evaluated by utilizing key performance indicators e.g satellite availability, geometric distribution in terms of PDOP, and statistical accuracy measures (i.e., Circular Error Probable (CEP) and Distance Root Mean Square (DRMS)). The experimental results shows that BDS-3 provides more number of satellites, favorable satellite geometry and reduced position error compared to GPS constellation in clear open sky environment. In urban environment it is observed that BDS-3 performance is reduced/dropped due to obstructions that leads to increase the positioning inaccuracies. It is comprehended that BDS-3 system performance is less affected in urban site in terms of satellite availability, PDOP and position error as compared to GPS system. The statistical positional accuracy for BDS-3 and GPS found to be similar at clear open sky environment. BDS is more resilient to environmental factors.","PeriodicalId":369792,"journal":{"name":"2023 4th International Conference on Computing, Mathematics and Engineering Technologies (iCoMET)","volume":"51 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128260076","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 : 2023-03-17DOI: 10.1109/iCoMET57998.2023.10099182
Komal Memon, F. Umrani, Attiya Baqai, Zafi Sherhan Syed
Agriculture is defined as a branch of science related to soil cultivation, crops growth and animals rearing for food supply and production of wool among other products. It plays pivotal role on the GDP (Gross Domestic Product) of any country. One way to increase the Agri-based GDP is to use Precision Agriculture (PA) whereby information technology is introduced into traditional agriculture. Agricultural expert systems have progressed from basic record keeping to large-scale farm management information systems (FMISs) that are utilized for crop prediction, crop disease detection, farm scheduling, and water monitoring. Precision Agriculture based framework allows us to use proper amount of water, fertilizers and seeds thereby, increasing the productivity of the agriculture fields by monitoring the environmental/soil parameters of agricultural land such as: pH, soil temperature, soil humidity, soil moisture and atmospheric pressure. In this paper an extensive literature review has been done on the architecture, hardware, communication protocol, and data acquisition infrastructure for crop monitoring systems along with the survey of different mobile applications and machine learning models used in precision agriculture.
{"title":"A Review Based On Comparative Analysis of Techniques Used in Precision Agriculture","authors":"Komal Memon, F. Umrani, Attiya Baqai, Zafi Sherhan Syed","doi":"10.1109/iCoMET57998.2023.10099182","DOIUrl":"https://doi.org/10.1109/iCoMET57998.2023.10099182","url":null,"abstract":"Agriculture is defined as a branch of science related to soil cultivation, crops growth and animals rearing for food supply and production of wool among other products. It plays pivotal role on the GDP (Gross Domestic Product) of any country. One way to increase the Agri-based GDP is to use Precision Agriculture (PA) whereby information technology is introduced into traditional agriculture. Agricultural expert systems have progressed from basic record keeping to large-scale farm management information systems (FMISs) that are utilized for crop prediction, crop disease detection, farm scheduling, and water monitoring. Precision Agriculture based framework allows us to use proper amount of water, fertilizers and seeds thereby, increasing the productivity of the agriculture fields by monitoring the environmental/soil parameters of agricultural land such as: pH, soil temperature, soil humidity, soil moisture and atmospheric pressure. In this paper an extensive literature review has been done on the architecture, hardware, communication protocol, and data acquisition infrastructure for crop monitoring systems along with the survey of different mobile applications and machine learning models used in precision agriculture.","PeriodicalId":369792,"journal":{"name":"2023 4th International Conference on Computing, Mathematics and Engineering Technologies (iCoMET)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130078576","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 : 2023-03-17DOI: 10.1109/iCoMET57998.2023.10099311
Wajahat Akbar, Muhammad Inam Ul Haq, A. Soomro, Sher Muhammad Daudpota, Ali Shariq Imran, M. Ullah
Radiology reports are the primary medium through which physicians communicate with patients and share diagnoses from medical scans. Examples include radiology reports for chest X-Rays and CT scans. Chest X-Ray images are frequently employed in clinical screening and diagnosis. However, writing medical reports for the X-Ray is tedious, error-prone, and time-consuming, even for experienced radiologists. The modern world of clinical practice demands that a radiologist with specialized training manually evaluate chest X-Ray and report the findings. Therefore, this paper explores the ability of artificial intelligence (AI) to automate diagnosing diseases through chest X-Rays and accurately generate radiology reports to alleviate the burdens of medical doctors. Automating this manual process could streamline a clinical workflow, and healthcare quality could be improved. The conventional AI-based abstract methods provide fluent but clinically incorrect radiology reports. The proposed Gated Recurrent Unit (GRU) based model provides both stan-dard language generation and clinical coherence. The model is evaluated on the Indiana University dataset with commonly-used metrics BLEU and ROUGE-L. Empirical evaluations illustrate that the proposed approach can make more precise diagnoses and generate more fluent and precise reports than existing baselines.
{"title":"Automated Report Generation: A GRU Based Method for Chest X-Rays","authors":"Wajahat Akbar, Muhammad Inam Ul Haq, A. Soomro, Sher Muhammad Daudpota, Ali Shariq Imran, M. Ullah","doi":"10.1109/iCoMET57998.2023.10099311","DOIUrl":"https://doi.org/10.1109/iCoMET57998.2023.10099311","url":null,"abstract":"Radiology reports are the primary medium through which physicians communicate with patients and share diagnoses from medical scans. Examples include radiology reports for chest X-Rays and CT scans. Chest X-Ray images are frequently employed in clinical screening and diagnosis. However, writing medical reports for the X-Ray is tedious, error-prone, and time-consuming, even for experienced radiologists. The modern world of clinical practice demands that a radiologist with specialized training manually evaluate chest X-Ray and report the findings. Therefore, this paper explores the ability of artificial intelligence (AI) to automate diagnosing diseases through chest X-Rays and accurately generate radiology reports to alleviate the burdens of medical doctors. Automating this manual process could streamline a clinical workflow, and healthcare quality could be improved. The conventional AI-based abstract methods provide fluent but clinically incorrect radiology reports. The proposed Gated Recurrent Unit (GRU) based model provides both stan-dard language generation and clinical coherence. The model is evaluated on the Indiana University dataset with commonly-used metrics BLEU and ROUGE-L. Empirical evaluations illustrate that the proposed approach can make more precise diagnoses and generate more fluent and precise reports than existing baselines.","PeriodicalId":369792,"journal":{"name":"2023 4th International Conference on Computing, Mathematics and Engineering Technologies (iCoMET)","volume":"71 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126966661","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 : 2023-03-17DOI: 10.1109/iCoMET57998.2023.10099072
Muhammad Waseem, I. Ali
In this study, we propose the OISMC approach for robust control under various unpredictable uncertainties. The approach uses optimal control (LQR gains) to deliver state feedback gains that help fulfill one of our objectives, which is to reduce the cost function in the presence of leading response, coupling effects on each link, and external control input disturbances. The approach offers several major benefits, including fast recovery reaction, reduced chattering, excellent tracking performance, low energy consumption, easy implementation, and solidity against uncertainties. Furthermore, the approach is theoretically sound and supported by extensive simulation testing and the Lyapunov stability theory. In our study, we conducted simulations using Mathematica computer software. We anticipate that the proposed approach will be helpful in developing reliable and superior tracking control for all types of uncertain multivariate systems.
{"title":"Tracking error control of robotic manipulator using optimal integral sliding mode control in the presence of external disturbances","authors":"Muhammad Waseem, I. Ali","doi":"10.1109/iCoMET57998.2023.10099072","DOIUrl":"https://doi.org/10.1109/iCoMET57998.2023.10099072","url":null,"abstract":"In this study, we propose the OISMC approach for robust control under various unpredictable uncertainties. The approach uses optimal control (LQR gains) to deliver state feedback gains that help fulfill one of our objectives, which is to reduce the cost function in the presence of leading response, coupling effects on each link, and external control input disturbances. The approach offers several major benefits, including fast recovery reaction, reduced chattering, excellent tracking performance, low energy consumption, easy implementation, and solidity against uncertainties. Furthermore, the approach is theoretically sound and supported by extensive simulation testing and the Lyapunov stability theory. In our study, we conducted simulations using Mathematica computer software. We anticipate that the proposed approach will be helpful in developing reliable and superior tracking control for all types of uncertain multivariate systems.","PeriodicalId":369792,"journal":{"name":"2023 4th International Conference on Computing, Mathematics and Engineering Technologies (iCoMET)","volume":"62 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130835915","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 : 2023-03-17DOI: 10.1109/iCoMET57998.2023.10099251
S. Kumar, Asif Ali Shaikh, Syed Feroz Shah, Hazoor Bux Lanjwani
In the present article, heat, and mass transfer features of MHD Casson nanofluid flow, is studied with thermal radiation, concentration and thermal slip effects over permeable stretching/shrinking surface by applying Buongiorno model. Governing equations of the present problem which are later reduced by similarity transformations into form of ordinary differential equations. The numerical solutions are achieved utilizing shooting method in MAPLE software. For validation of obtained results by shooting method, obtained results are matched with previously obtained results present in literature. The impact of Casson, magnetic, Brownian motion, Lewis number, porosity, thermophoresis, thermal radiation, Prandtl number, thermal and concentration slip parameters on velocity, temperature and the concentration profiles are examined. The results show that velocity profiles decrease by increasing magnetic, Casson, suction parameters and porosity. Furthermore, temperature profiles increases with rise in thermophoresis, radiation, and Brownian motion parameters. In last, skin friction, Nusselt and Sherwood numbers are acquired at several values of used parameters displayed in graphs.
{"title":"Numerical Investigation of the MHD Casson Nanofluid Flow over Permeable Stretching/Shrinking Surface with Radiation Effects","authors":"S. Kumar, Asif Ali Shaikh, Syed Feroz Shah, Hazoor Bux Lanjwani","doi":"10.1109/iCoMET57998.2023.10099251","DOIUrl":"https://doi.org/10.1109/iCoMET57998.2023.10099251","url":null,"abstract":"In the present article, heat, and mass transfer features of MHD Casson nanofluid flow, is studied with thermal radiation, concentration and thermal slip effects over permeable stretching/shrinking surface by applying Buongiorno model. Governing equations of the present problem which are later reduced by similarity transformations into form of ordinary differential equations. The numerical solutions are achieved utilizing shooting method in MAPLE software. For validation of obtained results by shooting method, obtained results are matched with previously obtained results present in literature. The impact of Casson, magnetic, Brownian motion, Lewis number, porosity, thermophoresis, thermal radiation, Prandtl number, thermal and concentration slip parameters on velocity, temperature and the concentration profiles are examined. The results show that velocity profiles decrease by increasing magnetic, Casson, suction parameters and porosity. Furthermore, temperature profiles increases with rise in thermophoresis, radiation, and Brownian motion parameters. In last, skin friction, Nusselt and Sherwood numbers are acquired at several values of used parameters displayed in graphs.","PeriodicalId":369792,"journal":{"name":"2023 4th International Conference on Computing, Mathematics and Engineering Technologies (iCoMET)","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130863941","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 : 2023-03-17DOI: 10.1109/iCoMET57998.2023.10099055
Mehwish Wahid, Ghufran Ahmed, Shahid Hussain, Asad Ahmed Ansari
Deep learning(DL) is a sub-field of artificial intelligence that mimics the human brain through computation. It has proven its proficiency in different domains, including healthcare. It has shown promising results in various health-care applications, including cancer classification, prognosis, and molecular sub-typing of cancer. Molecular sub-typing provides biological insights regarding cancer heterogeneity that may lead to personalized medicines. The objective of this review is to discuss and compare the different deep learning models used for molecular subtyping along with the different types of omics data used like gene expression data, RNA sequence data, mRNA, and miRNA. We compared and summarized the different models and data types used for the cancer molecular subtyping in a tabular format.
{"title":"A Survey on Cancer Molecular Subtype Classification using Deep learning","authors":"Mehwish Wahid, Ghufran Ahmed, Shahid Hussain, Asad Ahmed Ansari","doi":"10.1109/iCoMET57998.2023.10099055","DOIUrl":"https://doi.org/10.1109/iCoMET57998.2023.10099055","url":null,"abstract":"Deep learning(DL) is a sub-field of artificial intelligence that mimics the human brain through computation. It has proven its proficiency in different domains, including healthcare. It has shown promising results in various health-care applications, including cancer classification, prognosis, and molecular sub-typing of cancer. Molecular sub-typing provides biological insights regarding cancer heterogeneity that may lead to personalized medicines. The objective of this review is to discuss and compare the different deep learning models used for molecular subtyping along with the different types of omics data used like gene expression data, RNA sequence data, mRNA, and miRNA. We compared and summarized the different models and data types used for the cancer molecular subtyping in a tabular format.","PeriodicalId":369792,"journal":{"name":"2023 4th International Conference on Computing, Mathematics and Engineering Technologies (iCoMET)","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122616327","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 : 2023-03-17DOI: 10.1109/iCoMET57998.2023.10099075
Hina Umbrin, M. Aamir, Javed Ferzund, H. Tahir, R. Latif
The field of data science has facilitated the extraction of information from organized and unstructured data. It utilizes several approaches, algorithms, and processes to evaluate complex data effectively. Protein-Protein Interactions (PPIs) are crucial for a variety of chemical processes. This initiative will build predictive models that give a more efficient and straightforward way for PPI prediction to enhance the PPI prediction for high throughput. This work uses the PageRank algorithm for PPI systems' organic properties. PageRank is a method for ranking that can rate the interaction in MIPS datasets. It assigns a value to each interaction and determines the protein IDs with the most significant number of interactions. We have used the Perl programming language, Mlib, and GraphX libraries for PPI predictions. The data suggest that this method yields quicker execution times and good outcomes.
{"title":"Towards a Protein-Protein Interactions Framework using Graph Analytics on Apache Spark","authors":"Hina Umbrin, M. Aamir, Javed Ferzund, H. Tahir, R. Latif","doi":"10.1109/iCoMET57998.2023.10099075","DOIUrl":"https://doi.org/10.1109/iCoMET57998.2023.10099075","url":null,"abstract":"The field of data science has facilitated the extraction of information from organized and unstructured data. It utilizes several approaches, algorithms, and processes to evaluate complex data effectively. Protein-Protein Interactions (PPIs) are crucial for a variety of chemical processes. This initiative will build predictive models that give a more efficient and straightforward way for PPI prediction to enhance the PPI prediction for high throughput. This work uses the PageRank algorithm for PPI systems' organic properties. PageRank is a method for ranking that can rate the interaction in MIPS datasets. It assigns a value to each interaction and determines the protein IDs with the most significant number of interactions. We have used the Perl programming language, Mlib, and GraphX libraries for PPI predictions. The data suggest that this method yields quicker execution times and good outcomes.","PeriodicalId":369792,"journal":{"name":"2023 4th International Conference on Computing, Mathematics and Engineering Technologies (iCoMET)","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122733985","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 : 2023-03-17DOI: 10.1109/iCoMET57998.2023.10099250
Shaharyar Ahmed Khan Tareen, R. H. Raza
Extremely variant image pairs include distorted, deteriorated, and corrupted scenes that have experienced severe geometric, photometric, or non-geometric-non-photometric transformations with respect to their originals. Real world visual data can become extremely dusty, smoky, dark, noisy, motion-blurred, affine, JPEG compressed, occluded, shadowed, virtually invisible, etc. Therefore, matching of extremely variant scenes is an important problem and computer vision solutions must have the capability to yield robust results no matter how complex the visual input is. Similarly, there is a need to evaluate feature detectors for such complex conditions. With standard settings, feature detection, description, and matching algorithms typically fail to produce significant number of correct matches in these types of images. Though, if full potential of the algorithms is applied by using extremely low thresholds, very encouraging results are obtained. In this paper, potential of 14 feature detectors: SIFT, SURF, KAZE, AKAZE, ORB, BRISK, AGAST, FAST, MSER, MSD, GFTT, Harris Corner Detector based GFTT, Harris Laplace Detector, and CenSurE has been evaluated for matching 10 extremely variant image pairs. MSD detected more than 1 million keypoints in one of the images and SIFT exhibited a repeatability score of 99.76% for the extremely noisy image pair but failed to yield high quantity of correct matches. Rich information is presented in terms of feature quantity, total feature matches, correct matches, and repeatability scores. Moreover, computational costs of 25 diverse feature detectors are reported towards the end, which can be used as a benchmark for comparison studies.
{"title":"Potential of SIFT, SURF, KAZE, AKAZE, ORB, BRISK, AGAST, and 7 More Algorithms for Matching Extremely Variant Image Pairs","authors":"Shaharyar Ahmed Khan Tareen, R. H. Raza","doi":"10.1109/iCoMET57998.2023.10099250","DOIUrl":"https://doi.org/10.1109/iCoMET57998.2023.10099250","url":null,"abstract":"Extremely variant image pairs include distorted, deteriorated, and corrupted scenes that have experienced severe geometric, photometric, or non-geometric-non-photometric transformations with respect to their originals. Real world visual data can become extremely dusty, smoky, dark, noisy, motion-blurred, affine, JPEG compressed, occluded, shadowed, virtually invisible, etc. Therefore, matching of extremely variant scenes is an important problem and computer vision solutions must have the capability to yield robust results no matter how complex the visual input is. Similarly, there is a need to evaluate feature detectors for such complex conditions. With standard settings, feature detection, description, and matching algorithms typically fail to produce significant number of correct matches in these types of images. Though, if full potential of the algorithms is applied by using extremely low thresholds, very encouraging results are obtained. In this paper, potential of 14 feature detectors: SIFT, SURF, KAZE, AKAZE, ORB, BRISK, AGAST, FAST, MSER, MSD, GFTT, Harris Corner Detector based GFTT, Harris Laplace Detector, and CenSurE has been evaluated for matching 10 extremely variant image pairs. MSD detected more than 1 million keypoints in one of the images and SIFT exhibited a repeatability score of 99.76% for the extremely noisy image pair but failed to yield high quantity of correct matches. Rich information is presented in terms of feature quantity, total feature matches, correct matches, and repeatability scores. Moreover, computational costs of 25 diverse feature detectors are reported towards the end, which can be used as a benchmark for comparison studies.","PeriodicalId":369792,"journal":{"name":"2023 4th International Conference on Computing, Mathematics and Engineering Technologies (iCoMET)","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128148880","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 : 2023-03-17DOI: 10.1109/iCoMET57998.2023.10099177
Muhammad Umair, Syed Aun Irtaza, Shahid Salim
Now a days world is look right on digitalized. Social media is captivating in this digital age through the accessibility of consumer's feedback. The recent work in the field of classification based on comments on social media is gaining appeal on a global scale. Unfortunately, the study does not offer better accuracy in terms of toxic comments. On social media platforms, hateful and abusive language has a detrimental effect on users' mental health and involvement from people from all diverse backgrounds. Automatic methods is most commonly used datasets with categorical labels to detect foul language. The level of offensiveness of comments varies. In NLP we use binary classification like either a comment is offensive or not and leave continues classification. In continues classification one can identify the severity level of comments, can set a threshold, and by using Deep Learning and modeling techniques can directly identify the severity level of comments by considering context. The review of related literature shows that identification of toxicity of user comments can be improved by pre-processing methods, such as deleting null values and anomies from the dataset, to refine the dataset and increase its accuracy by applying different algorithm techniques to make feature more valuables. This research provides analysis of user comments datasets and study's user comments toxicity with different machine learning approaches. First, we need to do pre-processing steps including punctuations, stop words, null entries, and duplicate removal to remove anomalies. After that we need to apply different methods like count vectorizer and bag of words to extract features. After that, we MCPL algorithm applied on these datasets to predicts results. By applying MCPL model on user comments dataset 88.5% accuracy were founded.
{"title":"User Feedback Severity Level Identification and Classification through Deeper Analysis of Text","authors":"Muhammad Umair, Syed Aun Irtaza, Shahid Salim","doi":"10.1109/iCoMET57998.2023.10099177","DOIUrl":"https://doi.org/10.1109/iCoMET57998.2023.10099177","url":null,"abstract":"Now a days world is look right on digitalized. Social media is captivating in this digital age through the accessibility of consumer's feedback. The recent work in the field of classification based on comments on social media is gaining appeal on a global scale. Unfortunately, the study does not offer better accuracy in terms of toxic comments. On social media platforms, hateful and abusive language has a detrimental effect on users' mental health and involvement from people from all diverse backgrounds. Automatic methods is most commonly used datasets with categorical labels to detect foul language. The level of offensiveness of comments varies. In NLP we use binary classification like either a comment is offensive or not and leave continues classification. In continues classification one can identify the severity level of comments, can set a threshold, and by using Deep Learning and modeling techniques can directly identify the severity level of comments by considering context. The review of related literature shows that identification of toxicity of user comments can be improved by pre-processing methods, such as deleting null values and anomies from the dataset, to refine the dataset and increase its accuracy by applying different algorithm techniques to make feature more valuables. This research provides analysis of user comments datasets and study's user comments toxicity with different machine learning approaches. First, we need to do pre-processing steps including punctuations, stop words, null entries, and duplicate removal to remove anomalies. After that we need to apply different methods like count vectorizer and bag of words to extract features. After that, we MCPL algorithm applied on these datasets to predicts results. By applying MCPL model on user comments dataset 88.5% accuracy were founded.","PeriodicalId":369792,"journal":{"name":"2023 4th International Conference on Computing, Mathematics and Engineering Technologies (iCoMET)","volume":"88 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134541016","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：