Pub Date : 2021-04-06DOI: 10.1109/CAIDA51941.2021.9425075
Fares Yousefi, H. Kolivand
Nowadays, to access any digital device we use authentication techniques, which is a critical technology in terms of security. Present biometric authentications such as fingerprints or face recognition are the most used methods in our digitalized world, which are impressively advantageous in terms of security. However, there are still some flaws in using these methods like not being useful for physical disabilities, environment usage matters, and most importantly the possibility of replicating them with some new technologies because of their visibility. Brain signal is another human biometric that could cover the issues of other types in terms of security and visibility. There are different perspectives about the EEG authentication challenges, including ease of use, privacy, and confirmation necessities like comprehensiveness, uniqueness, collectability, and most importantly permanency which is a big challenge for EEG-based authentications specifically. In this paper, we proposed a method using the deep breath strategy to use brain signals for authentication purposes regardless of brain situation. The result shows that our proposal accomplishment can alter the entire cycle of brain-based authentication when compared with other techniques and EEG-based authentication methods according to the parameter of permanency of the technique in many different brain states.
{"title":"A New Solution to the Brain State Permanency for Brain-Based Authentication Methods","authors":"Fares Yousefi, H. Kolivand","doi":"10.1109/CAIDA51941.2021.9425075","DOIUrl":"https://doi.org/10.1109/CAIDA51941.2021.9425075","url":null,"abstract":"Nowadays, to access any digital device we use authentication techniques, which is a critical technology in terms of security. Present biometric authentications such as fingerprints or face recognition are the most used methods in our digitalized world, which are impressively advantageous in terms of security. However, there are still some flaws in using these methods like not being useful for physical disabilities, environment usage matters, and most importantly the possibility of replicating them with some new technologies because of their visibility. Brain signal is another human biometric that could cover the issues of other types in terms of security and visibility. There are different perspectives about the EEG authentication challenges, including ease of use, privacy, and confirmation necessities like comprehensiveness, uniqueness, collectability, and most importantly permanency which is a big challenge for EEG-based authentications specifically. In this paper, we proposed a method using the deep breath strategy to use brain signals for authentication purposes regardless of brain situation. The result shows that our proposal accomplishment can alter the entire cycle of brain-based authentication when compared with other techniques and EEG-based authentication methods according to the parameter of permanency of the technique in many different brain states.","PeriodicalId":272573,"journal":{"name":"2021 1st International Conference on Artificial Intelligence and Data Analytics (CAIDA)","volume":"32 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-04-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130050748","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 : 2021-04-06DOI: 10.1109/CAIDA51941.2021.9425144
Elrasheed Ismail Mohommoud ZAYID, Nadir Abdelrahman Ahmed FARAH, Turki Mohammed Abdullah AL-SHEHRI, Ali Mohammed Saeed ALRAYAYAEI, Somaia Mohammed Ali ELIMAM
In terms of fatalities caused by Traffic Accidents (TAs), Saudi Arabia occupies a top position in the world. Notably, the most affected age group is young people (i.e. age 18 -25 years). In this study, we addressed the major causes of TAs in the Aseer Region, which is the southern Saudi area. To accurately perform the correct simulation, we used two different datasets. The first one was called ataset1 (DS1), which was extracted from 116 volunteer young participants in the region, each of whom reported a full TA they had survived or witnessed. The second dataset (DS2) was generated using a powerful simulation and modeling algorithm. DS1 was used as input for performing the simulation. For accurate statistical computing TA simulation and modeling purposes, the MATLAB 2018 computing environment was the best candidate to validate our postulates. More than 14 different TA factors were examined to prove the extent of devastation caused by TAs. The contributions of several TA metrics were calculated, and smartphone use while driving and speeding were found to be the primary factors that contributed to TAs. The number of death(s) registered from the public hospital(s) in the region was greater than 22%, which is extraordinary, making this the highest traffic risk in the area. Comparing this study findings with those of previous related ones, the study offers promising results that could be utilized to secure the local community from the current and frequent TA threats. One of the main recommendations in this study, is to incorporate the driving behavior of young people in the primary educational systems.
{"title":"Simulation-based Traffic Accident Testing in the Aseer Region of Saudi Arabia","authors":"Elrasheed Ismail Mohommoud ZAYID, Nadir Abdelrahman Ahmed FARAH, Turki Mohammed Abdullah AL-SHEHRI, Ali Mohammed Saeed ALRAYAYAEI, Somaia Mohammed Ali ELIMAM","doi":"10.1109/CAIDA51941.2021.9425144","DOIUrl":"https://doi.org/10.1109/CAIDA51941.2021.9425144","url":null,"abstract":"In terms of fatalities caused by Traffic Accidents (TAs), Saudi Arabia occupies a top position in the world. Notably, the most affected age group is young people (i.e. age 18 -25 years). In this study, we addressed the major causes of TAs in the Aseer Region, which is the southern Saudi area. To accurately perform the correct simulation, we used two different datasets. The first one was called ataset1 (DS1), which was extracted from 116 volunteer young participants in the region, each of whom reported a full TA they had survived or witnessed. The second dataset (DS2) was generated using a powerful simulation and modeling algorithm. DS1 was used as input for performing the simulation. For accurate statistical computing TA simulation and modeling purposes, the MATLAB 2018 computing environment was the best candidate to validate our postulates. More than 14 different TA factors were examined to prove the extent of devastation caused by TAs. The contributions of several TA metrics were calculated, and smartphone use while driving and speeding were found to be the primary factors that contributed to TAs. The number of death(s) registered from the public hospital(s) in the region was greater than 22%, which is extraordinary, making this the highest traffic risk in the area. Comparing this study findings with those of previous related ones, the study offers promising results that could be utilized to secure the local community from the current and frequent TA threats. One of the main recommendations in this study, is to incorporate the driving behavior of young people in the primary educational systems.","PeriodicalId":272573,"journal":{"name":"2021 1st International Conference on Artificial Intelligence and Data Analytics (CAIDA)","volume":"47 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-04-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116624706","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 : 2021-04-06DOI: 10.1109/CAIDA51941.2021.9425215
M. Shafiullah, Md Juel Rana, Md. Ershadul Haque, Asif Islam, Syed Masiur Rahman, M. Shafiul Alam, Amjad Ali
This paper proposes an intelligent approach to detect and classify the power quality (PQ) events with the combination of machine learning and advanced signal processing techniques. It selects Stockwell transform, one of the efficient signal processing tools for feature extraction from the recorded signals. The extracted features are then fetched to one of the popular machine-learning tools, namely the artificial neural network (ANN), to develop the proposed intelligent PQ events detection and classification approach. This paper selects the hyper-parameters, e.g., number of hidden layer neurons, training algorithm, and activation functions through a systematic trial and error approach. To enhance the proposed approach performance, the weights and biases of the ANN are optimized using the grey wolf optimization (GWO) technique. Simulation results confirm the efficacy of the developed intelligent methodology in distinguishing PQ events from non-PQ events. Moreover, separates different PQ events, e.g., sag, swell, interruption, fluctuation, spike, notch, harmonics, from each other with reasonable accuracy. This research also investigates the efficacy of the proposed signal processing-based machine learning approach in the presence of measurement noises.
{"title":"An Intelligent Approach for Power Quality Events Detection and Classification","authors":"M. Shafiullah, Md Juel Rana, Md. Ershadul Haque, Asif Islam, Syed Masiur Rahman, M. Shafiul Alam, Amjad Ali","doi":"10.1109/CAIDA51941.2021.9425215","DOIUrl":"https://doi.org/10.1109/CAIDA51941.2021.9425215","url":null,"abstract":"This paper proposes an intelligent approach to detect and classify the power quality (PQ) events with the combination of machine learning and advanced signal processing techniques. It selects Stockwell transform, one of the efficient signal processing tools for feature extraction from the recorded signals. The extracted features are then fetched to one of the popular machine-learning tools, namely the artificial neural network (ANN), to develop the proposed intelligent PQ events detection and classification approach. This paper selects the hyper-parameters, e.g., number of hidden layer neurons, training algorithm, and activation functions through a systematic trial and error approach. To enhance the proposed approach performance, the weights and biases of the ANN are optimized using the grey wolf optimization (GWO) technique. Simulation results confirm the efficacy of the developed intelligent methodology in distinguishing PQ events from non-PQ events. Moreover, separates different PQ events, e.g., sag, swell, interruption, fluctuation, spike, notch, harmonics, from each other with reasonable accuracy. This research also investigates the efficacy of the proposed signal processing-based machine learning approach in the presence of measurement noises.","PeriodicalId":272573,"journal":{"name":"2021 1st International Conference on Artificial Intelligence and Data Analytics (CAIDA)","volume":"45 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-04-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130275226","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 : 2021-04-06DOI: 10.1109/CAIDA51941.2021.9425208
D. Bailey, N. Almusharraf
Pedagogical chatbots are used to elicit information from students. Yet, how the amount of student-generated content differs according to the type of chatbot-delivered questions is unknown. To this end, 19 South Korean English majors completed six chatbot assignments through an in-house developed Facebook Messenger chatbot. The chatbot activity entailed creating original stories for class presentations. In addition to directives requesting plot details, the chatbot used closed-ended button reply questions, open-ended questions, and fill-in-the-blank template statements to help students create stories. Results indicated that button reply questions allowed for pacing, recall and content assessment and required low levels of critical thinking. Next, open-ended questions and fill-in-the-blank template statements resulted in similar word count production but different levels of creativity and critical thinking. Lastly, directives requesting user input resulted in 35% more output, indicating students took more action when told to do something than when asked. Regarding the novelty effect, fewer students volunteered to do the sixth and final chatbot activity, but those who did produced word count on par with their initial chatbot activity.
{"title":"Investigating the Effect of Chatbot-to-User Questions and Directives on Student Participation","authors":"D. Bailey, N. Almusharraf","doi":"10.1109/CAIDA51941.2021.9425208","DOIUrl":"https://doi.org/10.1109/CAIDA51941.2021.9425208","url":null,"abstract":"Pedagogical chatbots are used to elicit information from students. Yet, how the amount of student-generated content differs according to the type of chatbot-delivered questions is unknown. To this end, 19 South Korean English majors completed six chatbot assignments through an in-house developed Facebook Messenger chatbot. The chatbot activity entailed creating original stories for class presentations. In addition to directives requesting plot details, the chatbot used closed-ended button reply questions, open-ended questions, and fill-in-the-blank template statements to help students create stories. Results indicated that button reply questions allowed for pacing, recall and content assessment and required low levels of critical thinking. Next, open-ended questions and fill-in-the-blank template statements resulted in similar word count production but different levels of creativity and critical thinking. Lastly, directives requesting user input resulted in 35% more output, indicating students took more action when told to do something than when asked. Regarding the novelty effect, fewer students volunteered to do the sixth and final chatbot activity, but those who did produced word count on par with their initial chatbot activity.","PeriodicalId":272573,"journal":{"name":"2021 1st International Conference on Artificial Intelligence and Data Analytics (CAIDA)","volume":"84 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-04-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124923415","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 : 2021-04-06DOI: 10.1109/CAIDA51941.2021.9425183
Safia Dawood, A. Dawood, Hind Alaskar, T. Saba
COVID-19 imposed huge burdens and obligations on public health and epidemiology centers to elevate the role of periodic surveillance and case tracing in order to cease the spread of the pandemic. As a result, nations globally are developing various digital solutions for accurate surveillance, reporting of new cases, tracing contacts, and monitoring public health. Traditional tracking and reporting methods have been replaced by intelligent solutions for accurate and efficient reporting. Tools such smart phones, portable devices, and drones have been incorporated in these solutions. These devices produce large amount of data on daily basis and need to be processed instantly to battle the spread of the virus and this is where AI is needed. While the need for AI in disease control and surveillance is clear, the application of AI methods and machine learning algorithms in this field needs further studies. This paper is a systematic review of using AI in COVID-19 surveillance literature to answer the following questions: 1. What AI-based methods are used globally for COVID-19 surveillance? 2. How effective are these methods, and 3. What are the methods used in the Kingdom of Saudi Arabia.
{"title":"COVID-19 Artificial Intelligence Based Surveillance Applications in The Kingdom of Saudi Arabia","authors":"Safia Dawood, A. Dawood, Hind Alaskar, T. Saba","doi":"10.1109/CAIDA51941.2021.9425183","DOIUrl":"https://doi.org/10.1109/CAIDA51941.2021.9425183","url":null,"abstract":"COVID-19 imposed huge burdens and obligations on public health and epidemiology centers to elevate the role of periodic surveillance and case tracing in order to cease the spread of the pandemic. As a result, nations globally are developing various digital solutions for accurate surveillance, reporting of new cases, tracing contacts, and monitoring public health. Traditional tracking and reporting methods have been replaced by intelligent solutions for accurate and efficient reporting. Tools such smart phones, portable devices, and drones have been incorporated in these solutions. These devices produce large amount of data on daily basis and need to be processed instantly to battle the spread of the virus and this is where AI is needed. While the need for AI in disease control and surveillance is clear, the application of AI methods and machine learning algorithms in this field needs further studies. This paper is a systematic review of using AI in COVID-19 surveillance literature to answer the following questions: 1. What AI-based methods are used globally for COVID-19 surveillance? 2. How effective are these methods, and 3. What are the methods used in the Kingdom of Saudi Arabia.","PeriodicalId":272573,"journal":{"name":"2021 1st International Conference on Artificial Intelligence and Data Analytics (CAIDA)","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-04-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116770622","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 : 2021-04-06DOI: 10.1109/CAIDA51941.2021.9425170
Sreejata Dutta, Soham Ghosh
Early detection and classification of wildfires using aerial image-based computer vision algorithms like convolution neural networks and image processing techniques have lately gained much attention due to the record-setting wildfire events worldwide. Past studies have demonstrated varying degrees of success in implementing forest fire classification algorithms using variants of well-known sophisticated convolutional neural network architectures, which require extensive computation time for training but demonstrate comparatively high false alarm rates and low predictive power. To accurately detect small-scale forest burns, which typically marks the onset of larger catastrophic events, a combined architecture of separable convolution neural network and digital image processing using thresholding and segmentation is proposed in this paper. The proposed architecture is simple and hence computationally less expensive. Performance evaluation on the test data yielded excellent results in terms of high sensitivity, of about 98.10%, and a low specificity of 87.09%.
{"title":"Forest Fire Detection Using Combined Architecture of Separable Convolution and Image Processing","authors":"Sreejata Dutta, Soham Ghosh","doi":"10.1109/CAIDA51941.2021.9425170","DOIUrl":"https://doi.org/10.1109/CAIDA51941.2021.9425170","url":null,"abstract":"Early detection and classification of wildfires using aerial image-based computer vision algorithms like convolution neural networks and image processing techniques have lately gained much attention due to the record-setting wildfire events worldwide. Past studies have demonstrated varying degrees of success in implementing forest fire classification algorithms using variants of well-known sophisticated convolutional neural network architectures, which require extensive computation time for training but demonstrate comparatively high false alarm rates and low predictive power. To accurately detect small-scale forest burns, which typically marks the onset of larger catastrophic events, a combined architecture of separable convolution neural network and digital image processing using thresholding and segmentation is proposed in this paper. The proposed architecture is simple and hence computationally less expensive. Performance evaluation on the test data yielded excellent results in terms of high sensitivity, of about 98.10%, and a low specificity of 87.09%.","PeriodicalId":272573,"journal":{"name":"2021 1st International Conference on Artificial Intelligence and Data Analytics (CAIDA)","volume":"95 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-04-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120997166","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 : 2021-04-06DOI: 10.1109/CAIDA51941.2021.9425202
Inzamam Mashood Nasir, M. Raza, J. H. Shah, Muhammad Attique Khan, A. Rehman
Video based Human Action Recognition (HAR) is an active research field of Machine Learning (ML) and human detection in videos is the most important step in action recognition. Recently, several techniques and algorithms have been proposed to increase the accuracy of HAR process, but margin of improvement still exists. Detection and classification of human actions is a challenging task due to random changes in human appearance, clothes, illumination, and background. In this article, an efficient technique to classify human actions by utilizing steps like removing redundant frames from videos, extracting Segments of Interest (SoIs), feature descriptor mining through Geodesic Distance (GD), 3D Cartesian-plane Features (3D-CF), Joints MOCAP (JMOCAP) and n-way Point Trajectory Generation (nPTG). A Neuro Fuzzy Classifier (NFC) is used at the end for the classification purpose. The proposed technique is tested on two publicly available datasets including HMDB-51 and Hollywood2, and achieved an accuracy of 82.55% and 91.99% respectively. These efficient results prove the validity of proposed model.
{"title":"Human Action Recognition using Machine Learning in Uncontrolled Environment","authors":"Inzamam Mashood Nasir, M. Raza, J. H. Shah, Muhammad Attique Khan, A. Rehman","doi":"10.1109/CAIDA51941.2021.9425202","DOIUrl":"https://doi.org/10.1109/CAIDA51941.2021.9425202","url":null,"abstract":"Video based Human Action Recognition (HAR) is an active research field of Machine Learning (ML) and human detection in videos is the most important step in action recognition. Recently, several techniques and algorithms have been proposed to increase the accuracy of HAR process, but margin of improvement still exists. Detection and classification of human actions is a challenging task due to random changes in human appearance, clothes, illumination, and background. In this article, an efficient technique to classify human actions by utilizing steps like removing redundant frames from videos, extracting Segments of Interest (SoIs), feature descriptor mining through Geodesic Distance (GD), 3D Cartesian-plane Features (3D-CF), Joints MOCAP (JMOCAP) and n-way Point Trajectory Generation (nPTG). A Neuro Fuzzy Classifier (NFC) is used at the end for the classification purpose. The proposed technique is tested on two publicly available datasets including HMDB-51 and Hollywood2, and achieved an accuracy of 82.55% and 91.99% respectively. These efficient results prove the validity of proposed model.","PeriodicalId":272573,"journal":{"name":"2021 1st International Conference on Artificial Intelligence and Data Analytics (CAIDA)","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-04-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133598325","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 : 2021-04-06DOI: 10.1109/CAIDA51941.2021.9425076
Shabana Habib, Altaf Hussain, Muhammad Islam, Sheroz Khan, Waleed Albattah
each year more than two million Muslims from around the world come to perform Hajj in Makah. It is considered the world's largest recorded human gathering during any worshiping event. Safety makes one of the main concerns with regards to managing such large crowds for ensuring that stampedes and other similar overcrowding accidents are avoided. For this purpose, 5000 cameras are installed around the holy sites for monitoring purposes. Due to the continuous nature of surveillance systems in generating video data, it is almost impossible to efficiently and accurately monitor an event of this size in real-time. Analyzing such huge data has required a lot of human resources. Therefore, there is a great need for advanced intelligent techniques to automatically count and manage such large crowds. In order to create an advanced intelligent system that contributes to crowds counting and managing through the surveillance system. In this paper, we propose an accurate computer vision-based approach to crowd management using Convolutional Neural Network (CNN). Our proposed framework is three folds. In the first fold, our own dataset for pilgrim detection is created, covering both sparse and dense crowds. In the second fold, a Faster-RCNN object detection model is trained to detect and count the number of pilgrims. In the third fold, utilizing the resources efficiently the surveillance system has used frame differencing technique to differentiate between motion and static video frames. Only in the case of some sort of motion, we will pass these frames to the pilgrims counting model to tell us about the number of pilgrims in the video. When the number of pilgrims counting is exceeded from the pre-defined threshold the system will automatically trigger the alarm pointing the camera to the location to inform the concerned authorities to take action appropriate measures. Along with that, only the dense crowd will be monitored by law enforcement and for better management. Our experiments show that Faster Region CNN (Faster RCNN) is suitable for accurate detection when compared with other state-of-art crowd management techniques so far reported.
每年有200多万来自世界各地的穆斯林来到麦加进行朝觐。这被认为是世界上有记录的在任何礼拜活动中最大的人类集会。在管理如此庞大的人群以确保避免踩踏和其他类似的过度拥挤事故时，安全是主要问题之一。为此目的，在圣地周围安装了5000个摄影机，以便进行监测。由于监控系统产生视频数据的连续性，几乎不可能高效、准确地实时监控如此大规模的事件。分析如此庞大的数据需要大量的人力资源。因此，非常需要先进的智能技术来自动计数和管理如此庞大的人群。为了创建一个先进的智能系统，有助于通过监控系统进行人群统计和管理。在本文中，我们提出了一种基于卷积神经网络(CNN)的精确计算机视觉的人群管理方法。我们提出的框架有三层。在第一个折叠中，我们创建了自己的朝圣者检测数据集，涵盖了稀疏和密集的人群。在第二部分中，训练了一个Faster-RCNN对象检测模型来检测和计数朝圣者的数量。在第三方面，系统利用帧差技术对动态视频帧和静态视频帧进行区分，有效地利用资源。只有在某种运动的情况下，我们才会将这些帧传递给朝圣者计数模型，以告诉我们视频中朝圣者的数量。当朝圣者人数超过预先设定的阈值时，系统会自动触发警报，将摄像头指向该地点，通知有关当局采取适当措施。与此同时，只有密集的人群才会受到执法部门的监控，并得到更好的管理。我们的实验表明，与目前报道的其他最先进的人群管理技术相比，Faster Region CNN (Faster RCNN)适合于准确的检测。
{"title":"Towards Efficient Detection and Crowd Management for Law Enforcing Agencies","authors":"Shabana Habib, Altaf Hussain, Muhammad Islam, Sheroz Khan, Waleed Albattah","doi":"10.1109/CAIDA51941.2021.9425076","DOIUrl":"https://doi.org/10.1109/CAIDA51941.2021.9425076","url":null,"abstract":"each year more than two million Muslims from around the world come to perform Hajj in Makah. It is considered the world's largest recorded human gathering during any worshiping event. Safety makes one of the main concerns with regards to managing such large crowds for ensuring that stampedes and other similar overcrowding accidents are avoided. For this purpose, 5000 cameras are installed around the holy sites for monitoring purposes. Due to the continuous nature of surveillance systems in generating video data, it is almost impossible to efficiently and accurately monitor an event of this size in real-time. Analyzing such huge data has required a lot of human resources. Therefore, there is a great need for advanced intelligent techniques to automatically count and manage such large crowds. In order to create an advanced intelligent system that contributes to crowds counting and managing through the surveillance system. In this paper, we propose an accurate computer vision-based approach to crowd management using Convolutional Neural Network (CNN). Our proposed framework is three folds. In the first fold, our own dataset for pilgrim detection is created, covering both sparse and dense crowds. In the second fold, a Faster-RCNN object detection model is trained to detect and count the number of pilgrims. In the third fold, utilizing the resources efficiently the surveillance system has used frame differencing technique to differentiate between motion and static video frames. Only in the case of some sort of motion, we will pass these frames to the pilgrims counting model to tell us about the number of pilgrims in the video. When the number of pilgrims counting is exceeded from the pre-defined threshold the system will automatically trigger the alarm pointing the camera to the location to inform the concerned authorities to take action appropriate measures. Along with that, only the dense crowd will be monitored by law enforcement and for better management. Our experiments show that Faster Region CNN (Faster RCNN) is suitable for accurate detection when compared with other state-of-art crowd management techniques so far reported.","PeriodicalId":272573,"journal":{"name":"2021 1st International Conference on Artificial Intelligence and Data Analytics (CAIDA)","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-04-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134083829","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 : 2021-04-06DOI: 10.1109/CAIDA51941.2021.9425120
Abrar A. Almuhanna, Marwa M. Alrehili, Samah H. Alsubhi, Liyakathunisa Syed
Crimes prediction is one of the most important topics in recent years that aim to protect people’s lives. These analytical studies for criminal hotspots are frequently demanded by law enforcement agencies hence, there is a huge requirement and demand for enhanced geographic information systems and innovative spatial data mining techniques in order to enhance crime detections and better protect their communities. In this paper, we propose a methodology to predict Spatio-temporal criminal patterns within the New York City neighbourhoods using a dataset from 2006 until 2019 with 2.2M criminal records for 25 different crimes type. In order to achieve the study objectives, the methodology passes through several stages until the final results are reached, starting with the visualization analysis of Spatio-temporal New York crime data which is important in decision-making, followed by, applying three different classifiers namely; Support Vector Machine (SVM), Random Forest (RF), and XGboost classifiers. After analysis, it is illustrated that XGboost has predicted the highest number of correct classifications out of 25 different crime types it has predicted 22 types of crime accurately, whereas Random Forest has predicted 21 types of crime accurately and SVM predicted accurately 17 types of crimes with lowest accuracy. Hence XGBoost outperformed all other models and can be considered for detection of crimes in the neighborhood.
{"title":"Prediction of Crime in Neighbourhoods of New York City using Spatial Data Analysis","authors":"Abrar A. Almuhanna, Marwa M. Alrehili, Samah H. Alsubhi, Liyakathunisa Syed","doi":"10.1109/CAIDA51941.2021.9425120","DOIUrl":"https://doi.org/10.1109/CAIDA51941.2021.9425120","url":null,"abstract":"Crimes prediction is one of the most important topics in recent years that aim to protect people’s lives. These analytical studies for criminal hotspots are frequently demanded by law enforcement agencies hence, there is a huge requirement and demand for enhanced geographic information systems and innovative spatial data mining techniques in order to enhance crime detections and better protect their communities. In this paper, we propose a methodology to predict Spatio-temporal criminal patterns within the New York City neighbourhoods using a dataset from 2006 until 2019 with 2.2M criminal records for 25 different crimes type. In order to achieve the study objectives, the methodology passes through several stages until the final results are reached, starting with the visualization analysis of Spatio-temporal New York crime data which is important in decision-making, followed by, applying three different classifiers namely; Support Vector Machine (SVM), Random Forest (RF), and XGboost classifiers. After analysis, it is illustrated that XGboost has predicted the highest number of correct classifications out of 25 different crime types it has predicted 22 types of crime accurately, whereas Random Forest has predicted 21 types of crime accurately and SVM predicted accurately 17 types of crimes with lowest accuracy. Hence XGBoost outperformed all other models and can be considered for detection of crimes in the neighborhood.","PeriodicalId":272573,"journal":{"name":"2021 1st International Conference on Artificial Intelligence and Data Analytics (CAIDA)","volume":"44 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-04-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130017544","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 : 2021-04-06DOI: 10.1109/CAIDA51941.2021.9425245
Atif Ali, Y. K. Jadoon, Malik Usman Dilawar, Muhammad Qasim, Shujah Ur Rehman, Muhammad Usama Nazir
The concept of hyper computing (BH) has been introduced to understand how living systems process information. This article presents the BH developments but supports the idea that living beings communicate in structures and not in the mode of signs and symbols. Genetic algorithms are bio-inspired optimization algorithms that simulate the process of the natural evolution of species. They make it possible to manipulate a set of solutions through several iterations to converge towards optimal solutions. This work allows us to study the efficiency of genetic algorithms for statistical machine translation. The bio-inspired communitarian literature proposes a communication model to capture the nature of individuals’ interaction. The research-based on metrics, topology, and algorithms are derived from the bio-inspired communication model’s visual investigation. The evaluation’s assumption is the choice of biologically inspired communication models can influence group performance for a specific task. The communication model was evaluated in two environments. Swarm mission: search for targets communicated among others and avoid opponents. Overall results of the survey Prove that the group agent has the best overall performance when used a bio-inspired communication model to look for specific tasks and avoid compliance with best practices than hostile tasks when using topology models. This study’s main cause is to magnify the group’s mission’s performance by deliberately selecting the bio-inspired communication model.
{"title":"Robotics: Biological Hypercomputation and Bio-Inspired Swarms Intelligence","authors":"Atif Ali, Y. K. Jadoon, Malik Usman Dilawar, Muhammad Qasim, Shujah Ur Rehman, Muhammad Usama Nazir","doi":"10.1109/CAIDA51941.2021.9425245","DOIUrl":"https://doi.org/10.1109/CAIDA51941.2021.9425245","url":null,"abstract":"The concept of hyper computing (BH) has been introduced to understand how living systems process information. This article presents the BH developments but supports the idea that living beings communicate in structures and not in the mode of signs and symbols. Genetic algorithms are bio-inspired optimization algorithms that simulate the process of the natural evolution of species. They make it possible to manipulate a set of solutions through several iterations to converge towards optimal solutions. This work allows us to study the efficiency of genetic algorithms for statistical machine translation. The bio-inspired communitarian literature proposes a communication model to capture the nature of individuals’ interaction. The research-based on metrics, topology, and algorithms are derived from the bio-inspired communication model’s visual investigation. The evaluation’s assumption is the choice of biologically inspired communication models can influence group performance for a specific task. The communication model was evaluated in two environments. Swarm mission: search for targets communicated among others and avoid opponents. Overall results of the survey Prove that the group agent has the best overall performance when used a bio-inspired communication model to look for specific tasks and avoid compliance with best practices than hostile tasks when using topology models. This study’s main cause is to magnify the group’s mission’s performance by deliberately selecting the bio-inspired communication model.","PeriodicalId":272573,"journal":{"name":"2021 1st International Conference on Artificial Intelligence and Data Analytics (CAIDA)","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-04-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115323395","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：