Maheshwari Kotha, Mounika Chadalavada, S. Karuturi, H. Venkataraman
{"title":"PotSense","authors":"Maheshwari Kotha, Mounika Chadalavada, S. Karuturi, H. Venkataraman","doi":"10.1145/3377283.3377286","DOIUrl":"https://doi.org/10.1145/3377283.3377286","url":null,"abstract":"An Intelligent Transport System is an essential facet of today's world. The Indian traffic scenario has many distinct challenges that are not common in other countries. These embody less-disciplined/chaotic lane traffic, varied vehicle types plying at the same time and poor road conditions such as potholes. As per MoRTH (Ministry of Road Transport and Highways) India, around a million accidents and nearly 10,000 road accident-based deaths happen every year solely owing to potholes. Given the humongous road network in India, combined with limitations of Government Departments, typically only a few roads are well-maintained. In this regard, this work proposes an economical yet efficient system called PotSense. PotSense uses smartphone-based sensors, particularly accelerometer and camera sensors, to crowd-source information of potholes on public roads. Furthermore, it is broadcasted to all road users to ensure their safety. PotSense also investigates the utilization of different neural network techniques for processing the data obtained from the camera. Moreover, the collected data is analyzed to perceive substantial insight on road quality and potholes. Notably, it is observed that irrespective of the dimension, depth, and other characteristics of the pothole, the proposed solution, \"PotSense\" accurately detects the pothole in more than 60% of the typical Indian road scenarios.","PeriodicalId":443854,"journal":{"name":"Proceedings of the 1st ACM Workshop on Autonomous and Intelligent Mobile Systems","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-01-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126916834","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}
S. Badri, Venkata Sunil Reddy Timma Reddy, Vijay Bhaskar Reddy Chintakunta, S. Kamini
{"title":"Note","authors":"S. Badri, Venkata Sunil Reddy Timma Reddy, Vijay Bhaskar Reddy Chintakunta, S. Kamini","doi":"10.1145/3377283.3377287","DOIUrl":"https://doi.org/10.1145/3377283.3377287","url":null,"abstract":"Unmanned Aerial Vehicles (UAV) or Drones have been developing in an exponential rise within a few decades, and they represent outstanding progress in several applications. Applications like human reach are either severe or venturous, particularly within the tedious and risky activities with advanced technology. However, we have stepped towards human health emergency cases like sudden strokes, natural disasters, and they need necessary treatment. A drone network with a high-speed that delivers emergency supplies to the location range with an adequate time constant. The primary objective of this work is to develop a drone system that will fly autonomously, travelling to multiple locations through the utilization of a Global positioning system (GPS) module with a minimum time constant. The second objective of this work is the installation of medical accessories for health care. Therefore, we can use a drone for medical assistance in emergency cases.","PeriodicalId":443854,"journal":{"name":"Proceedings of the 1st ACM Workshop on Autonomous and Intelligent Mobile Systems","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-01-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121829931","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}
Kartik Mundra, Rahul Modpur, Arpan Chattopadhyay, I. Kar
{"title":"Adversarial Image Detection in Cyber-Physical Systems","authors":"Kartik Mundra, Rahul Modpur, Arpan Chattopadhyay, I. Kar","doi":"10.1145/3377283.3377285","DOIUrl":"https://doi.org/10.1145/3377283.3377285","url":null,"abstract":"In this paper, detection of deception attack on deep neural network (DNN) based image classification in autonomous and cyber-physical systems is considered. Several studies have shown the vulnerability of DNN to malicious deception attack. In such attacks, some or all pixel values of an image are modified by an external attacker, so that the change is almost invisible to human eye but significant enough for a DNN-based classifier to misclassify it. This paper proposes a novel pre-processing technique that facilitates detection of such modified images under any DNN-based image classifier as well as attacker model. The proposed pre-processing algorithm involves a certain combination of principal component analysis (PCA)-based decomposition of the image, and random perturbation based detection to reduce computational complexity. Numerical experiments show that the proposed detection scheme outperforms a competing attack detection algorithm while achieving low false alarm rate and low computational complexity.","PeriodicalId":443854,"journal":{"name":"Proceedings of the 1st ACM Workshop on Autonomous and Intelligent Mobile Systems","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-01-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127524512","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}
Shivani Kapania, D. Saini, S. Goyal, Narina Thakur, Rachna Jain, P. Nagrath
{"title":"Multi Object Tracking with UAVs using Deep SORT and YOLOv3 RetinaNet Detection Framework","authors":"Shivani Kapania, D. Saini, S. Goyal, Narina Thakur, Rachna Jain, P. Nagrath","doi":"10.1145/3377283.3377284","DOIUrl":"https://doi.org/10.1145/3377283.3377284","url":null,"abstract":"Over the years, object tracking and detection has emerged as one of the most important aspects of UAV applications such as surveillance, reconnaissance, etc. In our paper, we present a tracking-by-detection approach for real-time Multiple Object Tracking (MOT) of footage from a drone-mounted camera. Tracking-by-detection is the leading paradigm considering its computational effectiveness and improved detection algorithms. Our algorithm builds on the baseline Deep SORT algorithm implemented for MOT benchmarks. However, to circumvent the challenges posed by videos captured from a significant height we use a combination of YOLOv3 and RetinaNet for generating detections in each frame. The results of our experiment on the VisDrone 2018 dataset exhibit competitive performance in comparison to the existing trackers.","PeriodicalId":443854,"journal":{"name":"Proceedings of the 1st ACM Workshop on Autonomous and Intelligent Mobile Systems","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-01-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129183439","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}
{"title":"A Neighborhood Overlap Based Approach for Service Provider Prioritization in a Directed Social IoT Service Network","authors":"Nishit Narang, Subrat Kar","doi":"10.1145/3377283.3377288","DOIUrl":"https://doi.org/10.1145/3377283.3377288","url":null,"abstract":"Social IoT (or SIoT) is an alternate architectural pattern for IoT, which involves imparting social behavioral attributes to IoT devices. An SIoT service network uses social collaboration between IoT devices (acting as service users or service providers or both), enabling low-latency collaborative services and applications. A key challenge in implementing an SIoT service network in a multi-vendor network of heterogeneous IoT devices is the issue of Trust. The problem is in prioritization and selection of trustworthy service provider(s) in an autonomous and independent manner. In a single-vendor network, the problem is handled via proprietary methods that do not scale for multi-vendor environments. The problem is further compounded in networks having IoT devices that are constrained in computational and storage resources. In this paper, we propose the use of Neighborhood Overlap for estimating tie-strengths and the consequent prioritization of service providers based on the estimated tie-strength. We verify the relationship between neighborhood overlap and tie-strength using three publicly available datasets. While past research on neighborhood-overlap and its relationship with tie-strength focuses on undirected social networks only, we extend the definition of neighborhood-overlap for directed networks. We further prove this extension with the help of two publicly available directed social network datasets. The idea proposed in this paper is fundamental and can be extended towards defining a trust framework for SIoT.","PeriodicalId":443854,"journal":{"name":"Proceedings of the 1st ACM Workshop on Autonomous and Intelligent Mobile Systems","volume":"106 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-01-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127658627","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}
{"title":"Proceedings of the 1st ACM Workshop on Autonomous and Intelligent Mobile Systems","authors":"","doi":"10.1145/3377283","DOIUrl":"https://doi.org/10.1145/3377283","url":null,"abstract":"","PeriodicalId":443854,"journal":{"name":"Proceedings of the 1st ACM Workshop on Autonomous and Intelligent Mobile Systems","volume":"847 2","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133842680","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}