{"title":"PCA Based Hierarchical CNN for the Classification of Mild Cognitive Impairments and the Role of SIREN Activations","authors":"Harsh Bhasin, R. Agarwal, For Alzheimer's Disease","doi":"10.1109/ACCC54619.2021.00031","DOIUrl":"https://doi.org/10.1109/ACCC54619.2021.00031","url":null,"abstract":"The use of Convolutional Neural Networks for the classification of volumetric data is contentious because 2-D convolutions miss out on the correlation between the slices of the volume, whilst 3-D networks guzzle extensive computing resources. Moreover, the advent of SIREN activations calls for the investigation regarding the role of activations in such networks. This work proposes a model that uses the Principal Component Analysis to reduce the given data, followed by a circumspectly designed CNN for extracting the pertinent features. The paper also investigates the role of activations in such networks. The method is used to classify the patients converted to Alzheimer's from Mild Cognitive Impairment from those who did not convert. The data is obtained from ADNI. The proposed work gives an accuracy of 94.29, which is better as compared to the state-of-the-art.","PeriodicalId":215546,"journal":{"name":"2021 2nd Asia Conference on Computers and Communications (ACCC)","volume":"86 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126073122","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":"Super-Resolution Estimation of Signal Direction Based on Unsupervised Learning","authors":"Jiawen He, Peishun Liu, Liang Wang, Ruichun Tang","doi":"10.1109/ACCC54619.2021.00019","DOIUrl":"https://doi.org/10.1109/ACCC54619.2021.00019","url":null,"abstract":"Target direction estimation is one of the main research directions in the field of array signal processing. In this paper, unsupervised learning method is adopted to study the multi-target direction estimation ability of Deep Neural Network (DNN), under low SNR without using a large amount of training data. The method in this paper is designed to estimate target direction by nonlinear least square spectrum estimation. It is found that when the SNR is -10dB, the precision rate of the DNN can still reach about 92%. Compared with the Conventional Beam Forming (CBF) method, the DNN has a significantly narrow main lobe, and the parameters obtained have the characteristics of sparse. In addition, when we explore whether adjacent targets have an impact on the results, we find that the method in this paper also has the ability of super-resolution. The above findings provide new ideas and experience for the further study of the relationship between array signals and deep learning. As well as for the design and improvement of relevant algorithms on this basis.","PeriodicalId":215546,"journal":{"name":"2021 2nd Asia Conference on Computers and Communications (ACCC)","volume":"221 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122865883","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":"Arithmetic Circuit Homomorphic Encryption Key Pairing Comparisons and Analysis between Elliptic Curve Diffie Hellman and Supersingular Isogeny Diffie Hellman","authors":"Wen Xin Khoo Joshua, Xin Wei Teoh Justin, C. Yap","doi":"10.1109/ACCC54619.2021.00030","DOIUrl":"https://doi.org/10.1109/ACCC54619.2021.00030","url":null,"abstract":"This project is an extension of ongoing research on Fully Homomorphic Encryption - Arithmetic Circuit Homomorphic Encryption. This paper focus on the implementation of pairing algorithm Supersingular Isogeny Diffie Hellman Key Exchange into Arithmetic Circuit Homomorphic Encryption as well as comparison and analyse with Elliptic Curve Diffie Hellman. Next, the paper will discuss on the latencies incurred due to pairing sessions between machines, key generations, key sizes, CPU usage and overall latency for the two respective key exchange methods to be compared against each other.","PeriodicalId":215546,"journal":{"name":"2021 2nd Asia Conference on Computers and Communications (ACCC)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130408295","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}
Xiaofeng Lu, Xuan Wang, Zhengyan Wang, Xinhong Hei
{"title":"Template Update Based on 3D-Convolutional Siamese Network for Object Tracking","authors":"Xiaofeng Lu, Xuan Wang, Zhengyan Wang, Xinhong Hei","doi":"10.1109/ACCC54619.2021.00016","DOIUrl":"https://doi.org/10.1109/ACCC54619.2021.00016","url":null,"abstract":"Object tracking is an important research area in the field of computer vision. In the past two years, the object tracking algorithms based on Siamese network have yielded brilliant results in CVPR. However, previous algorithms only extract the object feature of the first frame as a tracking template. In the process of tracking the object, the object template remains unchanged, leading to poor tracking accuracy. In view of this, the present paper proposes a new, end-to-end trained, fully convolutional 3D Siamese network-based tracking algorithm to extract multiple features. Logistic loss function and SGD are used to train the network. The trained network realizes the use of multi-frame features to update the object template in the process of tracking the video's object. The tracker in this paper can run at real-time frame-rates in OTB, VOT, and GOT-10k. The algorithm is applied to SiamFC, its accuracy is improved by 4% and 3% on the OTB and VOT-2016 data sets, respectively.","PeriodicalId":215546,"journal":{"name":"2021 2nd Asia Conference on Computers and Communications (ACCC)","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132218685","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":"Adaptive Digital Encounters: An approach for reducing digital impact on outpatient flow","authors":"Fahad Ahmed Satti, T. Chung, Sungyoung Lee","doi":"10.1109/ACCC54619.2021.00032","DOIUrl":"https://doi.org/10.1109/ACCC54619.2021.00032","url":null,"abstract":"Healthcare service delivery has been greatly impacted by the current Covid-19 pandemic. One of the key drawbacks of the current Healthcare Management Information Systems (HMIS) is the lack of research towards improving the user's experience before, during, or after interacting with the digital system, product, or service. This has further increased the amount of cognitive load experienced by healthcare providers. Adaptive Digital Encounters (ADE) provide a mechanism for dynamically generating and upgrading the user interfaces of healthcare and wellness applications, by incorporating past histories of the patient data. It also integrates various medical devices to automate the process of collecting vital signs and reduces the burden of inserting data. This paper provides the basic building blocks which were employed to incorporate the ADE into a live application. Our results indicate an above-average score of 1.13 (-3 to +3) using the UEQ-S questionnaire, indicating a positive UX evaluation from 11 participants.","PeriodicalId":215546,"journal":{"name":"2021 2nd Asia Conference on Computers and Communications (ACCC)","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132688633","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":"QR Code-Based Student Attendance System","authors":"Khang Jie Liew, Tee Hean Tan","doi":"10.1109/ACCC54619.2021.00009","DOIUrl":"https://doi.org/10.1109/ACCC54619.2021.00009","url":null,"abstract":"Student attendance in higher education institutions is a factor that impacts academic performance. Some higher education institutions have penalised students for poor attendance records. Most universities, however, have not implemented an automated attendance-taking system to deter poor attendance, leaving the task instead to instructors who manually record students' attendances into the system; a time-consuming and tedious undertaking, particularly, with a large number of students. This study primarily aimed to propose a Quick Response (QR) code-based attendance system, complete with several features to prevent attendance cheating. Students at the Centre for American Education, Sunway University tested a mobile application designed to scan QR codes, specifically generated for each classroom. The system checked three categories of data: subject class hour, registered mobile device, and geolocation. Once the information was verified, the student's attendance was recorded into the system. The proposed system succeeded in overcoming limitations encountered in the university's existing student attendance-taking system.","PeriodicalId":215546,"journal":{"name":"2021 2nd Asia Conference on Computers and Communications (ACCC)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130498959","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":"Design and Implementation of Safety Helmet Detection System Based on YOLOv5","authors":"Yaqi Guan, Wenqiang Li, Tianyu Hu, Qun Hou","doi":"10.1109/ACCC54619.2021.00018","DOIUrl":"https://doi.org/10.1109/ACCC54619.2021.00018","url":null,"abstract":"In order to reduce safety accidents caused by non-standard wearing of helmets, deep learning target detection technology is applied to construction safety detection scenarios, and a helmet detection algorithm based on YOLO v5 is proposed, which can realize real-time detection of helmet wearing. The deep learning part uses the K-means algorithm to cluster the dimensions of the target frame, and Yolov5s.pt is used for deep learning training. During training, the size of the input image is changed to increase the adaptability of the model, and the hyperparameters and optimizer are adjusted to be the best after improvement. The detection model has an accuracy rate of 90%, and the detection speed has reached 37.8fps, which meets the requirements of real-time detection of helmets. Through the combination of this model and hardware such as cameras, a real-time detection of whether a person wears a helmet is designed and implemented. The system realizes the three functions of picture detection, video detection and real-time monitoring.","PeriodicalId":215546,"journal":{"name":"2021 2nd Asia Conference on Computers and Communications (ACCC)","volume":"36 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125374517","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}
Valerio Mandarino, Giovanni Marotta, G. Pappalardo, E. Tramontana
{"title":"Issues Related to EHR Blockchain Applications","authors":"Valerio Mandarino, Giovanni Marotta, G. Pappalardo, E. Tramontana","doi":"10.1109/ACCC54619.2021.00029","DOIUrl":"https://doi.org/10.1109/ACCC54619.2021.00029","url":null,"abstract":"One of the most significant opportunities offered by the blockchain technology is to finally integrate the different processes in the healthcare ecosystem. Several companies are exploring the use of blockchain to manage digital medical data, and several solutions, leveraging the benefits of the blockchain, have been proposed for the implementation of the Electronic Health Record (EHR). However, the application of blockchain technology in the medical domain is not free from challenges. This paper describes the issues related to the adoption of the blockchain technology in the development of an EHR management system.","PeriodicalId":215546,"journal":{"name":"2021 2nd Asia Conference on Computers and Communications (ACCC)","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114055105","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":"Short Video Copyright Protection Based on Blockchain Technology","authors":"Yanfang Qi, Junbiao Liu, Fang Dong, Ping Dong, Yanyun Dai, Lurong Jiang","doi":"10.1109/ACCC54619.2021.00024","DOIUrl":"https://doi.org/10.1109/ACCC54619.2021.00024","url":null,"abstract":"The advent of the 5G era has promoted the rapid development of the short video industry. As the short video industry has become more prosperous, short video copyright disputes have occurred frequently, seriously jeopardizing the healthy development of the short video industry. How to protect the copyright of short videos has become an urgent problem to be solved. In order to solve the problems of long period of short video copyright identification, difficulty in rights protection, difficulty in infringement supervision, and complicated copyright transfer process, this paper combines blockchain technology and designs a short video copyright protection system based on blockchain. The system uses a multi-channel architecture, combined with the non-tamperable and traceable characteristics of the blockchain, and has designed a variety of functions to provide solutions to the above problems. In this paper, the system is actually built. Related functional tests and analysis of the system, have verified the feasibility of this scheme.","PeriodicalId":215546,"journal":{"name":"2021 2nd Asia Conference on Computers and Communications (ACCC)","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123048721","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":"Deep Adversarial Image Synthesis for Nuclei Segmentation of Histopathology Image","authors":"Jijun Cheng, Zimin Wang, Zhenbing Liu, Zhengyun Feng, Huadeng Wang, Xipeng Pan","doi":"10.1109/ACCC54619.2021.00017","DOIUrl":"https://doi.org/10.1109/ACCC54619.2021.00017","url":null,"abstract":"Nuclei segmentation is a fundamental upstream task of digital pathology image analysis. Existing nuclei segmentation methods usually require pixel-level labeled images from experienced pathologists. In this paper, we proposed an innovative data augmentation workflow for histopathology images: a) generates a set of initial central points randomly with existing human-annotated histopathology image datasets; b) generates nuclei segmentation masks based on the generated centroid points of step a); c) generates Haematoxylin and Eosin (H&E)-stained histopathology images corresponding to the generated nuclei masks. In addition, we proposed a deep attention feature fusion generative adversarial network (DAFF -GAN) to improve the image quality and the photorealism of the generated image. We conducted extensive experiments on several existing nuclei segmentation methods, comparing using raw data with the augmented data by our strategy. Extensive experiments proved the effectiveness of our proposed strategy.","PeriodicalId":215546,"journal":{"name":"2021 2nd Asia Conference on Computers and Communications (ACCC)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128318279","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}