2021 International Symposium on Intelligent Signal Processing and Communication Systems (ISPACS)最新文献_第5页

Modeling and readout of a InN humidity sensor InN湿度传感器的建模和读数

2021 International Symposium on Intelligent Signal Processing and Communication Systems (ISPACS) Pub Date : 2021-11-16 DOI: 10.1109/ISPACS51563.2021.9650928

Ta-Chi Jeang, Shu Mo, Hung-Yu Wang, Chih-Chin Yang, H. Tran

引用次数: 0

[ISPACS 2021 Front cover] [ISPACS 2021封面]

2021 International Symposium on Intelligent Signal Processing and Communication Systems (ISPACS) Pub Date : 2021-11-16 DOI: 10.1109/ispacs51563.2021.9651073

引用次数: 0

Humming-to-Instrument Conversion based on CycleGAN 基于CycleGAN的蜂鸣声-仪表转换

2021 International Symposium on Intelligent Signal Processing and Communication Systems (ISPACS) Pub Date : 2021-11-16 DOI: 10.1109/ISPACS51563.2021.9651076

Wen-Hsing Lai, Siou-Lin Wang, Zhi-Yao Xu

引用次数: 0

Optimum Assignment of Variable Code Length and Modulation Index for VLC-OFDM VLC-OFDM中可变码长和调制指标的优化分配

2021 International Symposium on Intelligent Signal Processing and Communication Systems (ISPACS) Pub Date : 2021-11-16 DOI: 10.1109/ISPACS51563.2021.9651065

Koki Yanashita, Kazushi Shimada, Seiya Hirano, C. Ahn

{"title":"Optimum Assignment of Variable Code Length and Modulation Index for VLC-OFDM","authors":"Koki Yanashita, Kazushi Shimada, Seiya Hirano, C. Ahn","doi":"10.1109/ISPACS51563.2021.9651065","DOIUrl":"https://doi.org/10.1109/ISPACS51563.2021.9651065","url":null,"abstract":"Over the last few years, wireless communication technology has been developed rapidly, and high-speed and high-capacity communications are expected to become possible. On the other hand, the available frequency band is limited, and the issue of frequency band exhaustion becomes more serious. Visible Light Communication (VLC) is attracting attention as a solution to these problems. VLC is a technology that transmits signals by means of flicker or obscuration of visible light. The advantage of this technology is that it does not impose restrictions on the frequency band because lighting can be used as communication. Another advantage is that it is safe for the human body and does not impact other wireless devices [1] . In a VLC system based on orthogonal frequency division multiplexing (OFDM), Intensity Modulation - Direct Detection (IM/DD) is applied as a cost effective design [2] . The intensity modulated optical signal is transmitted, which is directly detected by the photodiode on the receiving side. Although the system is quite simple, it is susceptible to noise [1] . Therefore, it is important to improve the performance in a low SNR situation.","PeriodicalId":359822,"journal":{"name":"2021 International Symposium on Intelligent Signal Processing and Communication Systems (ISPACS)","volume":"127 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130448750","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}

引用次数: 1

Design and development of a LoRa based Water Quality Monitoring System 基于LoRa的水质监测系统的设计与开发

2021 International Symposium on Intelligent Signal Processing and Communication Systems (ISPACS) Pub Date : 2021-11-16 DOI: 10.1109/ISPACS51563.2021.9651127

Sheng-Tao Chen, Shih-Sung Lin, Chien-Wu Lan, Tai-I Chou

引用次数: 1

Diabetic Retinopathy Detection using CNN, Transformer and MLP based Architectures 基于CNN、Transformer和MLP架构的糖尿病视网膜病变检测

2021 International Symposium on Intelligent Signal Processing and Communication Systems (ISPACS) Pub Date : 2021-11-16 DOI: 10.1109/ISPACS51563.2021.9651024

N. S. Kumar, Badri Karthikeyan

{"title":"Diabetic Retinopathy Detection using CNN, Transformer and MLP based Architectures","authors":"N. S. Kumar, Badri Karthikeyan","doi":"10.1109/ISPACS51563.2021.9651024","DOIUrl":"https://doi.org/10.1109/ISPACS51563.2021.9651024","url":null,"abstract":"Diabetic retinopathy is a chronic disease caused due to a long term accumulation of insulin in the retinal blood vessels. 2.6% of global blindness is a result of diabetic retinopathy (DR) with more than 150 million people affected. Early detection of DR plays an important role in preventing blindness. Use of deep learning is a long term solution to screen, diagnose and monitor patients within primary health centers. Attention based networks (Transformers), Convolutional neural networks (CNN) and multi-layered perceptrons (MPLs) are the current state-of-the-art architectures for addressing computer vision based problem statements. In this paper, we evaluate these three different architectures for the detection of DR. Model convegence time (training time), accuracy, model size are few of the metrics that have been used for this evaluation. State-of-the-art pre-trained models belonging to each of these architectures have been chosen for these experiments. The models include EfficientNet, ResNet, Swin-Transformer, Vision-Transformer (ViT) and MLP-Mixer. These models have been trained using Kaggle dataset, which contains more than 3600 annotated images with a resolution of 2416*1736. For fair comparisons, no augmentation techniques have been used to improve the performance. Results of the experiments indicate that the models based on Transformer based architecture are the most accurate and also have comparative model-convergence times compared to CNN and MLP architectures. Among all the state-of-the-art pre-trained models Swin-Transformer yields the best accuracy of 86.4% on test dataset and it takes around 12 minutes for training the model on a Tesla K80 GPU.","PeriodicalId":359822,"journal":{"name":"2021 International Symposium on Intelligent Signal Processing and Communication Systems (ISPACS)","volume":"168 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131146236","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}

引用次数: 7

Deep Learning Based Lane Line Detection and Segmentation Using Slice Image Feature 基于切片图像特征的深度学习车道线检测与分割

2021 International Symposium on Intelligent Signal Processing and Communication Systems (ISPACS) Pub Date : 2021-11-16 DOI: 10.1109/ISPACS51563.2021.9651012

Jing. Guo, Herleeyandi Markoni

引用次数: 4

Mapping 3D road model to 2D street-view video using Image and Semantic Feature Matching 使用图像和语义特征匹配将3D道路模型映射到2D街景视频

2021 International Symposium on Intelligent Signal Processing and Communication Systems (ISPACS) Pub Date : 2021-11-16 DOI: 10.1109/ISPACS51563.2021.9651097

Kuan-Ting Chen, Jheng-Wei Su, K. Hsiao, Kuo-Wei Chen, Chih-Yuan Yao, Ruen-Rone Lee, Hung-Kuo Chu

引用次数: 0

Occluded Face Recognition Using Sparse Complex Matrix Factorization with Ridge Regularization 基于脊正则化的稀疏复矩阵分解的遮挡人脸识别

2021 International Symposium on Intelligent Signal Processing and Communication Systems (ISPACS) Pub Date : 2021-11-16 DOI: 10.1109/ISPACS51563.2021.9651107

Diyah Utami Kusumaning Putri, Aina Musdholifah, Faizal Makhrus, Viet-Hang Duong, Phuong Thi Le, Bo-Wei Chen, Jia-Ching Wang

引用次数: 1

Learn from one image: Dynamic One-shot learning based on parameter generation 从一张图像中学习:基于参数生成的动态一次性学习

2021 International Symposium on Intelligent Signal Processing and Communication Systems (ISPACS) Pub Date : 2021-11-16 DOI: 10.1109/ISPACS51563.2021.9651100

N. S. Kumar, M. Phirke, Anupriya Jayapal

{"title":"Learn from one image: Dynamic One-shot learning based on parameter generation","authors":"N. S. Kumar, M. Phirke, Anupriya Jayapal","doi":"10.1109/ISPACS51563.2021.9651100","DOIUrl":"https://doi.org/10.1109/ISPACS51563.2021.9651100","url":null,"abstract":"State-of-the-art deep learning algorithms are usually pre-trained on datasets containing millions of images. Adding new classes to these pre-trained networks, require large number of images for each of the new classes. Formulation of such large scale datasets usually require a lot of effort and time. The aim of this paper is to develop novel deep learning based one-shot learning framework which can achieve state-of-the-art results on new classes (one-shot classes) which have only one image each during the training phase. Adding these new one-shot classes, should not degrade the performance of the model on pre-trained classes. Multi-layer transformation function has been proposed in this paper for one-shot learning, where activations of a class are converted to their corresponding parameters. The model is pre-trained on large-scale base classes and the model adapts to new classes with zero training. Experiments were conducted on opensource datasets like MiniImageNet and Pascal-VOC using Nvidia K80 GPU. The model achieves an accuracy of 93.14% for large scale base classes and 64.69% for one-shot classes which is more than 3% better than the current state-of-the-art models.","PeriodicalId":359822,"journal":{"name":"2021 International Symposium on Intelligent Signal Processing and Communication Systems (ISPACS)","volume":"57 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132055145","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}

引用次数: 0