2022 6th International Conference on Advanced Technologies for Signal and Image Processing (ATSIP)最新文献

筛选
英文 中文
Visual Tracking of mini-UAVs using Modified YOLOv5 and Improved DeepSORT Algorithms 基于改进YOLOv5和改进DeepSORT算法的小型无人机视觉跟踪
Tijeni Delleji, H. Fkih, A. Kallel, Zied Chtourou
{"title":"Visual Tracking of mini-UAVs using Modified YOLOv5 and Improved DeepSORT Algorithms","authors":"Tijeni Delleji, H. Fkih, A. Kallel, Zied Chtourou","doi":"10.1109/ATSIP55956.2022.9805923","DOIUrl":"https://doi.org/10.1109/ATSIP55956.2022.9805923","url":null,"abstract":"Over the past few years, unmanned aerial vehicles (UAVs) or drones have been deployed across the military services. At the same time, the civilian market has seen an exponential growth of essentially smaller systems (mini-UAVs) intended for public use. Recently, the malicious use of mini-UAVs has begun to emerge among terrorists, criminals and smugglers. The probability and frequency of the attacks of these devices, which have become more technologically advanced and cheaper, are both high and their impact can be very dangerous with devastating effects. Therefore, the need for counter-measure solution is highly required. For this purpose, the aim of this work is to propose a solution to early warning detect and track rogue mini-UAVs in restricted area. To meet this challenge, an online multiple object tracking (MOT) strategy is proposed based on YOLOv5 and the DeepSORT tracker. We adapt the former to the context of detecting small objects (ie.e. drones) by modifying the architecture. We enhance the discriminative power of DeppSORT tracker based on deep feature-based appearance. We combine detection and tracking that allows us the capability to handle small object localization. The experimental results show that the proposed multi-object tracking system is able to track mini-UAVs in real time.","PeriodicalId":328200,"journal":{"name":"2022 6th International Conference on Advanced Technologies for Signal and Image Processing (ATSIP)","volume":"76 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127078281","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: 3
Infrared-visible images’ fusion techniques for forest fire monitoring 森林火灾监测的红外-可见光图像融合技术
M. Tlig, M. Bouchouicha, M. Sayadi, E. Moreau
{"title":"Infrared-visible images’ fusion techniques for forest fire monitoring","authors":"M. Tlig, M. Bouchouicha, M. Sayadi, E. Moreau","doi":"10.1109/ATSIP55956.2022.9805812","DOIUrl":"https://doi.org/10.1109/ATSIP55956.2022.9805812","url":null,"abstract":"The forest fire represents a significant risk for many countries around the world. It often causes significant human, economic and ecological losses. Since only es or infrared images or visible images cannot offer accurate data and thanks to the technological advances of video-based fire detection systems, the fusion of visual and infrared images can play an essential role in fire detection. Therefore, the combination of visible and IR images is valuable because it combines the benefits of thermal radiation data in infrared images and detailed texture information in visible images.In this paper, we seek to compare infrared-visible fusion methods for forest fire imagery. First, we choose three image fusion approaches to test for the task of fire image fusion. Second, we introduce the fusion performance evaluation metrics. Finally, Infrared-visible fusion approaches are compared based on selected metrics.","PeriodicalId":328200,"journal":{"name":"2022 6th International Conference on Advanced Technologies for Signal and Image Processing (ATSIP)","volume":"76 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127445129","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: 2
Involving FCGR method in multiclass cancer diseases classification with transfer learning models 运用迁移学习模型将FCGR方法应用于多类肿瘤疾病分类
Ines Slimene, Imen Messaoudi, A. Oueslati, Z. Lachiri
{"title":"Involving FCGR method in multiclass cancer diseases classification with transfer learning models","authors":"Ines Slimene, Imen Messaoudi, A. Oueslati, Z. Lachiri","doi":"10.1109/ATSIP55956.2022.9805909","DOIUrl":"https://doi.org/10.1109/ATSIP55956.2022.9805909","url":null,"abstract":"Cancer, a widespread disease, can cause death if not diagnosed and treated in time. Any alteration of gene expression can be the cause of one or more cancers. In humans, the number of genes encoding proteins is estimated at 40,000 which presents a large data set. Analyzing these data helps in disease classification and identification. In this context, this paper will emphasize multi-class disease classification based on the gene expression alteration which is caused by miRNAs. The Frequency Chaos Game Representation technique (FCGR), which maps the occurrence frequency of words in the DNA sequence in the form of an image will define the gene’s features. Then, a transfer learning technique will be used to extract important features and classify our data in different disease types. The proposed approach achieved an overall accuracy of 0.96.","PeriodicalId":328200,"journal":{"name":"2022 6th International Conference on Advanced Technologies for Signal and Image Processing (ATSIP)","volume":"185 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123397746","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
Lung Disease Detection in Chest X-ray Images Using Transfer Learning 基于迁移学习的胸部x线图像肺部疾病检测
Ines Chouat, Amira Echtioui, R. Khemakhem, W. Zouch, M. Ghorbel, A. Hamida
{"title":"Lung Disease Detection in Chest X-ray Images Using Transfer Learning","authors":"Ines Chouat, Amira Echtioui, R. Khemakhem, W. Zouch, M. Ghorbel, A. Hamida","doi":"10.1109/ATSIP55956.2022.9805892","DOIUrl":"https://doi.org/10.1109/ATSIP55956.2022.9805892","url":null,"abstract":"Lung disease is a major health problem due to air pollution, smoking, and an aging population. For this reason, early and accurate detection is essential to achieve overall control of the disease and greatly increase the chances of successful medical treatment. This study proposes two convolutional neural network (CNN) models that rely on transfer learning to classify and detect the presence of pneumonia from a collection of chest X-ray (or CXR) images belonging to 4 classes. The data augmentation algorithm is used to increase the size of the training dataset in order to reduce overfitting and improve the model's generalization capacity.Using the data augmentation algorithm, we conducted a detailed evaluation of two pre-trained deep neural networks: VGGNet-16 and MobileNet. Deep learning (DL) is excellent in detecting infections, according to the findings. The MobileNet model outperforms the VGGNet-16 model where it achieved 82% accuracy, 83.5% recall, and 82.5% F1 score, while VGGNet-16 version gave the highest precision value which equals 84.5% but accuracy, recall and F1 score were respectively equal to 80%, 76% and 79%. On the other hand, and by calculating the AUC of the two models, MobileNet presents the best score which is 95% while VGGNet-16 has a score of 94%.","PeriodicalId":328200,"journal":{"name":"2022 6th International Conference on Advanced Technologies for Signal and Image Processing (ATSIP)","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117042441","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
Fusion Network for Super Resolution of UAVs Visible and Thermal Images 无人机可见光和热图像超分辨率融合网络
H. Fkih, T. Delleji, A. Kallel, Zied Chtourou
{"title":"Fusion Network for Super Resolution of UAVs Visible and Thermal Images","authors":"H. Fkih, T. Delleji, A. Kallel, Zied Chtourou","doi":"10.1109/ATSIP55956.2022.9805951","DOIUrl":"https://doi.org/10.1109/ATSIP55956.2022.9805951","url":null,"abstract":"Object identification is a fundamental and important problem in computer vision. Although impressive results have been obtained for medium to large objects in visible images, the performance on small objects, including mini/micro-UAVs (Unmanned Air Vehicle) in visible and thermal images, has received less attention. Therefore, due to the different appearances of these small flying objects, the identification of these objects, especially under poor early illumination conditions is currently a challenge. To overcome this challenge, we propose an increase in the resolution of the captured images by adding high frequency details which allows to produce better visual quality. In this work, we propose a Fusion Network for Super Resolution of UAVs Visible and Thermal Images (FNSR), which aims to produce a Hight Resolution image (HR) using visible and infrared (IR) reference images (Ref). FNSR is made up of five related modules, including a multi-scale texture extractor for feature extraction from visible and thermal images, a spatial fusion module, a relevance embedding module, a features transfer block and a spatial decoder module. We introduce channel attention mechanisms to highlight the most promising features, as well as a Spatial Total Variation (STV) loss to preserve spatial information. Experimental results on image pairs (IR/Visible) show that the FNSR is capable of producing high quality super-resolved images. Compared to the State-Of-The-Art (SOTA), it presents better performances.","PeriodicalId":328200,"journal":{"name":"2022 6th International Conference on Advanced Technologies for Signal and Image Processing (ATSIP)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128494252","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: 2
Effects of Spatial Filtering on PolSAR Parameter Estimation 空间滤波对PolSAR参数估计的影响
M. Yahia, Tarig Ali, M. Mortula, R. Abdelfattah
{"title":"Effects of Spatial Filtering on PolSAR Parameter Estimation","authors":"M. Yahia, Tarig Ali, M. Mortula, R. Abdelfattah","doi":"10.1109/ATSIP55956.2022.9805982","DOIUrl":"https://doi.org/10.1109/ATSIP55956.2022.9805982","url":null,"abstract":"Speckle filtering is essential for accurate parameter estimation from polarimetric synthetic aperture (PolSAR) data. To reduce speckle, sufficiently large independent samples (i. e. N looks) are generally averaged. However, in real scene, these N-looks are not provided. To surmount this limitation, by exploiting the ergodicity of the SAR data, the multilooking is substituted by the spatial averaging of N homogenous neighboring pixels. In this letter, it has been demonstrated that the multilooking and spatial averaging are quite different. First, the spatial averaging introduced a correlation between the pixels whereas in the multilooking they remain independent. Secondly, in real scene, the spatial averaging estimation of the polarimetric parameters (i.e. intensity mean and variance, equivalent number of looks (ENL), entropy and alpha angle) is nosier and more biased than the multilooked ones. These results should be taken into account when dealing with PolSAR parameter estimation of real data from multilooked PolSAR data models.","PeriodicalId":328200,"journal":{"name":"2022 6th International Conference on Advanced Technologies for Signal and Image Processing (ATSIP)","volume":"97 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129245786","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
Combination of Volumetric and Topologic Brain Characteristics Towards a Diagnosis of Alzheimer’s Disease in his Earlier Stage 结合脑体积和拓扑特征对早期阿尔茨海默病的诊断
Besma Mabrouk, N. Bouzidi, C. Mhiri, A. Hamida
{"title":"Combination of Volumetric and Topologic Brain Characteristics Towards a Diagnosis of Alzheimer’s Disease in his Earlier Stage","authors":"Besma Mabrouk, N. Bouzidi, C. Mhiri, A. Hamida","doi":"10.1109/ATSIP55956.2022.9805914","DOIUrl":"https://doi.org/10.1109/ATSIP55956.2022.9805914","url":null,"abstract":"The longevity of life and the rapid advances in the medical field have accentuated more and more neurodegenerative diseases. More than 60% of cases are affected by Alzheimer, an incurable disease. Which imposes special attention for the diagnosis of the dementia in his earlier stage. Voxel-based morphometry (VBM) has been effectively used, in association with Structural Magnetic Resonance Imaging (MRI), to predict multiple neurodegenerative diseases. Recently, the emergence of Diffusion-Weighted imaging opened access to brain comprehension. In this study, we focus on this two-imaging modality to emphasize the contribution of hemispheric asymmetries in the earlier-stage diagnosis of Alzheimer. For this aim, 209 T1 weighted brain MRIs, have been examined: 55 normal control (NC), 106 mild cognitive impairment (MCI), and 48 AD. Volumetric measurements of subcortical regions and topologic proprieties of brain mapping were combined. All in all, features used in this work are the average shortest path, the Average Clustering coefficient, and the small world graph characteristics, although, Hippocampus and Amygdala Volume. When left and right characteristic values were determined, asymmetry scores were calculated. For the classification task, a random forest classifier and support vector machine (SVM) were performed, then compared. They provided respectively an average of 0.72/0.60 for accuracy and an equality of 0.77 for the Area under the curve. Both used algorithms have proven their efficiency in the MCI group prediction, which confirms hemisphere asymmetry as a biomarker of disease in his prodromal stage.","PeriodicalId":328200,"journal":{"name":"2022 6th International Conference on Advanced Technologies for Signal and Image Processing (ATSIP)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123141070","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
Multiple Sclerosis Segmentation using Deep Learning Models : Comparative Study 使用深度学习模型的多发性硬化症分割:比较研究
Abdelkader Alrabai, Amira Echtioui, A. Hamida
{"title":"Multiple Sclerosis Segmentation using Deep Learning Models : Comparative Study","authors":"Abdelkader Alrabai, Amira Echtioui, A. Hamida","doi":"10.1109/ATSIP55956.2022.9805983","DOIUrl":"https://doi.org/10.1109/ATSIP55956.2022.9805983","url":null,"abstract":"Central Nervous System (CNS) has directly been affected by many demyelination diseases. Multiple Sclerosis (MS) is one among these diseases. Such a series pathologies could be explored of the CNS which can be imagined through Magnetic Resonance Imaging (MRI) scans is more prone to be affected. The progression of the disease could be determined by the detection of all (MS) lesions and especially their multiplication. This will also help to supervise the efficacy of a candidate treatment. Medical image segmentation is said to be another great contributor in the field of medicine. The recent methods like Deep Learning (DL), specifically, Convolutional Neural Networks (CNNs) associated with various applications has been excellent too. This study reviews the technique of MS Segmentation based on (DL) in MRI slices. This research study reviews several techniques of MS segmentation based on (DL) in MRI slices and the focus is on using U-Net architecture. It also focuses on simple adjustments to the well-known U-net and evaluates the impact of modifications. An overview of current (DL) based medical image segmentation techniques is provided to aid researchers address existing problems. Significant improvements in segmentation quality compared to competing methods are demonstrated by using the Dice metric.","PeriodicalId":328200,"journal":{"name":"2022 6th International Conference on Advanced Technologies for Signal and Image Processing (ATSIP)","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126513759","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: 2
A combined CNN-LSTM Network for Audio Emotion Recognition using Speech and Song attributs 基于语音和歌曲属性的音频情感识别的CNN-LSTM组合网络
Souha Ayadi, Z. Lachiri
{"title":"A combined CNN-LSTM Network for Audio Emotion Recognition using Speech and Song attributs","authors":"Souha Ayadi, Z. Lachiri","doi":"10.1109/ATSIP55956.2022.9805924","DOIUrl":"https://doi.org/10.1109/ATSIP55956.2022.9805924","url":null,"abstract":"Audio emotion recognition is a very active topic. It has many applications in various domains like health care, information technology and defence. This paper presents a hybrid CNN-LSTM Deep Network for audio emotion classification using speech and song description. The model combines the temporal modelling ability of a long shirt term memory (LSTM) with the CNN ability to learn invariant features. We evaluate our experements on RAVDESS Database that contains audio song and audio speech formats. We show the affectiveness of both CNN and LSTM neural networks for feature extraction and memorizing a long sentence in order to provide the stability and a good performance of the model by using MFCC which reinforces the idea of using frames instead of waves to improve the training process. We compare the obtained results for both modalities for the same amount number of emotions which are “happy”, “sad”, “angree”, “neutral”, “calm” and “fear”.","PeriodicalId":328200,"journal":{"name":"2022 6th International Conference on Advanced Technologies for Signal and Image Processing (ATSIP)","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132964703","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: 2
Training Hacks and a Frugal Man’s Net with Application to Glioblastoma Segmentation 训练技巧和一个节俭的人的网与应用胶质母细胞瘤分割
Jawher Ben Abdallah, L. Marrakchi-Kacem, I. Rekik
{"title":"Training Hacks and a Frugal Man’s Net with Application to Glioblastoma Segmentation","authors":"Jawher Ben Abdallah, L. Marrakchi-Kacem, I. Rekik","doi":"10.1109/ATSIP55956.2022.9805950","DOIUrl":"https://doi.org/10.1109/ATSIP55956.2022.9805950","url":null,"abstract":"In this paper, we investigate the effectiveness of training a sparse Neural Network on a limited number of samples in the context of brain tumor segmentation. Nowadays, Deep Learning architectures are getting deeper, more sophisticated and environmentally unfriendly in an effort to improve their segmentation performance. We use a brain tumor segmentation dataset and apply simple practices to reduce the needed computational resources to allow cheap and fast training. We also present a lighter, cheaper version of the U-Net dubbed Frugal U-Net stemming from our investigation on how far we can push the original U-Net by decreasing its parameter count using Depth-Wise Separable Convolutions instead of regular ones, all the while preserving the minimum levels of accuracy required in Medical Imaging. Our methodology is useful in clinical facilities where high-computation resources are limited.","PeriodicalId":328200,"journal":{"name":"2022 6th International Conference on Advanced Technologies for Signal and Image Processing (ATSIP)","volume":"113 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131008354","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
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
相关产品
×
本文献相关产品
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:604180095
Book学术官方微信
小红书