Towards Uncovering Feature Extraction From Temporal Signals in Deep CNN: the ECG Case Study

Author(s):  
Jacopo Ferretti ◽  
Pietro Barbiero ◽  
Vincenzo Randazzo ◽  
Giansalvo Cirrincione ◽  
Eros Pasero
Keyword(s):  
2016 ◽  
Vol 83 ◽  
pp. 1262-1267 ◽  
Author(s):  
Leonardo Feltrin ◽  
João Gabriel Motta ◽  
Feras Al-Obeidat ◽  
Farhi Marir ◽  
Martina Bertelli

Research problem introduction. The main research goal of this paper is to provide the urban geosystem research concept with both the theoretical basics presentation of GIS involvement in urban studies, and with examples of its practical applications. An urbogeosystem (UGS) has been presented not as a simple aggregate of cities, but as the emergent entity that produced complicated interconnections and interdependencies among its constituents. By the urbogeosystem concept the authors attempt to introduce a reliable research approach that has been deliberately developed to identify the nature and spatial peculiarities of the urbanization process in a given area. The exigency of this concept elaboration is listed by the number of needs and illustrated with ordinary 2D digital city cadaster limitations. The methodological background has been proposed, and its derivative applied solutions meet the number of necessities for more efficient urban mapping, city understanding, and municipal mana-gement. The geoinformation concept of the urban geographic system research. External and internal urbogeosystems. The authors explain why an UGS can be formalized as three major components: an aggregate of point features, a set of lines, an aggregate of areal features. The external UGS represents a set of cities, the internal one – a set of delineated areas within one urban territory. Algorithmic sequence of the urbogeosystem study with a GIS. The authors introduce algorithmic sequence of research provision with GIS, in which the LiDAR data processing block has been examined in the details with the procedure of the automated feature extraction explanation. Relevant software user interface sample of the visualization of the urban modeled feature attributes is provided. A case study of the external urbogeosystem. The regional case study of the external urbogeosystem modeling is introduced with GIS MapInfo Professional. The authors present the spatial econometric analysis for commuting study directed to a regional workforce market. The results of the external UGS research mainly correspond to some published social economic regularities in the area, but nonetheless it also demonstrates significant deviations that may be explained by this system’s emergent properties. Case studies of the internal urbogeosystem of Kharkiv-City. Two case studies of the internal urbogeosystem of Kharkiv City have been demonstrated, too. In the first one, automated feature extraction provided by the authors’ original software from LiDAR data has been applied for modeling this UGS content throughout a densely built-up urban parcel. In another case the GIS-analysis of the urbogeosystem functional impact on the catering services spatial distribution has been provided with the ArcGIS software. Results and conclusion. Summarizing all primary and derivative data processed with this technique as well as generalizing key ideas discussed in the text, the authors underline this whole methodological approach as such that can be considered as a general outlining showing how to use geoinformation software for the analysis of urban areas. Concluding their research, the authors emphasize that the urbogeosystem concept may be quite useful for visualization and different analysis applied for urban areas, including city planning, facility and other municipal management methods. The short list of the obtained results has been provided at the end of the text.


Author(s):  
Najme Mansouri ◽  
Gholam Reza Khayati ◽  
Behnam Mohammad Hasani Zade ◽  
Seyed Mohammad Javad Khorasani ◽  
Roya Kafi Hernashki

Author(s):  
Chunmian Lin ◽  
Lin Li ◽  
Zhixing Cai ◽  
Kelvin C. P. Wang ◽  
Danny Xiao ◽  
...  

Automated lane marking detection is essential for advanced driver assistance system (ADAS) and pavement management work. However, prior research has mostly detected lane marking segments from a front-view image, which easily suffers from occlusion or noise disturbance. In this paper, we aim at accurate and robust lane marking detection from a top-view perspective, and propose a deep learning-based detector with adaptive anchor scheme, referred to as A2-LMDet. On the one hand, it is an end-to-end framework that fuses feature extraction and object detection into a single deep convolutional neural network. On the other hand, the adaptive anchor scheme is designed by formulating a bilinear interpolation algorithm, and is used to guide specific-anchor box generation and informative feature extraction. To validate the proposed method, a newly built lane marking dataset contained 24,000 high-resolution laser imaging data is further developed for case study. Quantitative and qualitative results demonstrate that A2-LMDet achieves highly accurate performance with 0.9927 precision, 0.9612 recall, and a 0.9767 [Formula: see text] score, which outperforms other advanced methods by a considerable margin. Moreover, ablation analysis illustrates the effectiveness of the adaptive anchor scheme for enhancing feature representation and performance improvement. We expect our work will help the development of related research.


2014 ◽  
Vol 24 (6) ◽  
pp. 2979-2986
Author(s):  
Fernando Jorge-Hernandez ◽  
Yolanda Garcia Chimeno ◽  
Begonya Garcia-Zapirain ◽  
Alberto Cabrera Zubizarreta ◽  
Maria Angeles Gomez Beldarrain ◽  
...  

Sign in / Sign up

Export Citation Format

Share Document