Person classification from aerial imagery using local convolutional neural network features

被引:8
|
作者
Marasovic, Tea [1 ]
Papic, Vladan [1 ]
机构
[1] Univ Split, Fac Elect Engn Mech Engn & Naval Architecture, Rudera Boskovica 32, Split 21000, Croatia
关键词
Image classification - Network architecture - Image recognition - Classification (of information) - Convolution - Image representation - Disasters - Object recognition - Unmanned aerial vehicles (UAV) - Large dataset - Aerial photography - Chemical activation - Deep neural networks;
D O I
10.1080/01431161.2019.1597312
中图分类号
TP7 [遥感技术];
学科分类号
081102 ; 0816 ; 081602 ; 083002 ; 1404 ;
摘要
The need for search and rescue is not one that will go away. Extending beyond the concerns of hikers, natural disasters have brought the necessity for search and rescue in the heart of civilization. The advent of new technologies, such as unmanned aerial vehicles, can help decrease the cost of search and rescue operations and increase survival rates by finding the lost individuals in a faster manner. Visual object recognition is key to successful drone application for assisting search and rescue activities, and it is critical to develop a fully autonomous system. In recent years, deep convolutional neural networks have proven themselves as a powerful class of models and have become de facto a standard in a computer vision community. Moreover, a number of studies have shown that intermediate activations extracted from a deep convolutional neural network, pre-trained on large image dataset, can be adopted as universal image representation and transferred to other image classification tasks, leading to striking performances. In this paper, we investigate the effectiveness of different layer activations on the performance of convolutional features for binary person classification from UAV imagery, using various deep network architectures. Experiments have demonstrated that convolutional network generated features can deliver extremely competitive results, comparable to human-level performance. Furthermore, they reveal that in said context fully connected layer activations generalize well and are more expressive than the mid- or higher-level layers activations. Our work provides guidance for transferring pre-trained deep convolutional neural networks to address remote person classification tasks.
引用
收藏
页码:9084 / 9102
页数:19
相关论文
共 50 条
  • [1] Hyperspectral Imagery Classification Using Sparse Representations of Convolutional Neural Network Features
    Liang, Heming
    Li, Qi
    [J]. REMOTE SENSING, 2016, 8 (02)
  • [2] Classification of Local SAR Features with Convolutional Neural Network
    Arisoy, Sertac
    Kayabol, Koray
    [J]. 2017 25TH SIGNAL PROCESSING AND COMMUNICATIONS APPLICATIONS CONFERENCE (SIU), 2017,
  • [3] Disaster Detection from Aerial Imagery with Convolutional Neural Network
    Amit, Siti Nor Khuzaimah Binti
    Aoki, Yoshimitsu
    [J]. 2017 INTERNATIONAL ELECTRONICS SYMPOSIUM ON KNOWLEDGE CREATION AND INTELLIGENT COMPUTING (IES-KCIC), 2017, : 239 - 245
  • [4] Classification of Vehicles in Aerial Imagery using Deep Convolutional Neural Networks
    Mazurek, Przemyslaw
    Oszutowska-Mazurek, Dorota
    [J]. 2018 23RD INTERNATIONAL CONFERENCE ON METHODS & MODELS IN AUTOMATION & ROBOTICS (MMAR), 2018, : 413 - 418
  • [5] Peat Drainage Ditch Mapping from Aerial Imagery Using a Convolutional Neural Network
    Robb, Ciaran
    Pickard, Amy
    Williamson, Jennifer L.
    Fitch, Alice
    Evans, Chris
    [J]. REMOTE SENSING, 2023, 15 (02)
  • [6] Simultaneous classification of several features of a person’s appearance using a deep convolutional neural network
    Kukharenko A.I.
    Konushin A.S.
    [J]. Pattern Recognition and Image Analysis, 2015, 25 (3) : 461 - 465
  • [7] Habitat classification using convolutional neural networks and multitemporal multispectral aerial imagery
    Perez-Carabaza, Sara
    Boydell, Oisin
    O'Connell, Jerome
    [J]. JOURNAL OF APPLIED REMOTE SENSING, 2021, 15 (04)
  • [8] Multiple Pedestrians and Vehicles Tracking in Aerial Imagery Using a Convolutional Neural Network
    Azimi, Seyed Majid
    Kraus, Maximilian
    Bahmanyar, Reza
    Reinartz, Peter
    [J]. REMOTE SENSING, 2021, 13 (10)
  • [9] Classification of Hyperspectral Imagery Using a New Fully Convolutional Neural Network
    Li, Jiaojiao
    Zhao, Xi
    Li, Yunsong
    Du, Qian
    Xi, Bobo
    Hu, Jing
    [J]. IEEE GEOSCIENCE AND REMOTE SENSING LETTERS, 2018, 15 (02) : 292 - 296
  • [10] EEG Motor Imagery Classification using Fusion Convolutional Neural Network
    Zouch, Wassim
    Echtioui, Amira
    [J]. ICAART: PROCEEDINGS OF THE 14TH INTERNATIONAL CONFERENCE ON AGENTS AND ARTIFICIAL INTELLIGENCE - VOL 1, 2022, : 548 - 553