Deep ensemble model-based moving object detection and classification using SAR images

被引:1
|
作者
Paramasivam, Ramya [1 ]
Kumar, Prashanth [2 ]
Lai, Wen-Cheng [3 ]
Bidare Divakarachari, Parameshachari [4 ]
机构
[1] Mahendra Engn Coll, Dept Comp Sci Engn, Mallasamudram, India
[2] Univ Newcastle, Coll Engn Sci & Environm, Sch Informat & Phys Sci, Callaghan, NSW, Australia
[3] Ming Chi Univ Technol, Dept Elect Engn, New Taipei City, Taiwan
[4] Nitte Meenakshi Inst Technol, Dept Elect & Commun Engn, Bengaluru, India
关键词
bidirectional long short-term memory; improved deep belief network; moving object detection; recurrent neural network; synthetic aperture radar images; SHIP DETECTION; NETWORK; TARGETS;
D O I
10.3389/feart.2023.1288003
中图分类号
P [天文学、地球科学];
学科分类号
07 ;
摘要
In recent decades, image processing and computer vision models have played a vital role in moving object detection on the synthetic aperture radar (SAR) images. Capturing of moving objects in the SAR images is a difficult task. In this study, a new automated model for detecting moving objects is proposed using SAR images. The proposed model has four main steps, namely, preprocessing, segmentation, feature extraction, and classification. Initially, the input SAR image is pre-processed using a histogram equalization technique. Then, the weighted Otsu-based segmentation algorithm is applied for segmenting the object regions from the pre-processed images. When using the weighted Otsu, the segmented grayscale images are not only clear but also retain the detailed features of grayscale images. Next, feature extraction is carried out by gray-level co-occurrence matrix (GLCM), median binary patterns (MBPs), and additive harmonic mean estimated local Gabor binary pattern (AHME-LGBP). The final step is classification using deep ensemble models, where the objects are classified by employing the ensemble deep learning technique, combining the models like the bidirectional long short-term memory (Bi-LSTM), recurrent neural network (RNN), and improved deep belief network (IDBN), which is trained with the features extracted previously. The combined models increase the accuracy of the results significantly. Furthermore, ensemble modeling reduces the variance and modeling method bias, which decreases the chances of overfitting. Compared to a single contributing model, ensemble models perform better and make better predictions. Additionally, an ensemble lessens the spread or dispersion of the model performance and prediction accuracy. Finally, the performance of the proposed model is related to the conventional models with respect to different measures. In the mean-case scenario, the proposed ensemble model has a minimum error value of 0.032, which is better related to other models. In both median- and best-case scenario studies, the ensemble model has a lower error value of 0.029 and 0.015.
引用
收藏
页数:18
相关论文
共 50 条
  • [1] Deep ensemble based object detection from aerial images
    Park J.-C.
    Son S.-B.
    Lee S.-H.
    Jung J.-U.
    Park Y.-J.
    Oh H.-S.
    Journal of Institute of Control, Robotics and Systems, 2021, 27 (12) : 944 - 952
  • [2] Model-Based Information Extraction From SAR Images Using Deep Learning
    Sipos, Danijel
    Gleich, Dusan
    IEEE GEOSCIENCE AND REMOTE SENSING LETTERS, 2022, 19
  • [3] Ensemble of deep neural networks using acoustic environment classification for statistical model-based voice activity detection
    Hwang, Inyoung
    Park, Hyung-Min
    Chang, Joon-Hyuk
    COMPUTER SPEECH AND LANGUAGE, 2016, 38 : 1 - 12
  • [4] Moving Vehicle Detection and Classification Using Gaussian Mixture Model and Ensemble Deep Learning Technique
    Jagannathan, Preetha
    Rajkumar, Sujatha
    Frnda, Jaroslav
    Divakarachari, Parameshachari Bidare
    Subramani, Prabu
    WIRELESS COMMUNICATIONS & MOBILE COMPUTING, 2021, 2021
  • [5] Model-based neural network for target detection in SAR images
    Perlovsky, LI
    Schoendorf, WH
    Burdick, BJ
    Tye, DM
    IEEE TRANSACTIONS ON IMAGE PROCESSING, 1997, 6 (01) : 203 - 216
  • [6] Model-based Detection and Tracking of Single Moving Object using Laser Range Finder
    Abd Rahman, Abdul Hadi
    Zamzuri, Hairi
    Mazlan, Saiful Amri
    Rahman, Mohd Azizi Abdul
    PROCEEDINGS FIFTH INTERNATIONAL CONFERENCE ON INTELLIGENT SYSTEMS, MODELLING AND SIMULATION, 2014, : 556 - 561
  • [7] A Deep Detection Network Based on Interaction of Instance Segmentation and Object Detection for SAR Images
    Wu, Zitong
    Hou, Biao
    Ren, Bo
    Ren, Zhongle
    Wang, Shuang
    Jiao, Licheng
    REMOTE SENSING, 2021, 13 (13)
  • [8] Model-based tracking of moving object
    Jang, DS
    Kim, GY
    Choi, HI
    PATTERN RECOGNITION, 1997, 30 (06) : 999 - 1008
  • [9] Preliminary results from the model-based classification of airborne polarimetric SAR images
    Qong, M
    Nurmuhammad, A
    Tadono, T
    Wakabayashi, H
    Shimada, M
    Igarashi, T
    MICROWAVE REMOTE SENSING OF THE ATMOSPHERE AND ENVIRONMENT II, 2000, 4152 : 327 - 338
  • [10] Deep Transfer Model Based Local Climate Zone Classification Using SAR/Multispectral Images
    Nawaz, Amjad
    Chen, Jingyi
    Yang, Wei
    2024 4TH INTERNATIONAL CONFERENCE ON COMPUTER COMMUNICATION AND ARTIFICIAL INTELLIGENCE, CCAI 2024, 2024, : 81 - 86