Double Weight-Based SAR and Infrared Sensor Fusion for Automatic Ground Target Recognition with Deep Learning

A hierarchical recognition system (HRS) based on constrained Deep Belief Network (DBN) is proposed for SAR Automatic Target Recognition (SAR ATR). As a classical Deep Learning method, DBN has shown great performance on data reconstruction, big data mining, and classification. However, few works have been carried out to solve small data problems (like SAR ATR) by Deep Learning method. In HRS, the deep structure and pattern classifier are combined to solve small data classification problems. After building the DBN with multiple Restricted Boltzmann Machines (RBMs), hierarchical features can be obtained, and then they are fed to classifier directly. To obtain more natural sparse feature representation, the Constrained RBM (CRBM) is proposed with solving a generalized optimization problem. Three RBM variants,L1-RNM,L2-RBM, andL1/2-RBM, are presented and introduced to HRS in this paper. The experiments on MSTAR public dataset show that the performance of the proposed HRS with CRBM outperforms current pattern recognition methods in SAR ATR, like PCA + SVM, LDA + SVM, and NMF + SVM.

Download Full-text

Deep Learning and Multi-Sensor Fusion for Glioma Classification Using Multistream 2D Convolutional Networks

2018 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC) ◽

10.1109/embc.2018.8513556 ◽

2018 ◽

Cited By ~ 10

Author(s):

Chenjie Ge ◽

Irene Yu-Hua Gu ◽

Asgeir Store Jakola ◽

Jie Yang

Keyword(s):

Deep Learning ◽

Sensor Fusion ◽

Convolutional Networks ◽

Glioma Classification

Download Full-text

Underwater Target Recognition Based on Multi-Decision LOFAR Spectrum Enhancement: A Deep-Learning Approach

Future Internet ◽

10.3390/fi13100265 ◽

2021 ◽

Vol 13 (10) ◽

pp. 265

Author(s):

Jie Chen ◽

Bing Han ◽

Xufeng Ma ◽

Jian Zhang

Keyword(s):

Feature Extraction ◽

Deep Learning ◽

Recognition Accuracy ◽

Target Recognition ◽

Signal To Noise Ratio ◽

Recognition Performance ◽

Low Frequency ◽

Learning Approach ◽

Decision Algorithm ◽

Underwater Target

Underwater target recognition is an important supporting technology for the development of marine resources, which is mainly limited by the purity of feature extraction and the universality of recognition schemes. The low-frequency analysis and recording (LOFAR) spectrum is one of the key features of the underwater target, which can be used for feature extraction. However, the complex underwater environment noise and the extremely low signal-to-noise ratio of the target signal lead to breakpoints in the LOFAR spectrum, which seriously hinders the underwater target recognition. To overcome this issue and to further improve the recognition performance, we adopted a deep-learning approach for underwater target recognition, and a novel LOFAR spectrum enhancement (LSE)-based underwater target-recognition scheme was proposed, which consists of preprocessing, offline training, and online testing. In preprocessing, we specifically design a LOFAR spectrum enhancement based on multi-step decision algorithm to recover the breakpoints in LOFAR spectrum. In offline training, the enhanced LOFAR spectrum is adopted as the input of convolutional neural network (CNN) and a LOFAR-based CNN (LOFAR-CNN) for online recognition is developed. Taking advantage of the powerful capability of CNN in feature extraction, the recognition accuracy can be further improved by the proposed LOFAR-CNN. Finally, extensive simulation results demonstrate that the LOFAR-CNN network can achieve a recognition accuracy of 95.22%, which outperforms the state-of-the-art methods.

Download Full-text

Deep learning techniques‐based perfection of multi‐sensor fusion oriented human‐robot interaction system for identification of dense organisms

Cognitive Computation and Systems ◽

10.1049/ccs2.12010 ◽

2021 ◽

Author(s):

Haiju Li ◽

Chuntang Zhang ◽

Jingwen Bo ◽

Zhongjun Ding

Keyword(s):

Deep Learning ◽

Sensor Fusion ◽

Human Robot Interaction ◽

Robot Interaction ◽

Interaction System ◽

Learning Techniques

Download Full-text

COVIDSAVIOR: A Novel Sensor-Fusion and Deep Learning Based Framework for Virus Outbreaks

Frontiers in Public Health ◽

10.3389/fpubh.2021.797808 ◽

2021 ◽

Vol 9 ◽

Author(s):

Sharnil Pandya ◽

Anirban Sur ◽

Nitin Solke

Keyword(s):

Deep Learning ◽

Body Temperature ◽

Real Time ◽

Sensor Fusion ◽

Face Mask ◽

Temperature Threshold ◽

Mask Condition ◽

Essential Information ◽

The Individual ◽

Scanning Mechanism

The presented deep learning and sensor-fusion based assistive technology (Smart Facemask and Thermal scanning kiosk) will protect the individual using auto face-mask detection and auto thermal scanning to detect the current body temperature. Furthermore, the presented system also facilitates a variety of notifications, such as an alarm, if an individual is not wearing a mask and detects thermal temperature beyond the standard body temperature threshold, such as 98.6°F (37°C). Design/methodology/approach—The presented deep Learning and sensor-fusion-based approach can also detect an individual in with or without mask situations and provide appropriate notification to the security personnel by raising the alarm. Moreover, the smart tunnel is also equipped with a thermal sensing unit embedded with a camera, which can detect the real-time body temperature of an individual concerning the prescribed body temperature limits as prescribed by WHO reports. Findings—The investigation results validate the performance evaluation of the presented smart face-mask and thermal scanning mechanism. The presented system can also detect an outsider entering the building with or without mask condition and be aware of the security control room by raising appropriate alarms. Furthermore, the presented smart epidemic tunnel is embedded with an intelligent algorithm that can perform real-time thermal scanning of an individual and store essential information in a cloud platform, such as Google firebase. Thus, the proposed system favors society by saving time and helps in lowering the spread of coronavirus.

Download Full-text