A Three-Stage Low-Complexity Human Fall Detection Method Using IR-UWB Radar

Chen, Mengxia; Yang, Zhaocheng; Lai, Jialei; Chu, Ping; Lin, Jian

doi:10.1109/jsen.2022.3184513

Cited by 16 publications

(5 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The same argument can be applied to [41], where the study proposed the lightest model yet in literature at the cost of dropping the F1 score to 90.98%, considering only a simple dataset for testing. Moreover, in [42], a lightweight model with 259K parameters was developed and achieved 97.6% recall, but the dataset was simple and contained only 9 movements. On the contrary, some previous works [28] used a high level of engineering for the radar signal, where the range, velocity, and AoA are extracted and processed to generate a 3D point cloud localization for movements.…”

Section: Literature Reviewmentioning

confidence: 99%

“…Current systems, despite being designed for simple life scenarios, claim high performance but reach high hardware and/or software complexity [38]- [40]. On the other hand, some low-complexity systems were proposed, but they suffer from low performance in complex real-world and even in ideal scenarios [41], [42]. This unoptimized performancecomplexity balance results from approaching the radar classification problem similar to approaching image classification problems [43].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

IntNet: Lightweight yet High-Performance Deep Learning System for Intuitive Radar Patterns Analysis and Human Fall Detection

Almallah,

Sababha

2023

Preprint

View full text Add to dashboard Cite

show abstract

Section: Literature Reviewmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

IntNet: Lightweight yet High-Performance Deep Learning System for Intuitive Radar Patterns Analysis and Human Fall Detection

Almallah,

Sababha

2023

Preprint

View full text Add to dashboard Cite

show abstract

“…For traditional machine learning methods, low complexity and easily integrable classifiers such as SVM, decision trees, and k-nearest neighbors (KNN) are often used in combination with radar signal processing techniques to detect fall events. Chen used UWB radar with cascaded support vector data description (SVDD) and Mahalanobis distance classifier to detect falls using Doppler and distance features, achieving a recognition accuracy of 97% 16 . Compared to other sensors, radar sensors do not capture human image or audio information, thus ensuring privacy protection 17 .…”

Section: Introductionmentioning

confidence: 99%

A robust fall detection approach based on 4D imaging millimeter-wave radar

Xiao,

Yang,

Chu

et al. 2024

Fifteenth International Conference on Signal Processing Systems (ICSPS 2023)

Self Cite

View full text Add to dashboard Cite

Falls are the leading cause of injuries and even fatalities among elderly individuals in home environments, resulting in the development of fall detection technology particularly crucial. In this paper, we propose a robust human fall detection method based on the millimeter-wave radar and 4D point cloud imaging technology. The main objective of this method is to detect various types of fall actions in real-time and provide timely alerts to assist the fallen individuals. In our proposed method, we first perform range-FFT and static clutter suppression on the radar echo data. Subsequently, we conduct rangedomain target detection and angle estimation to generate initial point cloud information. Next, we introduce the median absolute deviation (MAD) based outlier removal method to eliminate non-human body outliers from the point cloud. Lastly, we present a suspected fall detection (SFD) method and a secondary fall detection method based on support vector machines (SVM) to maintain high detection accuracy while minimizing false alarms. The experimental results demonstrate that the average detection accuracy of our method for different types of falls is 97.5%, with an average false alarm rate of 0.4%.

show abstract

“…With the information obtained from human targets after processing radar echoes, real-time monitoring and analyses of the human body status have become a possible noninvasive healthcare approach. Furthermore, UWB radar has great research value and application potential because of its noncontact detection and privacy protection abilities in healthcare [13,14], clinical monitoring [10,15], the smart home industry [16,17], and various indoor applications [18,19], which arouses continuous interest among researchers around the globe.…”

Section: Introductionmentioning

confidence: 99%

Two-Step Accuracy Improvement for Multitarget Detection in Complex Environment Using UWB Radar

Liang,

Jin,

Yang

et al. 2024

Remote Sensing

View full text Add to dashboard Cite

Detecting multiple human targets in indoor scenarios using ultra-wideband (UWB) radar usually involves false detection results caused by the secondary reflections, which might reduce the target detection accuracy and cause a more severe deterioration when the number of targets increases. This article proposed a two-step accuracy improvement method for multitarget detection in environments with multiple human targets of more than three and strong secondary reflections by the surroundings, especially the walls. Based on the rough detection results acquired by the modified CA-CFAR (MCA-CFAR) processing, the first step achieves the primary false alarm suppression using a short-window accumulation in the time domain. Then, the second step applies the decision confidence on the detection results from the first step to assess the reliability of results for improved accuracy. The two-step accuracy improvement could thus have a higher accuracy through cascading false alarm suppression. The effectiveness and accuracy of the proposed algorithm are verified based on the experimental results.

show abstract

A Three-Stage Low-Complexity Human Fall Detection Method Using IR-UWB Radar

Cited by 16 publications

References 38 publications

IntNet: Lightweight yet High-Performance Deep Learning System for Intuitive Radar Patterns Analysis and Human Fall Detection

IntNet: Lightweight yet High-Performance Deep Learning System for Intuitive Radar Patterns Analysis and Human Fall Detection

A robust fall detection approach based on 4D imaging millimeter-wave radar

Two-Step Accuracy Improvement for Multitarget Detection in Complex Environment Using UWB Radar

Contact Info

Product

Resources

About