Non-contact penetrating detection and classification of human activities based on Micro-Doppler signatures (MDs) using UWB bio-radars are valuable tasks in various practical applications such as post-disaster search-and-rescue operations and urban military operations. However, for all classifiers, MD features of different magnitude activities at different positions are likely to result in classification errors due to MD attenuation and confusions. This paper proposes a classifier improving method called position-information-indexed classifier (PIIC). It aims at enhancing the performance of various classifiers in terms of recognition and classification. This method fully exploits the position information acquired by UWB bio-radar to create position-labeled modularized database of MD features. It also guides searching adaptively for optimal predict sub-model of PIICs for activity classification at a random position. We report through-wall detection and classification experimental results related to five activities within a range of 6 m. These results, based on four typical classifiers, demonstrates that PIIC-based classifiers can avoid those classification errors in an effective manner. Moreover, all PIIC-based classifiers presents a better classification performance with an average accuracy rise of 8.16% compared with those of overall-model-based classifiers. These performance evaluation experiments suggest that this method is strongly robust and stable, presenting wide applicability to various classifiers.

Radar has been widely applied in many scenarios as a critical remote sensing tool for non-contact vital sign monitoring, particularly for sleep monitoring and heart rate measurement within the home environment. For non-contact monitoring with radar, interference from house pets is an important issue that has been neglected in the past. Many animals have respiratory frequencies similar to those of humans, and they are easily mistaken for human targets in non-contact monitoring, which would trigger a false alarm because of incorrect physiological parameters from the animal. In this study, humans and common pets in families, such as dogs, cats, and rabbits, were detected using an impulse radio ultrawideband (IR-UWB) radar, and the echo signals were analyzed in the time and frequency domains. Subsequently, based on the distinct in-body structure between humans and animals, we propose a parameter, the respiratory and heartbeat energy ratio (RHER), which reflects the contribution rate of breathing and heartbeat in the detected vital signs. Combining this parameter with the energy index, we developed a novel scheme to distinguish between humans and animals. In the developed scheme, after background noise removal and direct-current component suppression, an energy indicator is used to initially identify the target. The signal is then decomposed using a variational mode decomposition algorithm, and the variational intrinsic mode functions that represent human respiration and heartbeat components are obtained and utilized to calculate the RHER parameter. Finally, the RHER index is applied to rapidly distinguish between humans and animals. Our experimental results demonstrate that the proposed approach more effectively distinguishes between humans and animals in terms of monitoring vital signs than the existing methods. Furthermore, its rapidity and need for only minimal calculation resources enable it to meet the needs of real-time monitoring.

Accurate segmentation of thyroid nodules in ultrasound images is crucial for the diagnosis of thyroid cancer and preoperative planning. However, the segmentation of thyroid nodules is challenging due to their irregular shape, blurred boundary, and uneven echo texture. To address these challenges, a novel Mamba- and ResNet-based dual-branch network (MRDB) is proposed. Specifically, the visual state space block (VSSB) from Mamba and ResNet-34 are utilized to construct a dual encoder for extracting global semantics and local details, and establishing multi-dimensional feature connections. Meanwhile, an upsampling–convolution strategy is employed in the left decoder focusing on image size and detail reconstruction. A convolution–upsampling strategy is used in the right decoder to emphasize gradual feature refinement and recovery. To facilitate the interaction between local details and global context within the encoder and decoder, cross-skip connection is introduced. Additionally, a novel hybrid loss function is proposed to improve the boundary segmentation performance of thyroid nodules. Experimental results show that MRDB outperforms the state-of-the-art approaches with DSC of 90.02% and 80.6% on two public thyroid nodule datasets, TN3K and TNUI-2021, respectively. Furthermore, experiments on a third external dataset, DDTI, demonstrate that our method improves the DSC by 10.8% compared to baseline and exhibits good generalization to clinical small-scale thyroid nodule datasets. The proposed MRDB can effectively improve thyroid nodule segmentation accuracy and has great potential for clinical applications.

In order to improve the speech acquisition ability of a non-contact method, a 94 GHz millimeter wave (MMW) radar sensor was employed to detect speech signals. This novel non-contact speech acquisition method was shown to have high directional sensitivity, and to be immune to strong acoustical disturbance. However, MMW radar speech is often degraded by combined sources of noise, which mainly include harmonic, electrical circuit and channel noise. In this paper, an algorithm combining empirical mode decomposition (EMD) and mutual information entropy (MIE) was proposed for enhancing the perceptibility and intelligibility of radar speech. Firstly, the radar speech signal was adaptively decomposed into oscillatory components called intrinsic mode functions (IMFs) by EMD. Secondly, MIE was used to determine the number of reconstructive components, and then an adaptive threshold was employed to remove the noise from the radar speech. The experimental results show that human speech can be effectively acquired by a 94 GHz MMW radar sensor when the detection distance is 20 m. Moreover, the noise of the radar speech is greatly suppressed and the speech sounds become more pleasant to human listeners after being enhanced by the proposed algorithm, suggesting that this novel speech acquisition and enhancement method will provide a promising alternative for various applications associated with speech detection.

Ultra-wideband (UWB) radar has been widely used for detecting human physiological signals (respiration, movement, etc.) in the fields of rescue, security, and medicine owing to its high penetrability and range resolution. In these applications, especially in rescue after disaster (earthquake, collapse, mine accident, etc.), the presence, number, and location of the trapped victims to be detected and rescued are the key issues of concern. Ample research has been done on the first issue, whereas the identification and localization of multi-targets remains a challenge. False positive and negative identification results are two common problems associated with the detection of multiple stationary human targets. This is mainly because the energy of the signal reflected from the target close to the receiving antenna is considerably stronger than those of the targets at further range, often leading to missing or false recognition if the identification method is based on the energy of the respiratory signal. Therefore, a novel method based on cross-correlation is proposed in this paper that is based on the relativity and periodicity of the signals, rather than on the energy. The validity of this method is confirmed through experiments using different scenarios; the results indicate a discernible improvement in the detection precision and identification of the multiple stationary targets.

Ultra-wide band (UWB) radar for short-range human target detection is widely used to find and locate survivors in some rescue missions after a disaster. The results of the application of bistatic UWB radar for detecting multi-stationary human targets have shown that human targets close to the radar antennas are very often visible, while those farther from radar antennas are detected with less reliability. In this paper, on account of the significant difference of frequency content between the echo signal of the human target and that of noise in the shadowing region, an algorithm based on wavelet entropy is proposed to detect multiple targets. Our findings indicate that the entropy value of human targets was much lower than that of noise. Compared with the method of adaptive filtering and the energy spectrum, wavelet entropy can accurately detect the person farther from the radar antennas, and it can be employed as a useful tool in detecting multiple targets by bistatic UWB radar.