This research proposes a nondestructive single-sensor acoustic emission (AE) scheme for the detection and localization of cracks in steel rail under loads. In the operation, AE signals were captured by the AE sensor and converted into digital signal data by AE data acquisition module. The digital data were denoised to remove ambient and wheel/rail contact noises, and the denoised data were processed and classified to localize cracks in the steel rail using a deep learning algorithmic model. The AE signals of pencil lead break at the head, web, and foot of steel rail were used to train and test the algorithmic model. In training and testing the algorithm, the AE signals were divided into two groupings (150 and 300 AE signals) and the classification accuracy compared. The deep learning-based AE scheme was also implemented onsite to detect cracks in the steel rail. The total accuracy (average F1 score) under the first and second groupings were 86.6% and 96.6%, and that of the onsite experiment was 77.33%. The novelty of this research lies in the use of a single AE sensor and AE signal-based deep learning algorithm to efficiently detect and localize cracks in the steel rail, unlike existing AE crack-localization technology that relies on two or more sensors and human interpretation.

This research proposes a through-wall S-band ultra-wideband (UWB) switched-antenna-array radar scheme for detection of stationary human subjects from respiration. The proposed antenna-array radar consists of one transmitting (Tx) and five receiving antennas (Rx). The Tx and Rx antennas are of Vivaldi type with high antenna gain (10 dBi) and narrow-angle directivity. The S-band frequency (2–4 GHz) is capable of penetrating non-metal solid objects and detecting human respiration behind a solid wall. Under the proposed radar scheme, the reflected signals are algorithmically preprocessed and filtered to remove unwanted signals, and 3D signal array is converted into 2D array using statistical variance. The images are reconstructed using back-projection algorithm prior to Sinc-filtered refinement. To validate the detection performance of the through-wall UWB radar scheme, simulations are carried out and experiments performed with single and multiple real stationary human subjects and a mannequin behind the concrete wall. Although the proposed method is an odd concept, the interest of this paper is applying the 1-Tx/5-Rx UWB switched-antenna array radar with the proposed method that is capable of distinguishing between the human subjects and the mannequin behind the concrete wall.

This research proposes a scheme of field programmable gate array (FPGA) to generate an impulse-radio ultra-wideband (IR-UWB) pulse. The FPGA scheme consists of three parts: digital clock manager, four-delay-paths stratagem, and edge combiner. The IR-UWB radar system is designed to detect human subjects from their respiration underneath the rubble in the aftermath of an earthquake and to locate the human subjects based on range estimation. The proposed IR-UWB radar system is experimented with human subjects lying underneath layers of stacked clay bricks in supine and prone position. The results reveal that the IR-UWB radar system achieves a pulse duration of 540 ps with a bandwidth of 2.073 GHz (fractional bandwidth of 1.797). In addition, the IR-UWB technology can detect human subjects underneath the rubble from respiration and identify the location of human subjects by range estimation. The novelty of this research lies in the use of the FPGA scheme to achieve an IR-UWB pulse with a 2.073 GHz (117 MHz–2.19 GHz) bandwidth, thereby rendering the technology suitable for a wide range of applications, in addition to through-obstacle detection.

Ultra-wideband (UWB) radar has become a critical remote-sensing tool for non-contact vital sign detection such as emergency rescues, securities, and biomedicines. Theoretically, the magnitude of the received reflected signal is dependent on the central frequency of mono-pulse waveform used as the transmitted signal. The research is based on the hypothesis that the stronger the received reflected signals, the greater the detectability of life signals. In this paper, we derive a new formula to compute the optimal central frequency to obtain as maximum received reflect signal as possible over the frequency up to the lower range of Ka-band. The proposed formula can be applicable in the optimization of hardware for UWB life detection and non-contact monitoring of vital signs. Furthermore, the vital sign detection results obtained by the UWB radar over a range of central frequency have been compared to those of the former continuous (CW) radar to provide additional information regarding the advantages and disadvantages of each radar.

This research proposes augmenting cropped computed tomography (CT) slices with data attributes to enhance the performance of a deep-learning-based automatic left-femur segmentation scheme. The data attribute is the lying position for the left-femur model. In the study, the deep-learning-based automatic left-femur segmentation scheme was trained, validated, and tested using eight categories of CT input datasets for the left femur (F-I–F-VIII). The segmentation performance was assessed by Dice similarity coefficient (DSC) and intersection over union (IoU); and the similarity between the predicted 3D reconstruction images and ground-truth images was determined by spectral angle mapper (SAM) and structural similarity index measure (SSIM). The left-femur segmentation model achieved the highest DSC (88.25%) and IoU (80.85%) under category F-IV (using cropped and augmented CT input datasets with large feature coefficients), with an SAM and SSIM of 0.117–0.215 and 0.701–0.732. The novelty of this research lies in the use of attribute augmentation in medical image preprocessing to enhance the performance of the deep-learning-based automatic left-femur segmentation scheme.

This research proposes a deep learning-driven impulse radio ultra-wideband (IR-UWB) multiantenna scheme for non-ionic breast tumor localization. The structure of the multiantenna scheme consists of one side slot Vivaldi transmitting (Tx) and nine side slot Vivaldi receiving (Rx₁ – Rx₉) antennas. To mitigate the attenuation and improve the diagnostic accuracy, the multiantenna scheme is rotated clockwise in 90° increments around the breast, with the angular position of the Tx antenna of 0°, 90°, 180°, and 270°. The deep learning algorithm is utilized to detect and localize the breast tumor, with 17 classification outputs, consisting of classifications 1 – 16 which correspond to 16 vertically discretized segments of the breast and classification 17 for cancer-free. Experiments were carried out using heterogenous breast replicas with a tumor of 1 cm in diameter, and the breast replicas possess the dielectric property and Hounsfield units (HU) similar to those of human breasts. The experimental results were compared with the computed tomography (CT) scan images. The results reveal that the multiantenna scheme could efficiently detect and accurately localize the breast tumor for nearly all classifications, with the total accuracy (average of F1 scores) of 99.11 %. Specifically, the novelty of this research lies in the use of deep learning with the IR-UWB technology to effectively localize breast tumors.