You Only Look Once (YOLO) has established itself as a prominent object detection framework due to its excellent balance between speed and accuracy. This article provides a thorough review of the YOLO series, from YOLOv1 to YOLOv10, including YOLOX, emphasizing their architectural advancements, loss function improvements, and performance enhancements. We have benchmarked the officially released versions from YOLOv3 to YOLOv10 and YOLOX, using widely recognized datasets VOC07+12 and COCO2017, on diverse hardware platforms: NVIDIA GTX Titan-X, RTX 3060, and Tesla V100. The benchmark provides significant insights, such as YOLOv9-E achieving the highest mean average precision (mAP) of 76.0% on VOC07+12 and also showing superior detection accuracy on COCO2017 with an mAP of 56.6% which is 1.2% higher than that of the latest YOLOv10-X. YOLOv9-E stands out for its superior detection accuracy making it more suitable for detection needs high accuracy such as analysis of medical images, while some lightweight versions like YOLOv5-S, YOLOv7-S, YOLOv8-S, and YOLOv10-S offer the great balance of accuracy and speed, making them ideal for real-time applications. Among them, YOLOv7-S has the highest mAP value among these lightweight models. Inference benchmarks highlight lightweight YOLO models such as YOLOv10-S for their exceptional inference speed on all GPUs and results of training time also indicate YOLOv9-E would take the longest time to converge among all versions using both datasets. This study would provide researchers and developers with some strategies in choosing appropriate YOLO models based on accuracy, resource availability, and application-specific needs.

This work reports both numerical and experimental studies of the reconditioning of a microwave plasma jet chemical vapor deposition (MPJCVD) system for the growth of diamond film. A three-dimensional plasma fluid model is constructed for investigating and conditioning the MPJCVD system and optimizing its operating conditions. The methodology solves electromagnetic wave and plasma dynamics self-consistently using an adaptive finite element method as implemented in COMSOL Multiphysics. The whole system has been modeled under varying parameters, including the reactor geometry, microwave power, and working gas pressure. Using an operating condition identical to the optimized simulation results, the MPJCVD system successfully fabricates a diamond-thin film on a graphite substrate. The SEM image reveals the presence of a diamond film uniformly distributed with particles of a size of ~1 μm. The field emission from the diamond film grown from our homemade MPJCVD system on the graphite substrate presents extraordinary properties, i.e., extremely high current density and relatively low turn-on voltage. The turn-on electric field observed could be as low as ~4 V/μm. This developed model provides valuable physical insights into the MPJCVD system, which guided performance improvements. The work may find applications in surface hardening and provide a better cold cathode for field electron emission.

Bilingual communication is increasingly prevalent in this globally connected world, where cultural exchanges and international interactions are unavoidable. Existing automatic speech recognition (ASR) systems are often limited to single languages. However, the growing demand for bilingual ASR in human–computer interactions, particularly in medical services, has become indispensable. This article addresses this need by creating an application programming interface (API)-based platform using VOSK, a popular open-source single-language ASR toolkit, to efficiently deploy a self-evaluated bilingual ASR system that seamlessly handles both primary and secondary languages in tasks like Mandarin–English mixed-speech recognition. The mixed error rate (MER) is used as a performance metric, and a workflow is outlined for its calculation using the edit distance algorithm. Results show a remarkable reduction in the Mandarin–English MER, dropping from ∼65% to under 13%, after implementing the self-evaluation framework and mixed-language algorithms. These findings highlight the importance of a well-designed system to manage the complexities of mixed-language speech recognition, offering a promising method for building a bilingual ASR system using existing monolingual models. The framework might be further extended to a trilingual or multilingual ASR system by preparing mixed-language datasets and computer development without involving complex training.

A capacitor is ideally considered as a pure capacitance, but under practical conditions, it also exhibits some resistive and inductive properties, called valent series resistance (ESR) and equivalent series inductance (ESL), respectively. These ESR and ESL characteristics are usually ignored in most cases as a first approximation. For high-power and/or high-frequency applications, especially pulsed systems, designed to achieve an accurate and reliable output, it is important to take these parasitic effects into consideration in circuit designs and operations. Conventionally, two capacitors are used, and two discharge tests are required to obtain the ESR and ESL values of the capacitors using a differential method. Besides, the two capacitors have to be identical. This makes an accurate determination very difficult and inefficient. In this work, we introduce a new method that can accurately and efficiently characterize the ESR, ESL, and capacitance of a high-voltage capacitor in a single discharge test employing only one capacitor and a spark gap. The electrical characterization is obtained by fitting the voltage drop across the capacitor and the current flowing through the spark gap during the discharge with simple circuit equations using a multiple linear regression method. Furthermore, the experiments have also been modeled more completely with PSpice simulations, resulting in good agreement and providing better understanding. It is found that the ESR, ESL, and capacitance obtained from the proposed methodology are consistent with the PSpice predictions, within differences of 5.71%, 4%, and 0.92%, respectively, in the demonstrated experiments.

This study investigates the feasibility and possible improvement of microwave power transmission (MPT) in a space-based solar power system (SSPS). SSPS, a concept proposed by Dr. Peter Glaser in 1968, aims to harness solar energy in space, free from atmospheric constraints, and transmit it to Earth using microwaves or lasers. Our focus is on enhancing the efficiency and cost-effectiveness of MPT, which accounts for a significant portion of SSPS expenses. In this work, we propose the use of a novel field emission-based rising-sun magnetron (RM) in the MPT subsystem, aiming to extend system longevity and simplify construction, while dramatically reducing the implementation cost. It is demonstrated that the optimization of the RM design at 2.45 GHz can achieve a high output power of >100 kW with a high efficiency of >85%, based on the well-established conformal finite-difference time-domain particle-in-cell simulations. This research provides valuable insights into improving SSPS, making it a more viable and sustainable renewable energy solution.

This work introduces a refined method for accurately and efficiently characterizing the electrical properties of a self-breakdown spark-gap switch using a single discharge test. In the experimental demonstration, a well-calibrated high-voltage (HV) capacitor is charged to approximately 10 kV and discharged through the spark gap under test. The simultaneously measured current and voltage waveforms across the spark gap are analyzed using a simple circuit model and a multiple-linear-regression method to concurrently determine the spark-gap inductance and resistance from a single discharge without requiring time-consuming plasma fluid or particle simulations. Ten sequential single-discharge tests yielded average values of 13.6 ± 0.3 nH and 0.202 ± 0.007 Ω for the inductance and resistance of the spark-gap, respectively. With the electrical properties of the HV capacitor known from previous characterization, the stray inductance and resistance of the connecting wires can be isolated and calculated by subtracting the capacitor and spark-gap contributions from the total circuit parameters. A PSpice parametric analysis is further employed to assess and verify the equivalent-circuit model by comparing the simulated waveforms with both the experimental waveforms and the fitting curves obtained from the simple circuit theory (SCT) analysis. This close agreement confirms the consistency of both the extracted spark-gap parameters and the calculated stray circuit parameters. These small statistical errors indicate that the proposed SCT method provides a practical approach for accurately and efficiently determining the inductance and resistance of a spark-gap for use in HV and pulsed-power applications.