Search

This paper proposes a highly sensitive and high-resolution resonant MEMS electrostatic field sensor based on electrostatic stiffness perturbation, which uses resonant frequency as an output signal to eliminate the feedthrough interference from the driving voltage. The sensor is composed of a resonator, driving electrode, detection electrode, transition electrode, and electrostatic field sensing plate. The working principle is that when there is an electrostatic field, an induction charge will appear at the surface of the electrostatic field sensing plate and induce electrostatic stiffness on the resonator, which will cause a resonant frequency shift. The resonant frequency is used as the output signal of the microsensor. The characteristics of the electrostatic field sensor are analyzed with a theoretical model and verified by finite element simulation. A device prototype is fabricated based on the Silicon on Insulator (SOI) process and tested under vacuum conditions. The results indicate that the sensitivity of the sensor is 0.1384 Hz / (kV / m) and the resolution is better than 10 V/m. [ABSTRACT FROM AUTHOR]

This paper proposes a novel wafer-level vacuum packaged electric field microsensor (EFM) featuring a high quality factor, low driving voltage, low noise, and low power consumption. The silicon-on-insulator (SOI) conductive handle layer was innovatively used as the sensing channel to transmit the external electric field to the surface of the sensitive structure, and the vacuum packaging was realized through anodic bonding between the SOI and glass-on-silicon (GOS). The fabrication process was designed and successfully realized, featured with a simplified process and highly efficient batch manufacturing, and the final chip size was only 5 × 5 mm. A theoretical model for the packaged device was set up. The influence of key parameters in the packaging structure on the output characteristics of the microsensor was analyzed on the basis of the proposed model. Experiments were conducted on the wafer-level vacuum-packaged EFM to characterize its performance. Experimental results show that, under the condition of applying 5 V DC driving voltage, the required AC driving voltage of the sensor was only 0.05 VP, and the feedthrough was only 4.2 mV. The quality factor was higher than 5000 and was maintained with no drop in the 50-day test. The vacuum in the chamber of the sensor was about 10 Pa. A sensitivity of 0.16 mV/(kV/m) was achieved within the electrostatic field range of 0–50 kV/m. The linearity of the microsensor was 1.62%, and the uncertainty was 4.42%. [ABSTRACT FROM AUTHOR]

One of the major concerns in the development of three-dimensional (3D) electric field sensors (EFSs) is their susceptibility to cross-axis coupling interference. The output signal for each sensing axis of a 3D EFS is often coupled by electric field components from the two other orthogonal sensing axes. In this paper, a one-dimensional (1D) electric field sensor chip (EFSC) with low cross-axis coupling interference is presented. It is designed to be symmetrical, forming a pair of in-plane symmetrically-located sensing structures. Using a difference circuit, the 1D EFSC is capable of sensing parallel electric fields along symmetrical structures and eliminating cross-axis coupling interference, which is contrast to previously reported 1D EFSCs designed for perpendicular electric field component measurement. Thus, a 3D EFS with low cross-axis coupling interference can be realized using three proposed 1D EFSCs. This 3D EFS has the advantages of low cross-axis coupling interference, small size, and high integration. The testing and calibration systems of the proposed 3D EFS were developed. Experimental results show that in the range of 0–120 kV/m, cross-axis sensitivities are within 5.48%, and the total measurement errors of this 3D EFS are within 6.16%. [ABSTRACT FROM AUTHOR]

This paper proposes a high sensitivity electric field microsensor (EFM) based on torsional resonance. The proposed microsensor adopts torsional shutter, which is composed of shielding electrodes and torsional beams. The movable shielding electrodes and the fixed sensing electrodes are fabricated on the same plane and interdigitally arranged. Push-pull electrostatic actuation method is employed to excite the torsional shutter. Simulation results proved that the torsional shutter has higher efficiency of charge induction. The optimization of structure parameters was conducted to improve its efficiency of charge induction further. A micromachining fabrication process was developed to fabricate the EFM. Experiments were conducted to characterize the EFM. A good linearity of 0.15% was achieved within an electrostatic field range of 0-50 kV/m, and the uncertainty was below 0.38% in the three roundtrip measurements. A high sensitivity of 4.82 mV/(kV/m) was achieved with the trans-resistance of 100 MO, which is improved by at least one order of magnitude compared with previously reported EFMs. The efficiency of charge induction for this microsensor reached 48.19 pA/(kV/m). [ABSTRACT FROM AUTHOR]

Electrostatic voltage is a vital parameter in industrial production lines, for reducing electrostatic discharge harms and improving yields. Due to such drawbacks as package shielding and low resolution, previously reported electric field microsensors are still not applicable for industrial static monitoring uses. In this paper, we introduce a newly designed microsensor package structure, which enhances the field strength inside the package cavity remarkably. This magnification effect was studied and optimized by both theoretical calculation and ANSYS simulation. By means of the digital synthesizer and digital coherent demodulation method, the compact signal processing circuit for the packaged microsensor was also developed. The meter prototype was calibrated above a charged metal plate, and the electric field resolution was 5 V/m, while the measuring error was less than 3 V, from −1 kV to 1 kV in a 2 cm distance. The meter was also installed into a production line and showed good consistency with, and better resolution than, a traditional vibratory capacitance sensor. [ABSTRACT FROM AUTHOR]

This paper proposes an electric field microsensor (EFM) with mutual shielding electrodes. Based on the charge-induction principle, the EFM consists of fixed electrodes and piezoelectric-driving vertically-movable electrodes. All the fixed electrodes and movable electrodes work as both sensing electrodes and shielding electrodes. In other words, all the fixed and movable electrodes are sensing electrodes, and they are mutually shielding electrodes simultaneously. The movable electrodes are driven to periodically modulate the electric field distribution at themselves and the fixed electrodes, and the induced currents from both movable and fixed electrodes are generated simultaneously. The electrode structure adopts an interdigital structure, and the EFM has been simulated by finite element methods. Simulation results show that, since the sensing area of this EFM is doubled, the variation of induced charge is twice, and therefore the output signal of the sensor is increased. The piezoelectric material, lead zirconate titanate (PZT), is prepared by the sol–gel method, and the microsensor chip is fabricated. [ABSTRACT FROM AUTHOR]

As one of the most important active remote sensing technologies, synthetic aperture radar (SAR) provides advanced advantages of all-day, all-weather, and strong penetration capabilities. Due to its unique electromagnetic spectrum and imaging mechanism, the dimensions of remote sensing data have been considerably expanded. Important for fundamental research in microwave remote sensing, SAR image classification has been proven to have great value in many remote sensing applications. Many widely used SAR image classification algorithms rely on the combination of hand-designed features and machine learning classifiers, which still experience many issues that remain to be resolved and overcome, including optimized feature representation, the fuzzy confusion of speckle noise, the widespread applicability, and so on. To mitigate some of the issues and to improve the pattern recognition of high-resolution SAR images, a ConvCRF model combined with superpixel boundary constraint is developed. The proposed algorithm can successfully combine the local and global advantages of fully connected conditional random fields and deep models. An optimizing strategy using a superpixel boundary constraint in the inference iterations more efficiently preserves structure details. The experimental results demonstrate that the proposed method provides competitive advantages over other widely used models. In the land cover classification experiments using the MSTAR, E-SAR and GF-3 datasets, the overall accuracy of our proposed method achieves 90.18 ± 0.37, 91.63 ± 0.27, and 90.91 ± 0.31, respectively. Regarding the issues of SAR image classification, a novel integrated learning containing local and global image features can bring practical implications. [ABSTRACT FROM AUTHOR]

Fine classification of vegetation types has always been the focus and difficulty in the application field of remote sensing. Unmanned Aerial Vehicle (UAV) sensors and platforms have become important data sources in various application fields due to their high spatial resolution and flexibility. Especially, UAV hyperspectral images can play a significant role in the fine classification of vegetation types. However, it is not clear how the ultrahigh resolution UAV hyperspectral images react in the fine classification of vegetation types in highly fragmented planting areas, and how the spatial resolution variation of UAV images will affect the classification accuracy. Based on UAV hyperspectral images obtained from a commercial hyperspectral imaging sensor (S185) onboard a UAV platform, this paper examines the impact of spatial resolution on the classification of vegetation types in highly fragmented planting areas in southern China by aggregating 0.025 m hyperspectral image to relatively coarse spatial resolutions (0.05, 0.1, 0.25, 0.5, 1, 2.5 m). The object-based image analysis (OBIA) method was used and the effects of several segmentation scale parameters and different number of features were discussed. Finally, the classification accuracies from 84.3% to 91.3% were obtained successfully for multi-scale images. The results show that with the decrease of spatial resolution, the classification accuracies show a stable and slight fluctuation and then gradually decrease since the 0.5 m spatial resolution. The best classification accuracy does not occur in the original image, but at an intermediate level of resolution. The study also proves that the appropriate feature parameters vary at different scales. With the decrease of spatial resolution, the importance of vegetation index features has increased, and that of textural features shows an opposite trend; the appropriate segmentation scale has gradually decreased, and the appropriate number of features is 30 to 40. Therefore, it is of vital importance to select appropriate feature parameters for images in different scales so as to ensure the accuracy of classification. [ABSTRACT FROM AUTHOR]

The scope of this research lies in the combination of pre-trained Convolutional Neural Networks (CNNs) and Quantum Convolutional Neural Networks (QCNN) in application to Remote Sensing Image Scene Classification(RSISC). Deep learning (RL) is improving by leaps and bounds pretrained CNNs in Remote Sensing Image (RSI) analysis, and pre-trained CNNs have shown remarkable performance in remote sensing image scene classification (RSISC). Nonetheless, CNNs training require massive, annotated data as samples. When labeled samples are not sufficient, the most common solution is using pre-trained CNNs with a great deal of natural image datasets (e.g., ImageNet). However, these pre-trained CNNs require a large quantity of labelled data for training, which is often not feasible in RSISC, especially when the target RSIs have different imaging mechanisms from RGB natural images. In this paper, we proposed an improved hybrid classical-quantum transfer learning CNNs composed of classical and quantum elements to classify open-source RSI dataset. The classical part of the model is made up of a ResNet network which extracts useful features from RSI datasets. To further refine the network performance, a tensor quantum circuit is subsequently employed by tuning parameters on near-term quantum processors. We tested our models on the open-source RSI dataset. In our comparative study, we have concluded that the hybrid classical-quantum transferring CNN has achieved better performance than other pre-trained CNNs based RSISC methods with small training samples. Moreover, it has been proven that the proposed algorithm improves the classification accuracy while greatly decreasing the amount of model parameters and the sum of training data.