In recent years, deep learning has been successfully applied to hyperspectral image classification (HSI) problems, with several convolutional neural network (CNN) based models achieving an appealing classification performance. However, due to the multi-band nature and the data redundancy of the hyperspectral data, the CNN model underperforms in such a continuous data domain. Thus, in this article, we propose an end-to-end transformer model entitled SAT Net that is appropriate for HSI classification and relies on the self-attention mechanism. The proposed model uses the spectral attention mechanism and the self-attention mechanism to extract the spectral–spatial features of the HSI image, respectively. Initially, the original HSI data are remapped into multiple vectors containing a series of planar 2D patches after passing through the spectral attention module. On each vector, we perform linear transformation compression to obtain the sequence vector length. During this process, we add the position–coding vector and the learnable–embedding vector to manage capturing the continuous spectrum relationship in the HSI at a long distance. Then, we employ several multiple multi-head self-attention modules to extract the image features and complete the proposed network with a residual network structure to solve the gradient dispersion and over-fitting problems. Finally, we employ a multilayer perceptron for the HSI classification. We evaluate SAT Net on three publicly available hyperspectral datasets and challenge our classification performance against five current classification methods employing several metrics, i.e., overall and average classification accuracy and Kappa coefficient. Our trials demonstrate that SAT Net attains a competitive classification highlighting that a Self-Attention Transformer network and is appealing for HSI classification.

Despite significant progress in object detection tasks, remote sensing image target detection is still challenging owing to complex backgrounds, large differences in target sizes, and uneven distribution of rotating objects. In this study, we consider model accuracy, inference speed, and detection of objects at any angle. We also propose a RepVGG-YOLO network using an improved RepVGG model as the backbone feature extraction network, which performs the initial feature extraction from the input image and considers network training accuracy and inference speed. We use an improved feature pyramid network (FPN) and path aggregation network (PANet) to reprocess feature output by the backbone network. The FPN and PANet module integrates feature maps of different layers, combines context information on multiple scales, accumulates multiple features, and strengthens feature information extraction. Finally, to maximize the detection accuracy of objects of all sizes, we use four target detection scales at the network output to enhance feature extraction from small remote sensing target pixels. To solve the angle problem of any object, we improved the loss function for classification using circular smooth label technology, turning the angle regression problem into a classification problem, and increasing the detection accuracy of objects at any angle. We conducted experiments on two public datasets, DOTA and HRSC2016. Our results show the proposed method performs better than previous methods.

In recent years, image classification on hyperspectral imagery utilizing deep learning algorithms has attained good results. Thus, spurred by that finding and to further improve the deep learning classification accuracy, we propose a multi-scale residual convolutional neural network model fused with an efficient channel attention network (MRA-NET) that is appropriate for hyperspectral image classification. The suggested technique comprises a multi-staged architecture, where initially the spectral information of the hyperspectral image is reduced into a two-dimensional tensor, utilizing a principal component analysis (PCA) scheme. Then, the constructed low-dimensional image is input to our proposed ECA-NET deep network, which exploits the advantages of its core components, i.e., multi-scale residual structure and attention mechanisms. We evaluate the performance of the proposed MRA-NET on three public available hyperspectral datasets and demonstrate that, overall, the classification accuracy of our method is 99.82 %, 99.81%, and 99.37, respectively, which is higher compared to the corresponding accuracy of current networks such as 3D convolutional neural network (CNN), three-dimensional residual convolution structure (RES-3D-CNN), and space–spectrum joint deep network (SSRN).

In this study, a fiber-optic Fabry–Perot (FP) high-temperature pressure sensor based on sapphire direct bonding is proposed and experimentally demonstrated. The sensor is fabricated by direct bonding of two-layer sapphire wafers, including a pressure diaphragm wafer and a cavity-etched wafer. The sensor is composed of a sensor head that contains a vacuum-sealed cavity arranged as an FP cavity and a multimode optical fiber. The external pressure can be measured by detecting the change in FP cavity length in the sensor. Experimental results demonstrate the sensing capabilities for pressures from 20 kPa to 700 kPa up to 800°C.

A intensity-modulated optical fiber relative humidity (RH) sensor based on the side coupling induction technology (SCIT) is presented and experimentally demonstrated. The agarose gel and the twisted macro-bend coupling structure are first combined for RH sensing applications. The refractive index (RI) of the agarose gel increases with the increase of the RH and is in linear proportion from 20 to 80%RH. The side coupling power, which changes directly with the RI of the agarose gel, can strip the source noise from the sensor signal and improve the signal to noise ratio substantially. The experiment results show that the sensitivity of the proposed sensor increases while the bend radius decreases. When the bend radius is 8 mm, the sensor has a linear response from 40% to 80% RH with the sensitivity of 4.23 nW/% and the limit of detection of 0.70%. A higher sensitivity of 12.49 nW/% is achieved when RH raises from 80% to 90% and the limit of detection decreases to 0.55%. Furthermore, the proposed sensor is a low-cost solution, offering advantages of good reversibility, fast response time, and compensable temperature dependence.

In the method of monitoring the liquid level based on ultrasonic impedance, the near-field effect can seriously affect the validity of the results. In this paper, we explore the factors affecting the length of the ultrasonic near field. Based on that, we propose the optimal length and the minimum length of the buffer block to avoid the near field. The evaluations show that when the parameters of the ultrasonic probe are 15 mm in diameter, 1 MHz in frequency, and ±15 V in emitted ultrasonic wave amplitude, the best results are obtained when the length of the buffer block is 22 mm. When the probe diameter is 10 mm, the buffer block length should be ≥5 mm to ensure the validity of the measured results. The evaluation precision is 1 mm. This research can effectively avoid the blind area of emitted waves when using ultrasonic to measure the liquid level. It provides an effective basis for the selection and design of ultrasonic probes.