The dependence of electronic products on printed circuit boards (PCBs) is increasing. In recent years, PCB defect detection technology has achieved considerable results, but the defect targets are small, the background is complex, and there are many integrated components, which poses great challenges to product quality control. Therefore, this paper proposes a printed circuit board defect detection network that integrates coordinate attention and multi-level gradient flow optimization, called EP-YOLO, to improve the detection accuracy of printed circuit board defects. Firstly, this paper improves the model's ability to capture small target details by reconstructing the detection head of the small target. Based on this, a Shallow Context Feature Extraction (SCFE) network is designed to fuse shallow features with multi-scale features, effectively preventing the loss of shallow detail information and texture features. At the same time, in order to suppress background noise, this paper designs a multi-level feature gradient flow optimization module (C2f_CA, abbreviated as CCA) that integrates coordinate attention and a Cross-Stage Partial Frequency-domain Omni-Kernel module (CSPFOK) to enhance feature extraction capability. Finally, the SCYLLA-IoU (SIoU) optimization model training process was introduced. The experimental results showed that EP-YOLO achieved 97.5% mAP50 on the PKU-Market-PCB dataset and 98.5% mAP50 on the DeepPCB dataset, with a parameter reduction of approximately 12.55%, outperforming popular object detection networks. The results highlight the potential capabilities of EP-YOLO, providing a powerful and effective solution for industrial PCB defect detection.

In direct current, there are not only DC signals but also ripple, distortion, and noise signals. In addition to avoiding various errors from the hardware, it is also necessary to compensate for errors in data processing. The algorithm is used to process the data collected by the signal acquisition module of the device to remove the interference in the original signal and reduce the signal error.

When faced with complex scenes containing strong edge contours and noise, there are still more background residuals in the detection results of traditional algorithms, leading to a high false alarm rate. To solve the above problems, we propose an infrared dim and small target detection method that combines local feature prior and tensor train nuclear norm(TTNN). To suppress the strong edge contours, we first establish a background edge contour suppression function based on the structure tensor. Secondly, we propose a multi-frame density peak search algorithm to obtain local feature information by combining the features associated with multiple contiguous frames of the target. Then we use the local feature information and reweighting strategy to constrain the sparse components of the target signal, and describe the background low-rank components by the tensor train nuclear norm. Finally, we separate the target image from the background image using the alternating direction multiplier method. As compared with eight advanced algorithms, the method in this paper has a better background strong edge contour and a stronger noise suppression.

To address the problem that traditional deep learning algorithms cannot fully utilize the correlation properties between spectral sequence information and the feature differences between different spectra, this paper proposes a parallel network architecture land-use classification based on a combined multi-head attention mechanism and multiscale residual cascade called MARC-Net. This parallel framework is firstly implemented by deeply mining the features generated by grouped spectral embedding for information among spectral features by adding a multi-head attention mechanism, which allows semantic features to have expressions from more subspaces while fully considering all spatial location interrelationships. Secondly, a multiscale residual cascade CNN (convolutional neural network) is designed to fully utilize the fused feature information at different scales, thereby improving the network’s ability to represent different levels of information. Lastly, the features obtained by the multi-head attention mechanism are fused with those obtained by the CNN, and the merged resultant features are downgraded through the fully connected layer to obtain the classification results and achieve pixel-level multispectral image classification. The findings show that the algorithm proposed in this paper has an aggregate precision of 97.22%, compared to that of the Vision Transformer (ViT) with 95.08%; its performance on the Sentinel-2 dataset shows a huge improvement. Moreover, this article mainly focuses on the change rate of forest land in the study area. The Forest land area was 125.1143 km2 in 2017, 105.6089 km2 in 2019, and 76.3699 km2 in 2021, with an increase of 15.59%, an decrease of 0.97%, and increase of 14.76% in 2017–2019, 2019–2021 and 2017–2021, respectively.

Rice is a primary food consumed daily by many people, and different samples of rice often show disparate quality levels due to different production environments. In the rice market, it is common to sell low-quality rice with high-quality origin labels. As a nondestructive testing technology, spectral analysis has been widely used in food quality supervision. In this work, a deep learning method was developed and combined with a hyperspectral imaging system to achieve a quality-based identification of rice samples from different origins. First, the hyperspectral system was used to obtain spectral information of rice samples from five different origins. Then, a multi-kernel channel attention (MKCA) was proposed to focus on the deep features of the spectral information. Finally, based on the classical deep learning network, combined with MKCA, the spectral characteristics of rice samples from different origins were effectively identified. The results showed that MKCA combined with the LeNet-5 network structure achieved 97.40% accuracy, 97.63% precision, 97.78% recall, and 97.70% F1-score. It provides an effective technical method for tracing rice.

Focusing on the MH-BCS-SPL framework, aiming at the problem that the theoretical conditions of the multi-hypothesis prediction algorithm can not be satisfied with high probability in the case of low sampling rate, resulting in poor reconstruction quality, a hybrid multi-hypothesis prediction reconstruction algorithm based on measurement value reorganization is proposed. On the basis of this algorithm, in order to prevent the negative impact of inaccurate edge information on the reconstruction quality, a mixed multiple hypothesis prediction reconstruction algorithm based on resampling is proposed. By measuring the similarity between the resampling measured value and the original measured value, multiple initial reconstruction results are adaptively mixed to obtain the final reconstruction results. Experimental results show that the algorithm can effectively improve the reconstruction quality under low sampling rate.

To effectively solve the problems that most convolutional neural networks cannot be applied to the pixelwise input in remote sensing (RS) classification and cannot adequately represent the spectral sequence information, we propose a new multispectral RS image classification framework called HyFormer based on Transformer. First, a network framework combining a fully connected layer (FC) and convolutional neural network (CNN) is designed, and the 1D pixelwise spectral sequences obtained from the fully connected layers are reshaped into a 3D spectral feature matrix for the input of CNN, which enhances the dimensionality of the features through FC as well as increasing the feature expressiveness, and can solve the problem that 2D CNN cannot achieve pixel-level classification. Secondly, the features of the three levels of CNN are extracted and combined with the linearly transformed spectral information to enhance the information expression capability, and also used as the input of the transformer encoder to improve the features of CNN using the powerful global modelling capability of the Transformer, and finally the skip connection of the adjacent encoders to enhance the fusion between different levels of information. The pixel classification results are obtained by MLP Head. In this paper, we mainly focus on the feature distribution in the eastern part of Changxing County and the central part of Nanxun District, Zhejiang Province, and conduct experiments based on Sentinel-2 multispectral RS images. The experimental results show that the overall accuracy of HyFormer for the study area classification in Changxing County is 95.37% and that of Transformer (ViT) is 94.15%. The experimental results show that the overall accuracy of HyFormer for the study area classification in Nanxun District is 95.4% and that of Transformer (ViT) is 94.69%, and the performance of HyFormer on the Sentinel-2 dataset is better than that of the Transformer.

It is difficult for traditional algorithms to remove cloud edge contours in multi-cloud scenarios. In order to improve the detection ability of dim and small targets in complex edge contour scenes, this paper proposes a new dim and small target detection algorithm based on local multi-directional gradient information energy perception. Herein, based on the information difference between the target area and the background area in the four direction neighborhood blocks, an energy enhancement model for multi-directional gray aggregation (EMDGA) is constructed to preliminarily enhance the target signal. Subsequently, a local multi-directional gradient reciprocal background suppression model (LMDGR) was constructed to model the background of the image. Furthermore, this paper proposes a multi-directional gradient scale segmentation model (MDGSS) to obtain candidate target points and then combines the proposed multi-frame energy-sensing (MFESD) detection algorithm to extract the true targets from sequence images. Finally, in order to better illustrate the effect of the algorithm proposed in this paper in detecting small targets in a cloudy background, four sequence images are selected for detection. The experimental results show that the proposed algorithm can effectively suppress the edge contour of complex clouds compared with the traditional algorithm. When the false alarm rate Pf is 0.005%, the detection rate Pd is greater than 95%.

In this paper, we propose a new filtering switch with excellent working performance made using an optimized coupled microstrip line. Upon analyzing the RF (radio frequency) front-end's system structure, the switching device was simplified to a diode, which was connected to the microstrip circuit we designed to become a filter switch with both filtering and shutdown functions. First, we obtained an equivalent schematic of this filtering switch based on the relevant microstrip line theory. This switch consists of two coupled microstrip circuits, parallel-coupled feed lines and coupled-line stub-load resonators (CLSs), and a PIN diode. Second, the operating principle is described by the switching of the operating states, with ideal shutdown performance in the off state and considerable selectivity and excellent out-of-band rejection performance in the filtered state. Finally, a prototype filtering switch with a center frequency of 0.8 GHz was designed and tested. After subsequent optimization and improvement, the simulation and test performance results were noticeably consistent, consequently verifying the performance requirements of this filtering switch in two operating states in the center frequency band.

With the development of intelligent transportation system, the detection method of traffic signs plays an important role in unmanned driving. However, due to the real-time and reliability characteristics of the automatic driving system, each traffic sign needs to be processed in a specific time interval to ensure the precision of the test results. Automatic driving is developing rapidly and has made great progress. Various traffic sign detection algorithms are proposed. Especially, convolutional neural network algorithm is concerned because of its fast execution and high recognition rate. But in the real world of complex traffic conditions, those algorithms still have problems such as poor real-time detection, low precision, false detection and high missed detection rate. To overcome those problems, this paper proposed an improved algorithm named as YOLO-RFB based on YOLO V4 network. Based on YOLO V4 network, the main feature extraction network is pruned, and convolution layer is replaced by RFB structure in two output feature layers. In the detection results of GTSDB data sets, the mAP of improved algorithm achieves 85.59

For the multi hidden layer LSTM recurrent neural network, the updating of its weights and thresholds depends on the gradient descent algorithm, the convergence speed of the model is slow, and the weight calculation of the network nodes is prone to local extremum, which leads to the global optimization of the LSTM neural network model and the decline of the generalization ability of the network model, which limits the application of LSTM recurrent neural network. Therefore, this paper proposes an improved LSTM neural network model based on APSO algorithm. In this model, the root mean square error is designed as an appropriate value function, and the APSO algorithm is used to construct an optimization strategy to globally optimize the weights of each neuron node, so as to improve the generalization and prediction performance of the model. The experimental results of classic data market and UCI data sets show that the prediction accuracy of APSO-LSTM model is significantly improved compared with the traditional LSTM model, which verifies the effectiveness and practicability of APSO-LSTM model.

With the development of intelligent transportation systems (ITSs), traffic sign detection and recognition methods play a vital role in unmanned vehicle driving (UVD) and traffic flow forecasting. However, due to the real-time and reliability features in traffic systems, each traffic sign should be handled in a specified interval to ensure the accuracy of the detection and recognition results. To this point, images captured by the vehicle’s image sensors recognize traffic signs on the road to help drivers correctly operate their vehicles and even drive themselves. In this research area, the first step is to locate the area of traffic signs as the detection range, and the second step is to recognize the types of traffic signs detected. This paper presents a survey of the research results of scholars over the years, analyses the research results and presents the findings in summarized categories. Then, the prospects of future research regarding intelligent system implementations for traffic sign detection and recognition are elaborated. We hope this summary will lead to a boom in more original research on our smart city.

The rapid development and application of AI in intelligent transportation systems has widely impacted daily life. The application of an intelligent visual aid for traffic sign information recognition can provide assistance and even control vehicles to ensure safe driving. The field of autonomous driving is booming, and great progress has been made. Many traffic sign recognition algorithms based on convolutional neural networks (CNNs) have been proposed because of the fast execution and high recognition rate of CNNs. However, this work addresses a challenging question in the autonomous driving field: how can traffic signs be recognized in real time and accurately? The proposed method designs an improved VGG convolutional neural network and has significantly superior performance compared with existing schemes. First, some redundant convolutional layers are removed efficiently from the VGG-16 network, and the number of parameters is greatly reduced to further optimize the overall architecture and accelerate calculation. Furthermore, the BN (batch normalization) layer and GAP (global average pooling) layer are added to the network to improve the accuracy without increasing the number of parameters. The proposed method needs only 1.15 M when using the improved VGG-16 network. Finally, extensive experiments on the German Traffic Sign Recognition Benchmark (GTSRB) Dataset are performed to evaluate our proposed scheme. Compared with traditional methods, our scheme significantly improves recognition accuracy while maintaining good real-time performance.