Jianshen Li

To address the limitation of measurement accuracy caused by the structural complexity of threads, this paper proposes an improved Foerstner corner detection method based on adaptive gradient weighting (AGW-Foerstner), combined with corner-level dynamic calibration and a neural network error compensation strategy, to effectively enhance measurement precision. First, the covariance matrix is weighted by gradient magnitude to accurately locate crest and root points of the thread. Second, a local homography matrix is employed to dynamically calculate the pixel equivalent for each corner. Finally, by deeply integrating vision-based measurements with a neural network error compensation model, a residual prediction network is constructed, where key measured parameters such as major diameter and pitch diameter serve as inputs and true values as outputs, thereby achieving error correction. Experimental results validate the applicability of the proposed method across thread samples of different specifications, showing a significant reduction in measurement errors of critical parameters. This approach realized stable high-precision measurement and provides a practical solution for industrial thread inspection.

With the acceleration of China’s urbanization process, the problem of environmental pollution is becoming more and more serious and has attracted more and more public attention. In industrial activities such as thermal power generation, the incomplete combustion of fossil fuels such as coal releases sulfur dioxide (SO2), causing harm to ecosystems. Due to the uneven development between regions in China, heavy industries such as thermal power generation will exist for a long time, which will lead to a long-term process of sulfur dioxide control. Therefore, there is an urgent need to develop a simple, accurate, and reliable SO2 concentration detection equipment in order to conduct grid monitoring and implement SO2 source control throughout the country. In addition, given the vast territory of China and the significant difference between the north and south environments, the equipment also needs to have good adaptability. In this paper, variational mode decomposition (VMD) and BP neural network are used to optimize the detection data, which effectively reduces the influence of electronic noise and temperature drift by about 66% and 12%, respectively, and significantly improves the accuracy and reliability of the detection equipment.

Convolutional neural networks (CNNs) have achieved remarkable success in fully supervised medical image segmentation tasks. However, the acquisition of large quantities of homogeneous labeled data is challenging, making semi-supervised training methods that rely on a small amount of labeled data and pseudo-labels increasingly popular in recent years. Most existing semi-supervised learning methods, however, underestimate the importance of the unlabeled regions during training. This paper posits that these regions may contain crucial information for minimizing the model’s uncertainty prediction. To enhance the segmentation performance of the left-atrium database, this paper proposes a triple consistency segmentation network based on the squeeze-and-excitation mechanism (SETC-Net). Specifically, the paper constructs a symmetric architectural unit called SEConv, which adaptively recalibrates the feature responses in the channel direction by modeling the inter-channel correlations. This allows the network to adaptively weigh each channel according to the task’s needs, thereby emphasizing or suppressing different feature channels. Moreover, SETC-Net is composed of an encoder and three slightly different decoders, which convert the prediction discrepancies among the three decoders into unsupervised loss through a constructed iterative pseudo-labeling scheme, thus encouraging consistent and low-entropy predictions. This allows the model to gradually capture generalized features from these challenging unmarked regions. We evaluated the proposed SETC-Net on the public left-atrium (LA) database. The proposed method achieved an excellent Dice score of 91.14% using only 20% of the labeled data. The experiments demonstrate that the proposed SETC-Net outperforms seven current semi-supervised methods in left-atrium segmentation and is one of the best semi-supervised segmentation methods on the LA database.

Currently, although semi-supervised image segmentation has achieved significant success in many aspects, further improvement in segmentation accuracy is necessary for practical applications. Additionally, there are fewer networks specifically designed for segmenting 3D images compared to those for 2D images, and their performance is notably inferior. To enhance the efficiency of network training, various attention mechanisms have been integrated into network models. However, these networks have not effectively extracted all the useful spatial or channel information. Particularly for 3D medical images, which contain rich spatial and channel information with tightly interconnected relationships between them, there remains a wealth of spatial and channel-specific information waiting to be explored and utilized. This paper proposes a bidirectional and efficient attention parallel network (BEAP-Net). Specifically, we introduce two modules: Supreme Channel Attention (SCA) and Parallel Spatial Attention (PSA). These modules aim to extract more spatial and channel-specific feature information and effectively utilize it. We combine the principles of consistency training and entropy regularization to enable mutual learning among sub-models. We evaluate the proposed BEAP-Net on two public 3D medical datasets, LA and Pancreas. The network outperforms the current state of the art in eight algorithms and is better suited for 3D medical images. It achieves the new best semi-supervised segmentation performance on the LA database. Ablation studies further validate the effectiveness of each component of the proposed model. Moreover, the SCA and PSA modules proposed can be seamlessly integrated into other 3D medical image segmentation networks to yield significant performance gains.

Air is the environmental foundation for human life and production, and its composition changes are closely related to human activities. Sulfur dioxide (SO₂) is one of the main atmospheric pollutants, mainly derived from the combustion of fossil fuels. But SO₂ is a trace gas in the atmosphere, and its concentration may be less than one part per billion (ppb). This paper is based on the principle of photoluminescence and uses a photomultiplier tube (PMT) as a photoelectric converter to develop a device for real-time detection of SO₂ concentration in the atmosphere. This paper focuses on the impact of noise interference on weak electrical signals and uses wavelet transform to denoise the signals. At the same time, considering that the photoelectric system is susceptible to temperature changes, a multi parameter fitting model is constructed, and a BP neural network is used to further process the signal, separating the real data from the original data. Finally, a high-precision and wide-range trace level sulfur dioxide concentration detection device and algorithm were obtained.

The leakage point location of the water supply pipeline (WSP) can be estimated by the time difference of the leakage characteristic signals obtained from both ends of the pipeline. It is very important to extract the characteristic signal in the frequency domain to accurately locate the leakage point. The variational mode decomposition (VMD) has been proven to be effective in extracting leakage characteristic frequency domain signals, but it needs to set parameters in advance to adjust the signal processing effect. Therefore, it is proposed to initialize the center frequency required for the VMD with the frequency corresponding to the peak of the power spectrum of the original signal and then solve the preset problem of VMD decomposition layers by the recursive method. In addition, mutual information is used as a screening criterion for composite pattern functions to select appropriate pattern components to reconstruct signals. The actual leak location results show that the location errors of the proposed algorithm are only 0.20 and 1.66% for 23.3 and 38.9 m pipeline lengths, respectively. This shows that the proposed method can achieve the accurate location of the water leakage point under long measurement distances and complex interference.

The asymmetric loads on wind turbines are becoming a fatal constraint factor with the increase of wind turbine size. In order to tackle this challenge, individual pitch control mechanism is introduced into the wind turbines. However, the application scopeof the traditional individual pitch control (IPC) is limited due to the complex dynamic coupling among wind turbine components, the strong nonlinearity of the pitch system and the inevitable unmodeled dynamics. Therefore, a multivariable model-free adaptive control (MIMO-DE-MFAC) strategy with differential characteristic is first proposed in this paper. Then, the proposed method is applied to the individual pitch control of the 5MW wind turbine. The simulation results from FAST demonstrate that the MIMO-DE-MFAC achieves a notable performance improvement on blade root compared with the PI. Furthermore, compared with other existing model-free adaptive control (MFAC) strategies, the proposed method not only achieves the optimal control results, but also has stronger robustness in speed-varying wind fields. Finally, the hardware-in-the-loop (HIL) experiment further verifies the reliability of the above simulation results, which provides a useful reference value for the field test.