Guang Huo

INTRODUCTION: Transmission line defect detection is one of the most conventional and fundamental maintenance tasks in power grids. Its performance in terms of detection accuracy, efficiency, and privacy protection directly affects the operational safety of power systems. Federated learning is considered a secure architecture for cross regional defect detection due to its advantages in data privacy and collaborative training. OBJECTIVES: The current mainstream two-layer federated learning architecture is difficult to balance detection accuracy with computational and communication efficiency in the face of limited on-site resources in the power grid, and cannot meet the requirements of collaborative transmission line defect detection across multiple regions. To address these issues, this paper proposes a circuit defect detection method based on HFL (Hierarchical Federated Learning, HFL). METHODS: First, a three-layer federated learning architecture is introduced, where an additional edge layer is incorporated to reduce the computational burden and communication load on the central server. Then, a client dynamic allocation method based on two-factor clustering is proposed. By calculating similarity between client models, clients with approximately independent and identically distributed data are assigned to corresponding edge servers, thereby enhancing training efficiency and detection accuracy for both client and edge models. RESULTS: Compared to the two-layer federated learning model, the proposed algorithm reduces the computational cost and communication load by 66.67% and 66.86% respectively, while maintaining accuracy and exhibiting stronger generalization ability. CONCLUSION: The proposed HFL with dynamic client grouping effectively addresses the challenges of data privacy and resource constraints in power grid scenarios. It successfully balances detection accuracy with computational and communication efficiency, offering a practical and scalable solution for collaborative transmission line defect detection across multiple regions.

In identifying crop leaf diseases, Due to the complex nature of the disease symptoms. There may be variations in disease symptoms with similar characterist

The existing iris recognition methods offer excellent recognition performance for known classes, but they do not perform well when faced with unknown classes. The process of identifying unknown classes is referred to as open-set recognition. To improve the robustness of iris recognition system, this work integrates a hash center to construct a deep metric learning method for open-set iris recognition, called central similarity based deep hash. It first maps each iris category into defined hash centers using a generation hash center algorithm. Then, OiNet is trained to each iris texture to cluster around the corresponding hash center. For testing, cosine similarity is calculated for each pair of iris textures to estimate their similarity. Based on experiments conducted on public datasets, along with evaluations of performance within the dataset and across different datasets, our method demonstrates substantial performance advantages compared with other algorithms for open-set iris recognition.

Compared with traditional substations, the secondary circuit of smart substations is quite different in terms of transmission channel and protection function realization, which puts forward higher requirements for the knowledge level and business capabilities of smart substation operation and maintenance personnel. It is necessary to develop a set of teaching tools suitable for the training of secondary circuits in smart substations. This paper proposes a set of smart substation simulation secondary circuit simulation training system. The system builds the dynamic drawing model of the secondary schematic diagram, secondary wiring diagram, and secondary expansion diagram of the smart substation through visual modeling tools; with the help of the drawing monitoring module, the system can monitor the status of drawing elements in real time, dynamically calculate the input and output of drawings, and record the action sequence of elements; the system can build training cases based on typical protection action events and alarm events of smart substations. Thus, the system can be used to carry out drawing recognition training and defect tracking training for the secondary circuit of smart substations. This system enriches the content of the simulation training of smart substation secondary circuit, and has achieved good results in practical applications.

Iris segmentation is a critical step in the iris recognition system. Since the quality of iris database taken under different camera sensors varies greatly, thus most existing iris segmentation methods are designed for a particular collection device. Meanwhile, many iris segmentation methods based on convolutional neural networks (CNNs) require a lot of computational costs and hardware costs (storage space), which are not suitable for deploying on low‐performance devices. To address the above problems, an accurate and efficient heterogeneous iris segmentation network is proposed in this paper. First, the authors design an efficient feature extraction network, which combines depth‐wise separable convolution with traditional convolution to greatly reduce model parameters and computational cost while maintaining segmentation accuracy. Then, a Multi‐scale Context Information Extraction Module (MCIEM) is proposed to extract multi‐scale spatial information at a more granular level and enhance the discriminability of the iris region. Finally, a Multi‐layer Feature Information Fusion Module (MFIFM) is proposed to reduce the loss of information during the downsampling process. Experimental results on multi‐source heterogeneous iris database show that the proposed network can not only achieve state‐of‐the‐art performance but is also more efficient in terms of required parameters, calculated load, and storage space.

Iris segmentation is an important task in the iris recognition system. Traditional iris segmentation methods are susceptible to irrelevant noise. At present, segmentation methods based on convolutional neural networks (CNNs) have become the main trend in addressing the task. Although these networks achieve promising segmentation performance on the advanced graphics processor, they require a lot of computation and time, which are far from meeting the standards of real-time iris segmentation. To this end, a precise and fast iris segmentation algorithm is proposed. First, an efficient feature extraction network that combines depth-wise separable convolution with dilated convolution is designed to reduce model parameters while maintaining segmentation accuracy. Then, an attention mechanism is introduced to suppress noise interference and enhance the discriminability of the iris region. Finally, an auxiliary training branch is proposed to overcome the vanishing gradient problem. Experimental results show that the proposed method not only achieves state-of-the-art performance but also is more efficient in terms of required parameters, calculated load, and storage space. Specifically, the model parameter is only 0.1 million (M), and the average prediction time is <0.03 seconds (S).

Iris recognition is used in applications of security, national defense, e-commerce, and mobile devices due to its convenience, stability, uniqueness, better anticounterfeiting, etc.^1,2 The iris recognition system includes iris image acquisition, preprocessing, iris segmentation and localization, iris normalization, iris feature extraction, and iris matching verification or identification. In an accurate iris segmentation method, the iris region includes more discriminative features and gets a higher final recognition accuracy.³ With the development of image processing technology, biometric verification products based on the iris have been gradually deployed on low-power hardware. However, many challenges remain in this line of research. The factors that render iris segmentation on low-power devices a challenging task can be summarized as follows:

An accurate iris segmentation method is crucial for the iris recognition system. The conventional iris segmentation algorithms have poor adaptability when applied to heterogeneous iris databases. Therefore, researchers have applied deep learning to the field of iris segmentation. A modified U-Net is proposed to perform the iris segmentation for heterogeneous iris databases, referred to as DropBlock and modified shortcut branch U-Net. The main work is as follows: first, EfficientNetV2 based on DropBlock is used as convolutional blocks of U-Net to improve the ability of feature extraction and generalization of the network. Second, an improved shortcut branch structure is proposed for U-Net to reduce the loss of information during the downsampling process. The experimental results on the CASIA-iris-interval-v4, IITD, and UBIRIS.v2 iris databases demonstrate that this method can not only have good versatility but also provide higher accuracy on heterogeneous databases. Compared with some state-of-the-art fusion methods, the proposed method has a significant performance advantage.

Convolutional neural networks (CNN) have achieved unprecedented results in the fields of pattern recognition and image processing. CNN methods have also been gradually applied to iris recognition. However, in the case of insufficient training samples, training deep CNN models is prone to overfitting. In order to solve the above problem, this paper proposes an iris recognition method based on fine-tune SquIrisNet. When the pre-trained SqueezNet is migrated to the iris dataset, the fully connected layers are used instead of the convolutional layer and the global pooling layer, then the parameters are adjusted using the error back propagation algorithm, and finally the images are classified by the Softmax classifier. The fully connected layers added in this model can play the role of “firewall” in the fine-tune process, retaining the model complexity to a certain extent. The experimental results on the IIT Delhi and SDUMLA-HMT iris databases show that the proposed method has higher correct recognition rate, faster convergence speed and better robustness than AlexNet and VGGNet.

With the development of iris recognition technology, sensors of iris images acquisition are being constantly developed and updated. Re-register users every time a new sensor is deployed is time-consuming and complicated, especially in applications with large-scale registered users. Therefore, it is a challenging problem to choose the common recognition model which is effective for multi-source heterogeneous iris recognition(MSH-IR). The paper proposes a efficient neural network model of stacked Convolutional Deep Belief Networks-Deep Belief Network (CDBNs-DBN) for MSH-IR. The main improvements are two parts: firstly, this model uses the region-by-region extraction method and positions the convolution kernel through the offset of the hidden layer to locate the effective local texture feature structure. Secondly, the model uses DBN as a classifier in order to reduce the reconstruction error through the negative feedback mechanism of the auto-encoder. Experimental results have been implemented on publicly available IIT Delhi iris database, which is captured by three different iris captured sensors. Experiments shows the model performs strong robustness performance and recognition ability.