Substations play a crucial role in the proper operation of power systems. Online fault diagnosis of substation equipment is critical for improving the safety and intelligence of power systems. Detecting the target equipment from an infrared image of substation equipment constitutes a pivotal step in online fault diagnosis. To address the challenges of missed detection, false detection, and low detection accuracy in the infrared image object detection in substation equipment, this paper proposes an infrared image object detection algorithm for substation equipment based on an improved YOLOv8n. Firstly, the DCNC2f module is built by combining deformable convolution with the C2f module, and the C2f module in the backbone is replaced by the DCNC2f module to enhance the ability of the model to extract relevant equipment features. Subsequently, the multi-scale convolutional attention module is introduced to improve the ability of the model to capture multi-scale information and enhance detection accuracy. The experimental results on the infrared image dataset of the substation equipment demonstrate that the improved YOLOv8n model achieves mAP@0.5 and mAP@0.5:0.95 of 92.7% and 68.5%, respectively, representing a 2.6% and 3.9% improvement over the baseline model. The improved model significantly enhances object detection accuracy and exhibits superior performance in infrared image object detection in substation equipment.

Substation equipment defect detection is a critical aspect of ensuring the reliability and stability of modern power grids. However, existing deep-learning-based detection methods often face significant challenges in real-world deployment, primarily due to low detection accuracy and inconsistent anomaly definitions across different substation environments. To address these limitations, this paper proposes the Language-Guided Enhanced Anomaly Power Equipment Detection Network (LEAD-Net), a novel framework that leverages text-guided learning during training to significantly improve defect detection performance. Unlike traditional methods, LEAD-Net integrates textual descriptions of defects, such as historical maintenance records or inspection reports, as auxiliary guidance during training. A key innovation is the Language-Guided Anomaly Feature Enhancement Module (LAFEM), which refines channel attention using these text features. Crucially, LEAD-Net operates solely on image data during inference, ensuring practical applicability. Experiments on a real-world substation dataset, comprising 8307 image–text pairs and encompassing a diverse range of defect categories encountered in operational substation environments, demonstrate that LEAD-Net significantly outperforms state-of-the-art object detection methods (Faster R-CNN, YOLOv9, DETR, and Deformable DETR), achieving a mean Average Precision (mAP) of 79.51%. Ablation studies confirm the contributions of both LAFEM and the training-time text guidance. The results highlight the effectiveness and novelty of using training-time defect descriptions to enhance visual anomaly detection without requiring text input at inference.

Visible and infrared thermal imaging are crucial techniques for detecting structural and temperature anomalies in electrical power system equipment. To meet the demand for multimodal infrared/visible light monitoring of target devices, this paper introduces CBAM-YOLOv4, an improved lightweight object detection model, which leverages a novel synergistic integration of the Convolutional Block Attention Module (CBAM) with YOLOv4. The model employs MobileNet-v3 as the backbone to reduce parameter count, applies depthwise separable convolution to decrease computational complexity, and incorporates the CBAM module to enhance the extraction of critical optical features under complex backgrounds. Furthermore, a feature-level fusion strategy is adopted to integrate visible and infrared image information effectively. Validation on public datasets demonstrates that the proposed model achieves an 18.05 frames per second increase in detection speed over the baseline, a 1.61% improvement in mean average precision (mAP), and a 2 MB reduction in model size, substantially improving both detection accuracy and efficiency through this optimized integration in anomaly inspection of electrical equipment. Validation on a representative edge device, the NVIDIA Jetson Nano, confirms the model’s practical applicability. After INT8 quantization, the model achieves a real-time inference speed of 40.8 FPS with a high mAP of 80.91%, while consuming only 5.2 W of power. Compared to the standard YOLOv4, our model demonstrates a significant improvement in both processing efficiency and detection accuracy, offering a uniquely balanced and deployable solution for mobile inspection platforms.

Detecting defects in aerial images of grading rings collected by drones poses challenges due to the structural similarity between normal and defective samples. The small visual differences make it hard to distinguish defects and extract key features. Additionally, critical defect features often become lost during feature fusion. To address these issues, this paper uses YOLOv8 as the baseline model and proposes an improved YOLOv8-based method for detecting grading ring defects in transmission lines. Our approach first integrates the CloAttention and C2f modules into the feature extraction network, enhancing the model’s ability to capture and identify defect features in grading rings. Additionally, we incorporate CARAFE into the feature fusion network to replace the original upsampling module, effectively reducing the loss of critical defect information during the fusion process. Experimental results demonstrate that our method achieves an average detection accuracy of 67.6% for grading ring defects, marking a 6.8% improvement over the baseline model. This improvement significantly enhances the effectiveness of defect detection in transmission line grading rings.

Autoregressive probability models improve the rate–distortion performance of learned image compression, but their spatial conditioning introduces serial neural inference and high decoding latency. This paper presents a latency-oriented learned image compression framework that jointly addresses sequential entropy estimation, decoder computation, and dense intermediate activation responses. First, a dependency-decoupled entropy model estimates all latent distribution parameters in parallel from hyperprior statistics, eliminating the neural autoregressive loop. Second, a lightweight synthesis transform uses compact convolutional mappings, reduced channel widths, and a short progressive reconstruction path. Third, a differentiable activation-magnitude regularizer penalizes the normalized $L_1$ magnitude of intermediate decoder activations and is introduced gradually during rate–distortion training to encourage sparse feature responses. The activation regularizer is treated as a representation-level training constraint rather than as a direct estimator of dense-convolution arithmetic. Experiments on Kodak and BSD100 show that the proposed model consistently improves on the hyperprior baseline over the evaluated bitrate range. On Kodak, it achieves BD-rate reductions of 7.9% and 9.7% relative to hyperprior in terms of PSNR and MS-SSIM, respectively. At approximately 0.30 bpp, the proposed model achieves 32.94 dB PSNR and 0.9745 MS-SSIM with 8.10 M parameters, 165.3 GMACs, and a decoding time of $35.6\pm 1.0$ ms. Compared with the serial autoregressive joint baseline at a comparable operating point, this corresponds to a 95.0% reduction in decoding latency.

Vision–language models (VLMs) are often reported to outperform vision-only backbones for UAV power-line inspection. We examine that claim using ElecVQA-Bench, a 56,972-item resource derived from InsPLAD, and an audit of partition, evaluated items, label space, replication, input resolution, and side information. The original split has substantial source-photograph dependence and is used for paired diagnostic comparisons. Reported tower-disjoint retraining provides a limited check on new-tower performance within the same source dataset. On the original binary task, no difference is detected by a paired item-level test between the selected Swin and Qwen runs; this does not establish equivalence. Fine-grained results change with resolution, prompt information, class support, and seed. A 14-tower, three-seed comparison changes direction across seeds, and classes represented by only a few items substantially affect macro recall. The contribution is a reproducible task representation and reporting protocol for examining these dependencies, rather than a general ranking of model families or evidence of end-to-end UAV deployment.