Electrical Thermal Image Semantic Segmentation: Large-Scale Dataset and Baseline

Wang, Futian; Guo, Yin; Li, Chenglong; Lu, Andong; Ding, Zhongfeng; Tang, Jin; Luo, Bin

doi:10.1109/tim.2022.3214605

Cited by 7 publications

(5 citation statements)

References 57 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…EC-CNN [11] designed a new network using the edge information of thermal image. CGNet [12] exposes a new benchmark data set for semantic segmentation of power thermal images, and proposes a cross-guide network based on it, which uses a cross guidance unit and a global information integration module in the network to improve segmentation effect of thermal infrared power image.…”

Section: Semantic Segmentation In Thermal Scenariosmentioning

confidence: 99%

“…Dataset: We evaluated the performance of the proposed infraredimage semantic segmentation method on LS-ETS [12] dataset. The LS-ETS dataset was created specifically for the semantic segmentation task of thermal infrared power equipment, it has a total of 4839 thermal infrared images, which contain a total of 18 categories.LS-ETS divides the data set into two parts according to category proportion, including 3226 training pictures and 1613 test pictures.…”

Section: Experimental Settingsmentioning

confidence: 99%

See 1 more Smart Citation

Thermal infrared image semantic segmentation with Iightweight edge assisted context guided network

Wang¹,

Zheng²,

Lv³

et al. 2023

Third International Conference on Machine Learning and Computer Application (ICMLCA 2022)

View full text Add to dashboard Cite

Section: Semantic Segmentation In Thermal Scenariosmentioning

confidence: 99%

Section: Experimental Settingsmentioning

confidence: 99%

Thermal infrared image semantic segmentation with Iightweight edge assisted context guided network

Wang¹,

Zheng²,

Lv³

et al. 2023

Third International Conference on Machine Learning and Computer Application (ICMLCA 2022)

View full text Add to dashboard Cite

“…A novel DE-ResUnet based on texture features and background knowledge is proposed by Wu et al [ 42 ] for brain tissue segmentation. Because thermal infrared information [ 43 ] is not affected by illumination changes and extreme weather, semantic segmentation using thermal images has attracted great attention. Maheswari et al [ 44 ] presented a top–down attention and gradient alignment-based graph neural network (AGAGNN) to discover the crucial semantic information.…”

Section: Related Workmentioning

confidence: 99%

Self-Enhanced Mixed Attention Network for Three-Modal Images Few-Shot Semantic Segmentation

Song,

Zhang,

Bao

et al. 2023

Sensors

View full text Add to dashboard Cite

As an important computer vision technique, image segmentation has been widely used in various tasks. However, in some extreme cases, the insufficient illumination would result in a great impact on the performance of the model. So more and more fully supervised methods use multi-modal images as their input. The dense annotated large datasets are difficult to obtain, but the few-shot methods still can have satisfactory results with few pixel-annotated samples. Therefore, we propose the Visible-Depth-Thermal (three-modal) images few-shot semantic segmentation method. It utilizes the homogeneous information of three-modal images and the complementary information of different modal images, which can improve the performance of few-shot segmentation tasks. We constructed a novel indoor dataset VDT-2048-5i for the three-modal images few-shot semantic segmentation task. We also proposed a Self-Enhanced Mixed Attention Network (SEMANet), which consists of a Self-Enhanced module (SE) and a Mixed Attention module (MA). The SE module amplifies the difference between the different kinds of features and strengthens the weak connection for the foreground features. The MA module fuses the three-modal feature to obtain a better feature. Compared with the most advanced methods before, our model improves mIoU by 3.8% and 3.3% in 1-shot and 5-shot settings, respectively, which achieves state-of-the-art performance. In the future, we will solve failure cases by obtaining more discriminative and robust feature representations, and explore achieving high performance with fewer parameters and computational costs.

show abstract

“…For instance, two-stage object detection methods such as Faster R-CNN [12] and one-stage methods like YOLO [13] and CenterNet [14] are commonly used to detect various substation or transmission equipment [15][16][17][18]. Segmentation methods such as DeepLab [19] and cross-guidance network [20] are also applied to segment equipment regions. However, as shown in Figure 2, these object detection methods producing upright bounding boxes may include irrelevant background regions when the equipment or part is slanted, leading to background interference.…”

Section: Electrical Equipment Localizationmentioning

confidence: 99%

Voltage-Induced Heating Defect Detection for Electrical Equipment in Thermal Images

Lin,

Li,

Sun

et al. 2023

Energies

View full text Add to dashboard Cite

Voltage-induced heating defect is a type of defect that may occur in transformation substation equipment. Although this type of defect is less common compared to current-induced heating defects, it is crucial to identify it due to its association with severe insulation degradation problems that require prompt intervention. However, the temperature variations caused by these defects may be relatively subtle, making it challenging to distinguish them in thermal images. In this work, considering the characteristics of voltage-induced heating defects and the scarcity of defect data, we propose a two-stage method for defect detection. In the first stage, we employ oriented R-CNN to detect oriented parts of the equipment, accurately localizing the centerline of each part. In the second stage, we extract the temperature distribution along the centerline of specific parts and discretize them as features. Subsequently, we train one-class support vector machines based on the features extracted from normal images for defect diagnosis. Experimental results demonstrate that the proposed method is capable of accurately detecting defects while maintaining a low false positive rate.

show abstract

Electrical Thermal Image Semantic Segmentation: Large-Scale Dataset and Baseline

Cited by 7 publications

References 57 publications

Thermal infrared image semantic segmentation with Iightweight edge assisted context guided network

Thermal infrared image semantic segmentation with Iightweight edge assisted context guided network

Self-Enhanced Mixed Attention Network for Three-Modal Images Few-Shot Semantic Segmentation

Voltage-Induced Heating Defect Detection for Electrical Equipment in Thermal Images

Contact Info

Product

Resources

About