No abstract

In recent years, the success of deep learning in natural scene image processing boosted its application in the analysis of remote sensing images. In this paper, we applied Convolutional Neural Networks (CNN) on the semantic segmentation of remote sensing images. We improve the Encoder- Decoder CNN structure SegNet with index pooling and U-net to make them suitable for multi-targets semantic segmentation of remote sensing images. The results show that these two models have their own advantages and disadvantages on the segmentation of different objects. In addition, we propose an integrated algorithm that integrates these two models. Experimental results show that the presented integrated algorithm can exploite the advantages of both the models for multi-target segmentation and achieve a better segmentation compared to these two models.

No abstract

No abstract

With the number of long-distance pipelines increasing in China,the excavation inspection and maintenance is the most important means to avoid the leakage of pipeline. However, the current excavation and maintenance decision-making models are absent to make satisfying results based on the comprehensive data of pipeline. To address this problem, a new quantitative risk-based model is proposed to guide decision-making on excavation inspection and maintenance. Based on previous failure cases, the model includes data about the surrounding soils as well as about the pipeline’s protective layer, cathodic protection and thickness readings. Case of the proposed model on previous failure cases shows that the new model can correctly predict a rational excavation inspection and maintenance span for a long-distance pipeline during its whole service life.