Search

Remote sensing (RS) cross-modal text–image retrieval has great application value in many fields such as military and civilian. Existing methods utilize the deep network to project the images and texts into a common space and measure the similarity. However, the majority of those methods only utilize the inter-modality information between different modalities, which ignores the rich semantic information within the specific modality. In addition, due to the complexity of the RS images, there exists a lot of interference relation information within the extracted representation from the original features. In this paper, we propose a jointly guided deep network for fine-grained cross-modal RS text–image retrieval. First, we capture the fine-grained semantic information within the specific modality and then guide the learning of another modality of representation, which can make full use of the intra- and inter-modality information. Second, to filter out the interference information within the representation extracted from the two modalities of data, we propose an interference filtration module based on the gated mechanism. According to our experimental results, significant improvements in terms of retrieval tasks can be achieved compared with state-of-the-art algorithms. The source code is available at https://github.com/CQULab/JGDN. [ABSTRACT FROM AUTHOR]

• Proposed a new phase-field modeling method based on spectral decomposition. • Verified the proposed method by loading tests of the segment and joint specimens. • Constructed a phase-field based stratum-structure model. • Successfully simulated the cracking feature and mechanical behavior of tunnel lining. Disturbed by the surrounding engineering construction, the shield tunnel lining may appear mixed-mode cracks such as segment cracking and joint crushing, which seriously damage its mechanical behaviors. To simulate the shield tunnel lining with mixed-mode cracks, a phase-field modeling method based on spectral decomposition is proposed, which unifies the critical energy release rate and crack-driving strain energy. The indoor full-scale tests are carried out to study the failure mechanism of the segment and joint. The phase-field modeling method is verified by the tests, and the influence of loading path on the cracking feature and mechanical behavior of the segment and joint are analyzed by the proposed numerical method. The test and numerical simulation indicate that the crack in the segment gradually develops from I-type crack (tensile crack) to Y-type crack (tensile-shear crack), and the joint is crushed due to local compression, and the crushed region expands gradually with the increase of the eccentricity. As the eccentricity decreases, the ultimate bearing capacity of the segment and joint is improved, while the ductility decreases. Then, the phase-field modeling method is extended to the stratum-structure model. As the surface load increases, the joints of the shield tunnel lining are crushed, and the top and bottom segments are cracked. Meanwhile, the bending moment is reduced due to the redistribution of internal forces caused by the segment cracking and joint crushing. [ABSTRACT FROM AUTHOR]