	US 12,493,752 B2
	Automatic concrete dam defect image description generation method based on graph attention network
Hua Zhou, Yunnan (CN); Fudong Chi, Yunnan (CN); Yingchi Mao, Yunnan (CN); Hao Chen, Yunnan (CN); Xu Wan, Yunnan (CN); Huan Zhao, Yunnan (CN); Bohui Pang, Yunnan (CN); Jiyuan Yu, Yunnan (CN); Rui Guo, Yunnan (CN); Guangyao Wu, Yunnan (CN); and Shunbo Wang, Yunnan (CN)
Assigned to Huaneng Lancang River Hydropower Inc, Kunming (CN); Hohai Univerdity, Nanjing (CN); and HUANENG GROUP R&D CENTER CO., LTD., Beijing (CN)
Filed by Huaneng Lancang River Hydropower Inc, Yunnan (CN); Hohai University, Jiangsu (CN); and HUANENG GROUP R&D CENTER CO., LTD., Beijing (CN)
Filed on Jun. 1, 2023, as Appl. No. 18/327,074.
Application 18/327,074 is a continuation of application No. PCT/CN2023/093220, filed on May 10, 2023.
Claims priority of application No. 202210664943.9 (CN), filed on Jun. 13, 2022.
Prior Publication US 2023/0401390 A1, Dec. 14, 2023
Int. Cl. G06F 40/40 (2020.01); G06T 7/00 (2017.01); G06V 10/42 (2022.01); G06V 10/774 (2022.01); G06V 10/82 (2022.01)

CPC G06F 40/40 (2020.01) [G06T 7/0002 (2013.01); G06V 10/42 (2022.01); G06V 10/774 (2022.01); G06V 10/82 (2022.01); G06T 2207/20081 (2013.01); G06T 2207/20084 (2013.01); G06T 2207/30184 (2013.01)]

10 Claims

1. An automatic concrete dam defect image description generation method based on graph attention network, characterized by including the following steps:

1) extracting global features and grid features of a defect image respectively by using a multi-layer convolutional neural network;

2) constructing a grid feature interaction graph, and inputting the global features and grid features as nodes;

3) updating and optimizing information of the nodes in the grid feature interaction graph constructed in Step 2) by using the graph attention network to get updated global features and grid features;

4) automatically generating an image description by using a sequence of the updated global features and grid features based on a Transformer decoding module.