[1]薛凌峰,宋炜,鲍建腾,等.基于元学习的少样本水闸图像识别方法研究[J].江苏水利,2024,(03):20-24.
 XUE Lingfeng,SONG Wei,BAO Jianteng,et al.Research of few-sample sluice image recognition method based on meta-learning[J].JIANGSU WATER RESOURCES,2024,(03):20-24.
点击复制

基于元学习的少样本水闸图像识别方法研究()
分享到:

《江苏水利》[ISSN:1006-6977/CN:61-1281/TN]

卷:
期数:
2024年03期
页码:
20-24
栏目:
水利信息化
出版日期:
2024-03-15

文章信息/Info

Title:
Research of few-sample sluice image recognition method based on meta-learning
文章编号:
1007-7839(2024)03-0020-0005
作者:
薛凌峰1宋炜1鲍建腾1焦野1戚荣志2
(1. 江苏省水旱灾害防御调度指挥中心,江苏 南京 210029;2. 河海大学 计算机与软件学院,江苏 南京 211100)
Author(s):
XUE Lingfeng1 SONG Wei1 BAO Jianteng1 JIAO Ye1 QI Rongzhi2
(1. Flood and Drought Disaster Prevention and Control Center of Jiangsu Province, Nanjing 210029, China;2. College of Computer Science and Software Engineering, Hohai University, Nanjing 2111000, China)
关键词:
图像识别少样本元学习多头注意力
Keywords:
image recognition few-sample meta-learning multi-head attention
分类号:
TV211
文献标志码:
B
摘要:
针对实际工程环境中采集的水闸图像样本不均衡、前后景混融导致的识别效果不好的问题,提出一种基于元学习的少样本水闸图像识别方法。首先构建水闸图像数据集,并使用图像增强和预处理对数据集进行优化;再使用多头注意力,提升网络准确捕捉多种与任务相关的关键特征信息的能力,更好地与时序卷积协作,进一步提高水闸图像的识别效果。在构建的sluice-ImageNet数据集上进行实验,实验结果表明,相比其他方法,所提方法在水闸启闭状态图像识别任务上更具有效性和优越性。该方法部署于重点水利工程视频监测平台,辅助人工监管,可实现对水闸异常运行情况的实时监测,为防汛决策提供智能化支持。
Abstract:
Aiming at the problem of poor recognition effect caused by uneven samples of sluice gate images collected in the actual engineering environment and the mixing of front and rear views, a few-sample sluice image recognition method based on meta-learning was proposed. First, the sluice image dataset was constructed, and the dataset was optimized using image enhancement and preprocessing. Then, multi-head attention was used to improve the ability of the network to accurately capture a variety of key feature information related to the task, and better cooperate with the temporal convolution to further improve the recognition effect of the sluice image. Experiments were carried out on the constructed sluice-ImageNet dataset. The experimental results show that the proposed method is more effective and superior than other methods in the sluice opening and closing recognition task. The method was deployed on the video monitoring platform of key water conservancy projects to assist manual supervision and realize real-time monitoring of abnormal operation of sluices, and provide intelligent support for flood control decision-making.

参考文献/References:

[1]牟舵,刘斌. 大藤峡水利枢纽施工过程智能视频监控与应用[J]. 水利信息化,2021(2):67-69,90.
[2]杨凯,郭振霆,杨甜,等. 视频监控在偏远水利工程管理的应用[J]. 海河水利,2021(1):118-120,126.
[3]陈述,纪勤,陈云,等. 基于知识图谱的智慧水利研究进展[J]. 河海大学学报(自然科学版),2023,51(3):143-153.
[4]余再康,程井. 贵州省水库运行及安全监测监管系统关键技术[J]. 河海大学学报(自然科学版),2023,51(4):46-54.
[5]VILALTA R,DRISSI Y. A Perspective View and Survey of Meta-learning[J]. Artificial Intelligence Review,2002,18(2):77-95.

备注/Memo

备注/Memo:
收稿日期:2024-01-15
基金项目:江苏省水利科技项目(2018057)
作者简介:薛凌峰(1985—),男,工程师,主要从事防汛抗旱工作。E-mail: 84713687@qq.com
通信作者:戚荣志(1980—),男,副教授,博士,主要从事水利信息化、智能软件工程等工作。E-mail: rzqi@hhu.edu.cn
更新日期/Last Update: 2024-03-15