摘要
海上运输业务的不断增多,使得航运领域的发展速度随之大幅度加快,图像识别技术在该领域中的应用范围大幅度拓宽。为在现有的基础上进一步提升图像识别的精度和效率,业内的专家学者进行大量的研究探索。与此同时,电子海图结构发生了一定程度的变化,传统的图像识别算法现已无法满足电子海图识别的需要。基于此,选取深度学习算法中的卷积神经网络,搭建识别模型,提出量化方法,增强识别的精度。通过量化处理之后,网络模型的性能得到显著提升,满足了舰船电子海图识别的要求。
The continuous increase of maritime transportation business has greatly accelerated the development of the shipping field,and the application range of image recognition technology in this field has been greatly expanded.In order to further improve the accuracy and efficiency of image recognition on the existing basis,experts and scholars in the industry have conducted a lot of research and exploration.At the same time,the structure of electronic charts has changed to a certain extent,and traditional image recognition algorithms can no longer meet the needs of electronic chart recognition.Based on this,the convolutional neural network in the deep learning algorithm is selected,the recognition model is built,and the quantization method is proposed to enhance the accuracy of recognition.After quantitative processing,the performance of the network model has been significantly improved,which meets the requirements of ship electronic chart recognition.
作者
曾海峰
ZENG Hai-feng(Guangdong Eco-engineering Polytechnic,Department of Information Engineering,Guangzhou 510520,China)
出处
《舰船科学技术》
北大核心
2021年第14期133-135,共3页
Ship Science and Technology
基金
广东省普通高校特色创新类项目(2019GKTSCX066)
广东省科技创新战略专项资金项目(pdjh2020a1102)
工业和信息化职业教育教学科研课题(GXHZW20201711)
关键词
深度学习
卷积神经网络
舰船
电子海图识别
deep learning
convolutional neural network
ship
electronic chart recognition
