首页|基于改进DeeplabV3+的水面多类型漂浮物分割方法研究

基于改进DeeplabV3+的水面多类型漂浮物分割方法研究

扫码查看
原文链接
  • 国家科技期刊平台
  • NETL
  • NSTL
  • 万方数据
[目的]为解决传统图像处理方法鲁棒性差、常用深度学习检测方法无法准确识别大片漂浮物的边界等问题，[方法]提出一种基于改进DeeplabV3+的水面多类型漂浮物识别的语义分割方法，提高水面漂浮的识别能力。对所收集实际水面漂浮物进行分类，采用自制数据集进行对比试验。算法选择xception网络作为主干网络以获得初步漂浮物特征，在加强特征提取网络部分引入注意力机制以强调有效特征信息，在后处理阶段加入全连接条件随机场模型，将单个像素点的局部信息与全局语义信息融合。[结果]对比图像分割性能指标，改进后的算法mPA(Mean Pixel Accuracy)提升了 5。73％，mIOU(Mean Intersection Over Union)提升了 4。37％。[结论]相比于其他算法模型，改进后的Deep-labV3+算法对漂浮物特征的获取能力更强，同时能获得丰富的细节信息以更精准地识别多类型水面漂浮物的边界与较难分类的漂浮物，在对多个水库场景测试后满足实际水域环境中漂浮物检测的需求。
Research on segmentation method of multiple types of floating objects on water surface based on improved DeeplabV3+
[Objective]In order to solve the problems of poor robustness of traditional image processing method and the inability of commonly used deep learning detection method to accurately recognize the boundaries of large floating objects,[Methods]a se-mantic segmentation method based on the improvement of DeeplabV3+for the recognition of multiple types of water surface floats was proposed,which improves the recognition ability of water surface floats.By classifying the collected actual water surface floats into categories,a homemade dataset is used for comparison experiments.The algorithm selects the xception network as the backbone network to obtain preliminary float features,introduces the attention mechanism in the enhanced feature extraction net-work part to emphasize the effective feature information,and incorporates the fully-connected conditional random field model in the post-processing stage to fuse the local information of a single pixel with the global semantic information.[Results]Comparing the image segmentation performance metrics,the improved algorithm mPA(Mean Pixel Accuracy)is improved by 5.73％and mIOU(Mean Intersection Over Union)is improved by 4.37％.[Conclusion]Compared with other algorithmic models,the im-proved DeeplabV3+algorithm is more capable of acquiring features of floating objects,and at the same time obtains rich detail in-formation to more accurately identify the boundaries of multiple types of water surface floating objects and more difficult to clas-sify floating objects,which meets the needs of floating object detection in the actual water environment after testing on multi-ple reservoir scenarios.

deep learningsemantic segmentationfeature extractionfloating object identificationattention mechanismfull connection condition random fieldalgorithmic modelinfluence factors

包学才、刘飞燕、聂菊根、许小华、柯华盛

展开 >

南昌工程学院江西省水信息协同感知与智能处理重点实验室,江西南昌 330099

南昌工程学院信息工程学院,江西南昌 330099

江西省水利科学院,江西南昌 330029

深度学习 语义分割 特征提取 漂浮物识别 注意力机制 全连接条件随机场 算法模型 影响因素

国家自然科学基金江西省水利厅科技项目江西省科技厅重大科技研发专项"揭榜挂帅"制项目

61961026202223YBKT1920213AAG01012

2024

10.13928/j.cnki.wrahe.2024.04.015

水利水电技术(中英文)
水利部发展研究中心

水利水电技术(中英文)

CSTPCD北大核心
影响因子：0.456
ISSN：1000-0860
年,卷(期)：2024.55(4)
  • 35