摘要
为了增强图像分割技术的准确性并优化图像分割技术的细节分割效果,提出了一种基于属性加权朴素贝叶斯算法的OTSU图像分割方法。将OTSU算法中依据图像灰度特征选取的图像中的前景和背景通过属性加权朴素贝叶斯算法进行分类处理,计算图像中前景和背景的概率,训练该模型以获得最佳阈值进行图像分割处理,优化图像分割的效果。利用无人机航拍采集的图像数据进行实验,结果显示基于属性加权朴素贝叶斯算法的OTSU图像分割方法优化了图像的分割效果,较完整地展示了分割后的图像细节,具有较好的应用价值。
In order to enhance the accuracy of image segmentation and optimize the detail segmentation effect of image segmentation,an improved OTSU image segmentation method based on attribute weighted naive Bayesian algorithm is proposed.The foreground and background of an image are selected according to the grayscale characteristics of the image as using OTSU algorithm,and then classified by using the attribute Weighted Naive Bayesian algorithm.Thus,the probability of the foreground and background of the image is calculated.By training this model to obtain the optimal threshold for the image segmentation process,the optimized effect of image segmentation can be obtained.Experiment with image data collected by drone aerial photography is conducted,and results show that the image segmentation of OTSU based on the attribute weighted naive Bayesian algorithm optimizes the image segmentation effect and shows much finer details of the image after segmentation,promising a prospective application value.
作者
马飞虎
曾聪
金依辰
孙翠羽
陈华鹏
MA Feihu;ZENG Cong;JIN Yichen;SUN Cuiyu;CHEN Huapeng(School of Civil Engineering and Architecture,East China Jiaotong University,Nanchang 330013,Jiangxi,China)
出处
《应用科学学报》
CAS
CSCD
北大核心
2022年第2期224-232,共9页
Journal of Applied Sciences
基金
国家重点研发计划项目(No.2021YFE0105600)
国家自然科学基金面上项目(No.51978263)
江西省自然科学基金重点项目(No.20192ACBL20008)资助
作者简介
通信作者:马飞虎,副教授,研究方向为3S技术集成、智能交通、工程测量等。E-mail:mfh3@163.com
