摘要
提高电气设备紫外图像分割精确度对设备放电程度的准确评估具有重要意义。用紫外成像仪拍摄电气设备放电图像时,由于拍摄背景的复杂性,一般的图像分割方法并不能快速准确地分割紫外放电区域,因此提出一种结合显著性检测及改进大津算法的紫外图像分割模型。首先,对紫外图像进行显著性检测,使得故障区域突出,提升分割准确性;其次,利用基于Lévy飞行特征的蝙蝠算法对大津算法进行改进后对图像进行分割,以达到快速分割图像的目的。实验结果表明,改进的大津算法在紫外图像分割效果上明显优于大津算法,且计算速度也有所提升。
Improving the accuracy of UV image segmentation for electrical equipment is of great significance for the accurate evaluation of equipment discharge levels.When using a UV imager to capture electrical equipment discharge images,due to the complexity of the shooting background,general image segmentation methods cannot quickly and accurately segment the UV discharge area.Therefore,a UV image segmentation model combining saliency detection and improved Otsu algorithm is proposed.Firstly,perform saliency detection on the UV image to highlight the fault area and improve segmentation accuracy;secondly,the Bat Algorithm based on Lévy flight features is used to improve the Otsu algorithm and segment the image to achieve the goal of fast image segmentation.The experimental results show that the improved Otsu algorithm outperforms the Otsu algorithm in UV image segmentation,and the computational speed has also been improved.
作者
杨强强
陈思林
秦伦明
张贝贝
YANG Qiangqiang;CHEN Silin;QIN Lunming;ZHANG Beibei(College of Electronics and Information Engineering,Shanghai University of Electric Power,Shanghai 201306,China)
出处
《现代信息科技》
2023年第20期50-53,共4页
Modern Information Technology
