首页|基于支持向量机的强噪声颗粒物图像识别

基于支持向量机的强噪声颗粒物图像识别

扫码查看
在观测等离子体中的颗粒物飞行轨迹时,等离子体的辐射光受到颗粒物的散射而产生较强的噪声干扰.针对这种情况,提出一种在强噪声干扰下对颗粒物图像进行识别的算法.利用 自适应滤波器和边缘检测对强噪声颗粒物图像进行预处理之后,进行霍夫圆变换提取颗粒物候选区域,支持向量机分类器以候选区域的灰度对比度和边缘强度为特征进行训练和测试.结果表明基于支持向量机的目标识别算法在强噪声干扰下对颗粒物图像进行识别是可行的,识别的准确率高达95%.
STRONG NOISE PARTICLE IMAGE RECOGNITION BASED ON SUPPORT VECTOR MACHINES
When observing the flight path of particles in plasma,the radiated light of plasma is scattered by particles,resulting in strong noise interference.In view of this situation,a particle image recognition algorithm under strong noise interference is proposed.After preprocessing the strong noise particle image with adaptive filter and edge detection,Hough circle transform was used to extract the candidate region of particle,and the support vector machine classifier was trained and tested based on the gray contrast and edge strength of the candidate region.The results show that the target recognition algorithm based on support vector machine is feasible under strong noise interference,and the recognition accuracy is as high as 95%.

Strong noiseParticle identificationHough circle transformationSupport vector machine

李冬冬、赵彦君、倪国华

展开 >

安徽大学物质科学与信息技术研究院 安徽合肥 230601

中国科学院等离子体物理研究所 安徽合肥 230031

强噪声 颗粒物识别 霍夫圆变换 支持向量机

国家自然科学基金项目国家重点研发计划重点项目

118752952019YFC0119000

2024

计算机应用与软件
上海市计算技术研究所 上海计算机软件技术开发中心

计算机应用与软件

CSTPCD北大核心
影响因子:0.615
ISSN:1000-386X
年,卷(期):2024.41(9)