摘要
针对烧结机机尾断面烧结状态识别分类效率低、成本高等问题,提出一种基于深度学习的烧结断面识别分类方法。首先,利用YCbCr颜色空间转换结合中值滤波的方法对断面图像预处理;然后,利用粒子群优化(PSO)算法优化的Canny算子进行边缘检测,得到较好的分割结果;最后,构建LetNet—5卷积神经网络模型对断面灰度图分类识别,实验结果表明:该方法具有较好的识别分类效果。
Aiming at the problems of low efficiency and high cost in recognition and classification of the sintering state of the tail section, a method for recognition and classification of sintering section based on deep learning is proposed.Firstly, the YCbCr color space conversion method and the median filtering are used to achieve cross section image preprocessing.And then, apply Canny operator optimized by particle swarm optimization(PSO) algorithm to perform edge detection and better segmentation effect is obtained.Finally, the LetNet-5 convolutional neural network(CNN)model is constructed for classification recognition of cross section grayscale image.Experimental results show that this method has better recognition and classification effect.
作者
阮志国
周敏
文喆皓
高强
RUAN Zhiguo;ZHOU Min;WEN Zhehao;GAO Qiang(Key Laboratory of Metallurgical Equipment and Control Technology,Ministry of Education,Wuhan University of Science and Technology,Wuhan 430081,China;Hubei Key Laboratory of Mechanical Transmission and Manufacturing Engineering,Wuhan University of Science and Technology,Wuhan 430081,China)
出处
《传感器与微系统》
CSCD
北大核心
2022年第12期51-54,共4页
Transducer and Microsystem Technologies
基金
国家自然科学基金面上资助项目(51975431)。
