摘要
文中采用红外摄像头拍摄不同情况下货厢内部的图片作为样本,提出一种HOG特征进行人形分类特征提取并进行降维处理的方法,选择常用于可见光识别领域的支持向量机作为优化分类模型,结合粒子群优化算法寻找支持向量机的核函数的核参数。最终成功实现了对简易升降机违规载人行为的监测,识别正确率达到了97%。
In this paper,taking the pictures of the interior of the cargo compartment taken by infrared cameras under different conditions as samples,a method of HOG feature extraction and dimensionality reduction is proposed.Support vector machine,which is commonly used in the field of visible light recognition,was selected as the optimal classification model,and the kernel parameters of the kernel function of support vector machine were found by combining the particle swarm optimization algorithm.Finally,the monitoring of manned operations against rules of the simple lift was realized,with a correct recognition rate of 97%.
作者
李贤
王国贤
朱春东
程茁
Li Xian;Wang Guoxian;Zhu Chundong;Cheng Zhuo
出处
《起重运输机械》
2024年第14期67-73,共7页
Hoisting and Conveying Machinery
关键词
简易升降机
红外识别
HOG特征
SVM
粒子群优化算法
simple lift
infrared identification
HOG features
SVM
particle swarm optimization algorithm
