摘要
在SSD(single shot multibox detector)模型基础上,提出一种采用多特征融合及递进池化技术的烟雾实时检测模型,用于对烟雾的实时检测,最终实现了火灾的前期预警.首先,采用MobileNet作为基础网络,实现对烟雾图像特征的逐层提取;然后利用递进池化技术实现对特征模型的压缩,通过反卷积操作实现关键特征的向前融合,避免关键特征的损失;最后经过1×1的卷积后对烟雾图像中不同类型的特征进行融合,借助SSD模型合并不同尺度特征的检测框,使模型目标框预测器统一,增强对模型正负样本的判断能力,实现对目标的类别和位置做出准确判断.实验结果表明,所改进的模型不仅能够对常规环境中的烟雾准确检测,而且对不同光照和尺度的烟雾图像检测也取得了较好的效果.
Aiming at the characteristics of variable signs and difficult to detect,on the basis of SSD(single shot multi-box detector)model,a real-time smoke detection model is proposed based on multi-feature fusion and progressive pooling technology,which can be used for real-time smoke detection and ultimately realize early warning of fire.MobileNet is used as the basic network to extract the smoke image features layer by layer.Then,the feature model is compressed by progressive pooling technology,and the key features are fused forward by deconvolution operation to avoid the loss of key features.Finally,after convolution of 1×1,the smoke image features of different types are fused,and SSD model is used to combine detection boxes with features of different scales,so as to unify the model target box predictor,enhance the ability to judge the positive and negative samples of the model,and realize the accurate judgment of the category and location of the target.Experimental results show that the improved model can not only accurately detect smoke in normal environment,but also achieve good results in smoke image detection with different illumination and scale.
作者
刘丽娟
陈松楠
LIU Lijuan;CHEN Songnan(College of Information Engineering,Xinyang Agriculture and Forestry University,Xinyang 464000,China;School of Technology,Beijing Forestry University,Beijing 100083,China)
出处
《信阳师范学院学报(自然科学版)》
CAS
北大核心
2020年第2期305-311,共7页
Journal of Xinyang Normal University(Natural Science Edition)
基金
国家自然科学基金项目(31570713)
河南省科技攻关项目(182102110160)
河南省重点研发与推广专项项目(182102210131)
信阳农林学院青年教师基金项目(201701013,2019LG014)
信阳农林学院青年骨干教师项目。
关键词
SSD模型
烟雾检测
多特征融合
数据增强
递进池化
SSD model
smoke detection
multiple features fusion
data augmentation
progressive pooling