基于改进Faster RCNN的毫米波图像实时目标检测  

Real-Time Object Detection for Millimeter-Wave Images Based on Improved Faster Regions with Convolutional Neural Networks

作　　者：侯冰基 杨明辉[1] 孙晓玮[1] HouBingji;Yang Minghui;Sun Xiaowei(Key Laboratory of Terahertz Solid Technology, Shanghai Institute of Microsystem and Inform ationTechnology (SIMIT ), Chinese Academy of Sciences, Shanghai 200050, China;University of Chinese Academy of Sciences, Beijing 100049, China;School of Inform ation Science and Technology, Shanghai Tech University, Shanghai 201210, China)

机构地区：[1]中国科学院上海微系统与信息技术研究所太赫兹固态技术重点实验室,上海200050 [2]中国科学院大学,北京100049 [3]上海科技大学信息科学与技术学院,上海201210

出　　处：《激光与光电子学进展》2019年第13期119-125,共7页Laser & Optoelectronics Progress

基　　金：国家自然科学基金(61731021,61671439)

摘　　要：采用反卷积与捷径连接,针对毫米波图像提出了一种高效、快速的卷积神经网络,在保留图像低阶细粒度特征的同时,检测速度由原框架的9frame/s大幅提升至27frame/s,并取消了Faster RCNN (Regions with Convolutional Neural Networks)中的RCNN部分。为了使网络更好地收敛,基于聚类思想设计了初始候选框的大小。使用在线困难样本挖掘(OHEM)优化了Faster RCNN的损失函数,解决了毫米波图像中正负样本失衡的问题,大幅提升了训练速度。所提算法在测试集上取得了87.6%的准确率和81.2%的检出率,F1分数相较于主流算法提升了5%左右。An efficient and fast convolution neural network for millimeter-wave images that uses deconvolution and a shortcut connection is proposed. The proposed network retains the low-order fine-grained features of the image and significantly improves the detection speed to 27 frame/s from 9 frame/s of original frame. The RCNN (Regions with Convolutional Neural Networks) part of the Faster RCNN is removed. To achieve better network convergence, the initial candidate box size is designed based on thought clustering. The online hard example mining is applied to optimize the loss function of the Faster RCNN such that the imbalance problem between positive and negative samples in millimeter wave images is solved and the training speed is improved significantly. By using the proposed algorithm, the accuracy of 87. 6 % and the detection rate of 81.2 % are obtained on the test set. Compared with mainstream algorithms, the proposed algorithm improves the Fi score by approximately 5%.

关 键 词：图像处理 图像识别 卷积神经网络 反卷积 毫米波图像 目标检测 

分 类 号：TP751.2[自动化与计算机技术—检测技术与自动化装置]