| 面向驾驶场景的多尺度特征融合目标检测方法 |
| |
| 引用本文： | 黄仝宇,胡斌杰,朱婷婷,黄哲文. 面向驾驶场景的多尺度特征融合目标检测方法[J]. 计算机工程与应用, 2021, 57(14): 134-141. DOI: 10.3778/j.issn.1002-8331.2101-0430 |
| |
| 作者姓名： | 黄仝宇 胡斌杰 朱婷婷 黄哲文 |
| |
| 作者单位： | 1.华南理工大学 电子与信息学院,广州 5106402.广东白云学院 大数据与计算机学院,广州 5104503.广州市生发科技服务有限公司 技术部,广州 510308 |
| |
| 摘 要： | 针对驾驶场景中目标检测卷积神经网络模型检测精度较低的问题,提出一种基于改进RefineDet网络结构的多尺度特征融合目标检测方法.在RefineDet网络结构中嵌入LFIP(Light-weight Featurized Image Pyramid,轻量级特征化的图像金字塔)网络,将LFIP网络生成的多尺度特征图与Re...
|
| 关 键 词： | 深度学习 卷积神经网络 目标检测 RefineDet算法 感受野模块(RFB) 轻量级特征化的图像金字塔(LFIP) 参数化修正线性单元(PReLU) 损失函数 遮挡目标 |
Object Detection Method Based on Multi-scale Feature Fusion for Driving Scene |
| |
| HUANG Tongyu,HU Binjie,ZHU Tingting,HUANG Zhewen. Object Detection Method Based on Multi-scale Feature Fusion for Driving Scene[J]. Computer Engineering and Applications, 2021, 57(14): 134-141. DOI: 10.3778/j.issn.1002-8331.2101-0430 |
| |
| Authors: | HUANG Tongyu HU Binjie ZHU Tingting HUANG Zhewen |
| |
| Affiliation: | 1.School of Electronic and Information Engineering, South China University of Technology, Guangzhou 510640, China2.Faculty of?Mega?Data?and Computer Science, Guangdong Baiyun University, Guangzhou 510450, China3.Department of Technology, Guangzhou Shengfa Technology Service Co., Ltd., Guangzhou 510308, China |
| |
| Abstract: | Aiming at the problem of low detection accuracy of convolutional neural network model for object detection in driving vision, a multi-scale feature fusion object detection method based on improved RefineDet is proposed. Firstly, the LFIP（Light-weight Featured Image Pyramid） network is embedded in the RefineDet, and the multi-scale feature map generated by LFIP network is integrated with the main feature map output from ARM（Anchor Refinement Module） in the RefineDet, which improves the output effect of anchors preliminary classification and regression in the convolutional layer, and provides refined anchors frame for ODM（Object Detection Module） for further regression and multi-class prediction. Secondly, after the ODM in the RefineDet, a multi-branch structure RFB（Receptive Field Block） is embedded to obtain receptive fields of different scale in the detection task to improve the features extracted from the backbone network. Thirdly, the activation function in the model is replaced by the nonlinear activation function PReLU（Parametric Rectified Linear Unit） with learnable parameters to speed up the convergence of the model. Then, the Bounding box loss function of RefineDet is replaced by the Repulsion Loss function to narrow the gap between a proposal and its designated target and increase the distance between the proposal and the surrounding non-target objects. Finally, an object detection dataset is constructed with 48 260 images in driving vision, including 38 608 as training set and 9 652 as test set, which are verified on mainstream GPU hardware platform. The mAP of this method is 85.59%, which is better than RefineDet and other improved algorithms;the FPS is 41.7 frame/s, which meets the application requirements of driving scene object detection. Experimental results show that the proposed method can improve the accuracy of object detection in driving vision, and solve the problems of occlusion object detection and small object detectionin driving vision to a certain extent. |
| |
| Keywords: | deep learning convolutional neural network object detection RefineDet algorithm Receptive Field Block（RFB） Light-weight Featured Image Pyramid（LFIP） Parametric Rectified Linear Unit（PReLU） loss function occlusion object |
| 本文献已被 万方数据 等数据库收录！ |
| 点击此处可从《计算机工程与应用》浏览原始摘要信息 |
|
点击此处可从《计算机工程与应用》下载免费的PDF全文 |
|
