Multiscale Feature Extraction Network for Real-time Semantic Segmentation of Road Scenes On the Autonomous Robot
文献类型: 外文期刊
作者: Xue, Junrui;Dai, Yingpeng;Wang, Yutan;Qu, Aili
作者机构:
关键词: Automation;autonomous robot;deep convolutional neural networks;real-time semantic segmentation
期刊名称: INTERNATIONAL JOURNAL OF CONTROL AUTOMATION AND SYSTEMS
ISSN: 1598-6446
年卷期: 2023 年
页码:
收录情况: SCIE(2023版) ; ; EI(2023版)
摘要: Semantic segmentation is an effective means for autonomous robots to understand the surrounding scenes. For autonomous robot, it requires the balance of accuracy and speed. Moreover, it is necessary to correctly extract environmental information in complex environments such as occlusion, poor illumination, and shadows condition. To solve above problems, a novel image-based Multi-scale Feature Extraction Network (MFENet) is designed for real-time semantic segmentation task. This network preserves different level features in the encoder and fuses those features to accurately segment each object. In addition, to enhance the representation ability, fusion module is introduced for information exchange between feature maps with different spatial resolution. Moreover, standard convolution is replaced by Multiscale Feature Extraction (MFE) module in intermediate layers, which could strengthen the feature extraction ability. On the Cityscapes dataset, MFENet achieves 72.4% Mean Intersection over Union (MIoU) with 8.0 million parameters at the speed of 30.5 FPS on a single GTX 1070Ti card. Finally, MFENet is deployed on an autonomous robot and tested in the real world. It produces good semantic segmentation results at the speed of 65.5 FPS. The results reveals the proposed MFENet could work well in real-world applications.
分类号:
- 相关文献
作者其他论文 更多>>
-
Lightweight multi-scale feature dense cascade neural network for scene understanding of intelligent autonomous platform
作者:Dai, Yingpeng;Meng, Lingfeng;Sun, Fushan;Wang, Songfeng
关键词:Lightweight neural network;Semantic segmentation;Classification;Autonomous platform
-
An ensemble multi-dimensional randomization network for intelligent recognition of tobacco baking stage
作者:Zhao, Panzhen;Wang, Songfeng;Hao, Xianwei;Wang, Zhisheng;Zou, Jun;Ren, Jie;Dai, Yingpeng
关键词:Image processing;Randomization network;Crop baking;Agricultural intelligent platform;Classification
-
Research on the Classification Method of Fresh Tobacco Leaf Maturity Based on Transfer Learning and Multi-Feature Fusion
作者:Zhao, Panzhen;Dai, Yingpeng
关键词:Transfer Learning;Feature Selection;Feature Fusion;Classification Model;Smart Agriculture
-
Lightweight multi-scale feature dense cascade neural network for scene understanding of intelligent autonomous platform
作者:Dai, Yingpeng;Meng, Lingfeng;Sun, Fushan;Wang, Songfeng
关键词:Lightweight neural network;Semantic segmentation;Classification;Autonomous platform
-
Design and implementation of flyback converter design and control of driving power supply for unmanned robot
作者:Meng, Lingfeng;Dai, Yingpeng;Wang, Songfeng;Zhuang, Keyu;Sun, Fushan;He, Dengfeng;Wang, Aihua
关键词:AC-DC power convertors;mobile robots
-
Efficient Dual-Branch Bottleneck Networks of Semantic Segmentation Based on CCD Camera
作者:Li, Jiehao;Dai, Yingpeng;Su, Xiaohang;Wu, Weibin
关键词:semantic segmentation;CCD camera;lightweight network;neural networks