本期推荐IEEE/CAA Journal of Automatica Sinica（自动化学报英文版）和自动化学报近年发表的目标检测相关论文。

Towards Collaborative Robotics in Top View Surveillance: A Framework for Multiple Object Tracking by Detection Using Deep Learning

In order to show the importance of top view surveillance, a collaborative robotics framework has been presented. It can assist in the detection and tracking of multiple objects in top view surveillance. The framework consists of a smart robotic camera embedded with the visual processing unit. 

I. Ahmed, S. Din, G. Jeon, F. Piccialli, and G. Fortino, "Towards Collaborative Robotics in Top View Surveillance: A Framework for Multiple Object Tracking by Detection Using Deep Learning," IEEE/CAA J. Autom. Sinica, vol. 8, no. 7, pp. 1253-1270, July 2021.

Automated Silicon-Substrate Ultra-Microtome for Automating the Collection of Brain Sections in Array Tomography

This paper presents the design of an automatic collection device for brain sections. First, a novel mechanism based on circular silicon substrates is proposed for collection of brain sections; second, an automatic collection system based on microscopic object detection and feedback control strategy is proposed. Experimental results verify the function of the proposed collection device. 

L. Cheng, W. Z. Liu, C. Zhou, Y. X. Zou, and Z.-G. Hou, "Automated Silicon-Substrate Ultra-Microtome for Automating the Collection of Brain Sections in Array Tomography," IEEE/CAA J. Autom. Sinica, vol. 8, no. 2, pp. 389-401, Feb. 2021.

Speed and Accuracy Tradeoff for LiDAR Data Based Road Boundary Detection

A speed and accuracy tradeoff method for LiDAR-based road boundary detection in structured environments is proposed. The proposed method consists of three main stages: 1) a multi-feature based method is applied to extract feature points; 2) a road-segmentation-line-based method is proposed for classifying left and right feature points; 3) an iterative Gaussian Process Regression (GPR) is employed for filtering out false points and extracting boundary points.

G. J. Wang, J. Wu, R. He, and B. Tian, "Speed and Accuracy Tradeoff for LiDAR Data Based Road Boundary Detection," IEEE/CAA J. Autom. Sinica, vol. 8, no. 6, pp. 1210-1220, June 2021.

光学遥感图像目标检测算法综述

对光学遥感图像中通用目标检测方法进行全面、细致的梳理, 既包括早期基于手工设计特征的方法, 也包括当前深度学习的方法, 同时重点对基于深度学习的方法及其针对遥感图像特点的改进进行更为精细的梳理, 此外还额外评估各个改进方法的性能定量改善, 并根据评估结果指出现存的问题和未来的发展方向.

聂光涛, 黄华. 光学遥感图像目标检测算法综述. 自动化学报, 2021, 47(8): 1749−1768

智能人机交互中第一视角手势表达的一次性学习分类识别

提出了一种基于深度神经网络级联组合解决复杂应用场景中第一视角下的一次性学习手势识别的算法. 首先, 借助采集的手势目标样本对改进的轻量级SSD网络进行迁移式强化训练. 接着, 在包含手势目标的图像中, 利用改进的U-Net模型对复杂背景下的手势目标实施高效精准分割, 以降低无关目标对手势识别结果的影响. 在此基础上,借助端到端2D关系网络并将其扩展为处理视频序列输入的3D关系网络, 同时采用3D残差卷积神经网络作为视频数据的特征提取模块. 在对相关类别的大样本数据集进行深度训练的基础上, 使用预训练模型初始化目标网络参数, 提升网络的学习能力, 减少过拟合的风险并加速网络收敛. 

鹿智, 秦世引, 李连伟, 张鼎豪. 智能人机交互中第一视角手势表达的一次性学习分类识别. 自动化学报, 2021, 47(6): 1284−1301

目标检测模型及其优化方法综述

针对近年来目标检测算法的最新研究进展, 从目标检测框架的子模块设计优化角度出发, 对该领域中一些有启发性的研究成果进行整理、归纳和分析, 并对目标检测模型的优化思路提出一些建议.

蒋弘毅, 王永娟, 康锦煜. 目标检测模型及其优化方法综述. 自动化学报, 2021, 47(6): 1232−1255

基于显著图融合的无人机载热红外图像目标检测方法

针对无人机平台内存和算力的局限性, 设计了YOLOv3-MobileNetv2网络, 利用轻量化网络MobileNetv2替代YOLOv3原有的特征提取网络DarkNet53, 在大量减少网络参数的同时显著提升运行速度. 此外, 使用Focal loss改进YOLOv3原有的损失函数, 解决正负样本不均衡问题, 使得网络专注于困难样本的计算.

赵兴科, 李明磊, 张弓, 黎宁, 李家松. 基于显著图融合的无人机载热红外图像目标检测方法. 自动化学报, 2021, 47(9): 2120−2131