Mechanical Particle Filter-Based Active Vision System for Fast Wide-Area Multiobject Detection
Long XL(龙宪磊)1,2; Ma, Liping1; Jiang, He1,2; Li, Zhikai1,2; Chen, Yunze1,2; Gu, Qingyi1
Source PublicationIEEE Transactions on Instrumentation and Measurement
ISSN0018-9456
2022-08
Volume71Pages:1-13
Abstract

Sliding window searching is a common paradigm used for large field-of-view (FOV) multiobject detection. However, low target resolution and irregular distribution of objects make the traditional method inefficient and hard to meet the demands of fast object detection in a wide area. To address the above issues, in this article we design an ultrafast switching galvano-mirror-based active vision system that can simultaneously search and detect multiobject at 300 frames per second (fps) in a large FOV, without the guidance of a high-resolution (HR) panoramic camera. Specifically, a novel mechanical particle filter (MPF) framework is proposed to improve the efficiency of searching potential objects by constructing the probability distribution model and then iteratively locating target objects, where each particle is represented by an image of a single small FOV that is controlled and captured by a switching galvano-mirror. The object detector is used to determine whether the particle image contains target objects. Then, to reduce the total scanning cost for capturing all the particles, we introduce a center partitioned scanning algorithm that significantly speeds up the scanning process by 10× or more. Moreover, we optimize the processing pipeline for low latency sensing using a high-parallelism visual feedback scheme to improve efficiency. Finally, an efficient and fast active vision system is implemented, which is based on a two-axis galvano-mirror and a high-speed camera without multicamera fusion for active object searching. Abundant experimental results demonstrate the effectiveness of our proposed system in indoor environments and real-world scenarios. We also verify the performance in terms of scanning speed and detection accuracy.

KeywordHigh-speed vision , mechanical particle filter (MPF) , object detection , visual feedback , wide-area surveillance
Indexed BySCI
Language英语
Sub direction classification机器人感知与决策
planning direction of the national heavy laboratory环境多维感知
Document Type期刊论文
Identifierhttp://ir.ia.ac.cn/handle/173211/50597
Collection中科院工业视觉智能装备工程实验室_精密感知与控制
Corresponding AuthorGu, Qingyi
Affiliation1.中国科学院自动化所研究所
2.中国科学院大学
Recommended Citation
GB/T 7714
Long XL,Ma, Liping,Jiang, He,et al. Mechanical Particle Filter-Based Active Vision System for Fast Wide-Area Multiobject Detection[J]. IEEE Transactions on Instrumentation and Measurement,2022,71:1-13.
APA Long XL,Ma, Liping,Jiang, He,Li, Zhikai,Chen, Yunze,&Gu, Qingyi.(2022).Mechanical Particle Filter-Based Active Vision System for Fast Wide-Area Multiobject Detection.IEEE Transactions on Instrumentation and Measurement,71,1-13.
MLA Long XL,et al."Mechanical Particle Filter-Based Active Vision System for Fast Wide-Area Multiobject Detection".IEEE Transactions on Instrumentation and Measurement 71(2022):1-13.
Files in This Item: Download All
File Name/Size DocType Version Access License
Mechanical_Particle_(4989KB)期刊论文作者接受稿开放获取CC BY-NC-SAView Download
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[Long XL(龙宪磊)]'s Articles
[Ma, Liping]'s Articles
[Jiang, He]'s Articles
Baidu academic
Similar articles in Baidu academic
[Long XL(龙宪磊)]'s Articles
[Ma, Liping]'s Articles
[Jiang, He]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[Long XL(龙宪磊)]'s Articles
[Ma, Liping]'s Articles
[Jiang, He]'s Articles
Terms of Use
No data!
Social Bookmark/Share
File name: Mechanical_Particle_Filter-Based_Active_Vision_System_for_Fast_Wide-Area_Multiobject_Detection.pdf
Format: Adobe PDF
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.