检索结果-内蒙古大学图书馆

Target-oriented deformable fast depth estimation based on stereo vision for space object detection

MEASUREMENT 2025年 245卷

作者： Xu, Chengcheng Zhao, Haiyan Gao, Bingzhao Liu, Hangyu Xie, Hongbin Jilin Univ Coll Commun Engn Changchun 130025 Peoples R China Tongji Univ Sch Automot Studies Shanghai 201804 Peoples R China

To address problems of the poor matching accuracy and speed in space object detection, this paper proposes a deformable fast object matching algorithm. It is a target-oriented depth estimation approach that calculates the disparity value by matching the object on the left and right images. The logical encoding masking layer is designed to achieve the deformable operation, which can fully fuse the semantic or contour feature information of the object. This effectively reduces the computational cost and improves the accuracy. The feature coding method is optimized and upgraded by integrating relative positions and global pixels in the image, solving the problem of mismatching in complex regions without obvious features. Based on the characteristics of stereo vision and the concept of regional matching, an optimal matching search range is proposed. Results show that the average time is less than 9ms and accuracy reaches the state-of-the-art.

关键词： Intelligent vehicles space object detection object matching Depth estimation

来源：评论

学校读者我要写书评

暂无评论

Backlight and dim space object detection based on a novel event camera

引用

PEERJ COMPUTER SCIENCE 2024年 10卷 e2192-e2192页

作者： Zhou, Xiaoli Bei, Chao CASIC Res Acad 2 Grad Sch Beijing Peoples R China CASIC Space Engn Dev Co Ltd Beijing Peoples R China

Background: For space object detection tasks, conventional optical cameras face various application challenges, including backlight issues and dim light conditions. As a novel optical camera, the event camera has the advantages of high temporal resolution and high dynamic range due to asynchronous output characteristics, which provides a new solution to the above challenges. However, the asynchronous output characteristic of event cameras makes them incompatible with conventional object detection methods designed for frame images. Methods: Asynchronous convolutional memory network (ACMNet) for processing event camera data is proposed to solve the problem of backlight and dim space object detection. The key idea of ACMNet is to fi rst characterize the asynchronous event streams with the Event Spike Tensor (EST) voxel grid through the exponential kernel function, then extract spatial features using a feed-forward feature extraction network, and aggregate temporal features using a proposed convolutional spatiotemporal memory module ConvLSTM, and fi nally, the end-to-end object detection using continuous event streams is realized. Results: Comparison experiments among ACMNet and classical object detection methods are carried out on Event_DVS_space7, which is a large-scale space synthetic event dataset based on event cameras. The results show that the performance of ACMNet is superior to the others, and the mAP is improved by 12.7% while maintaining the processing speed. Moreover, event cameras still have a good performance in backlight and dim light conditions where conventional optical cameras fail. This research offers a novel possibility for detection under intricate lighting and motion conditions, emphasizing the superior bene fi ts of event cameras in the realm of space object detection.

关键词： space object detection Event cameras Convolutional neural network (CNN) Long short-term memory network (LSTM)

来源：评论

学校读者我要写书评

暂无评论

Techniques for Improved space object detection Performance from Ground-Based Telescope Systems Using Long and Short Exposure Images

Techniques for Improved Space Object Detection Performance f...

引用

作者： David J. Becker Air Force Institute of Technology

学位级别：博士

space object detection is of great importance in the highly dependent yet competi- tive and congested space domain. detection algorithms employed play a crucial role in fulfilling the detection component in the space situational awareness mission to detect, track, characterize and catalog unknown space objects. Many current space detection algorithms use a matched filter or a spatial correlator on long exposure data to make a detection decision at a single pixel point of a spatial image based on the assumption that the data follows a Gaussian distribution. This research focuses on improving current space object detection algorithms and developing new algorithms that provide a greater detection performance, specifically with dim and small objects which are inherently difficult to detect. With a greater detection rate, a great num- ber of unknown objects will be detected, tracked and cataloged to deliver safer space operations. Three novel approaches to object detection using long and short exposure images obtained from ground-based telescopes are examined in this dissertation. Two of the approaches examined in this dissertation involved detection algorithms designed from the probability distribution of the real component of the Fourier trans- form of the intensity of the object. Models for the object and its statistics were calculated to build a likelihood ratio test. Additionally, spatial frequency correlation was examined to build a joint probability distribution function likelihood ratio test. Both the Fourier point detector and Fourier correlation algorithm demonstrated a significant increase in the probability of detection over their commonly used spatial counterpart algorithms. Long exposure imaging is critical to detection performance in these algorithms, however if imaging under daylight conditions it becomes necessary to create a long exposure image as the sum of many short exposure images. The third approach in this research explores the potential to increa

关键词： detection space Situational Awareness Telescope Optics space object detection

来源：评论

学校读者我要写书评

暂无评论

space object detection: Receiver Operating Characteristics for Poisson and Normally Distributed Data 12

Space Object Detection: Receiver Operating Characteristics f...

引用

Conference on Unconventional Imaging and Wavefront Sensing XII

作者： McMurry, Richard Air Force Inst Technol Dept Engn Wright Patterson AFB OH 45433 USA

ISBN: (纸本)9781510603554;9781510603561

object detection algorithms typically implement a Likelihood Ratio Test (LRT) for determining if an image-pixel contains a star. Implementing a LRT requires prior knowledge about the statistics of the data, such as the mean and variance. Assuming the background noise follows a Gaussian distribution, the mean and the variance have to be calculated separately. If the background data follows a Poisson distribution, the mean is the only calculation needed, as mean and variance of the Poisson distribution equal each other. This paper will compare the possible detection improvements when using a Poisson assumption. Many star detection LRTs will use a windowing technique to limit the amount of background data that is being tested. Various window sizes will also be tested to determine possible detection improvements that can be realized.

关键词： space object detection ROC Gaussian noise Poisson noise space situational awareness SSA outlier rejection window size LRT

来源：评论

学校读者我要写书评

暂无评论

Improving space object detection using a Fourier likelihood ratio detection algorithm 12

Improving space object detection using a Fourier likelihood ...

引用

Conference on Unconventional Imaging and Wavefront Sensing XII

作者： Becker, David J. Cain, Stephen C. Air Force Inst Technol 2950 Hobson Way Dayton OH 45433 USA

ISBN: (纸本)9781510603554;9781510603561

In this paper a new detection algorithm is proposed and developed for detecting space objects from images obtained using a ground-based telescope with the goal to improve space situational awareness. Most current space object detection algorithms rely on developing a likelihood ratio test (LRT) for the observed data based on a binary hypothesis test. These algorithms are based on the assumption that the observed data is Gaussian or Poisson distributed under both the hypothesis that a low signal-to-noise ratio (SNR) space object is present in the data and the hypothesis that an object is absent from the data. The LRT algorithm in this paper was developed based on the assumption that the distribution of the Fourier transform of the observed data will be different when a low SNR object is present in the data compared to when the data only contains background noise and known space objects. When an object is present the probability distribution of the real component of the Fourier transform of the intensity was found to follow a Gaussian distribution with a mean significantly different than in the data that doesn't contain an object even at low SNR levels. As the separation of these two probability distribution functions increases, it becomes more likely that an object can be detected. In this paper, simulated data are used to demonstrate the effectiveness and to highlight the benefits gained from this algorithm.

关键词： space object detection likelihood ratio test detection algorithms Fourier transform random phasor sum

来源：评论

学校读者我要写书评

暂无评论

Improved space object detection via scintillated short exposure image data 12

Improved space object detection via scintillated short expos...

引用

Conference on Unconventional Imaging and Wavefront Sensing XII

作者： Cain, Stephen Dept Elect & Comp Engn 2950 Hobson Way Wright Patterson AFB OH 45433 USA

ISBN: (纸本)9781510603554;9781510603561

Since telescopes were first trained on the heavens, nearly every attempt to locate new objects in the sky has been accomplished using long-exposure images (images taken with exposures longer than one second). For many decades astronomers have utilized short-exposure images to achieve high spatial resolution and mitigate the effect of Earth's atmosphere. This is due to the fact that short-exposure images contain higher spatial frequency content than their long exposure counterparts. New objects, such as dim companions of binary stars have been revealed using these techniques, thus achieving some discoveries via short-exposure images, but always in context of a detailed analysis of a specific area of the sky, not through a synoptic search of the heavens like the kind carried out by asteroid research programs like space Watch, PAN-STARRS or the Catalina Sky Survey. In this paper simulated short-exposure images are processed to detect simulated stars in the presence of strong atmospheric turbulence. The strong turbulence condition produces long-exposures with wide star patterns, while the short-exposure images feature better concentration of the available photons, but large random position variance. A new processing strategy involving a matched-filter technique using a short-exposure atmospheric impulse response model is utilized The results from this process are compared to efforts to detect stars in long exposure images taken over the same interval the short-exposure images are collected. The two techniques are compared and the short-exposure imaging technique is found to produce the higher probability of detection.

关键词： short-exposure image space situational awareness scintillation space object detection

来源：评论

学校读者我要写书评

暂无评论

A Semi-Supervised object detection Method for Close Range detection of spacecraft and space Debris

引用

INTERNATIONAL JOURNAL OF AERONAUTICAL AND space SCIENCES 2025年第2期26卷 773-784页

作者： Zhang, Huan Zhang, Yang Feng, Qingjuan Zhang, Kebei Beijing Informat Sci & Technol Univ Beijing 100192 Peoples R China Beijing Inst Control Engn Beijing 100094 Peoples R China

With the continuous development of space missions, a large number of artificial objects have been produced and deployed in Earth orbit, and there is an urgent need to equip satellites with the ability to autonomously detect nearby space targets to cope with the harm caused by non-cooperative targets to the safety of in orbit satellites. To meet the needs of intelligent detection of spacecraft and space debris, we propose a semi supervised target detection method with domain generalization capability. This method aims to improve the detection methods for spacecraft and debris, utilizing limited labeled data and a large amount of unlabeled data to train end-to-end networks and reduce dependence on data labeled information. In the Semi-Supervised Learning framework, we optimized and improved the dense detection network model, enhanced the Receptive field of the model, and optimized the feature fusion and representation capabilities of the model. The improved semi supervised object detection algorithm trained with only 10% annotation information achieved an mAP@0.50:0.95 metric of 69.2% on the SPARK space object detection data set, which is superior to object detection algorithms such as Fast R-CNN that utilize 100% annotation information. In addition, a fast image style transfer similar to target domain image augmentation method has been introduced to address the challenge of model domain generalization. The addition of generated style transfer images in the semi supervised method enhances the model's generalization ability towards the target domain, thereby improving the detection performance of the target domain.

关键词： space object detection Semi-supervised object detection Domain generalization

来源：评论

学校读者我要写书评

暂无评论

Characterizing Point Spread Function (PSF) fluctuations to improve Resident space object (RSO) detection 8

Characterizing Point Spread Function (PSF) fluctuations to i...

引用

Conference on Sensors and Systems for space Applications

作者： Hardy, Tyler J. Cain, Stephen C. Air Force Inst Technol Dayton OH 45433 USA

ISBN: (纸本)9781628415858

This research paper deals with methods for improving the performance of Electro-optical detection systems designed to find Resident space objects (RSOs). Some methods for detecting RSOs rely on accurate knowledge of the system Point Spread Function (PSF). The PSF is a function of the telescope optics, the atmosphere, and other factors including object intensity and noise present in the system. Due to the random photon arrival times, any observed data will contain Poisson noise. Assuming that other noise sources such as dark current and readout noise do not contribute significantly, the final source of intensity fluctuations in the data is the atmosphere. To quantify these fluctuations, an optical model of a telescope system is developed, and its PSF is simulated. In a long exposure image, the effects of the atmosphere are well characterized with the long exposure atmosphere Optical Transfer Function (OTF). In contrast, a short exposure image does not average the fluctuations as effectively. To model the atmosphere, random phase screens with Kolmogorov statistics are added to the optical model to observe PSF fluctuations in short exposure telescope data. The distribution of the peak intensity is analyzed for varying exposure times and atmospheric turbulence strengths. This distribution is combined with the Poisson random arrival times of photons to create a combined model for received data, which is then used to design a new detection algorithm. The performance of the new space object detection algorithm will be compared to a traditional algorithm using simulated telescope data.

关键词： space object detection Remote Sensing Telescope Signal Processing Intensity Fluctuations

来源：评论

学校读者我要写书评

暂无评论

A Long-Distance 6D Pose Estimation System for space object Recognition Using Region-scanning LiDAR 23

A Long-Distance 6D Pose Estimation System for Space Object R...

引用

Conference on MOEMS and Miniaturized Systems XXIII

作者： Lim, Yoon-Seop An, Seok-Jun Lee, Sung-Hyun Park, Yong-Hwa Korea Adv Inst Sci & Technol Dept Mech Engn Daejeon 34141 South Korea Samsung Elect Co Ltd Hwasung 445701 Gyeonggido South Korea

ISBN: (纸本)9781510670594;9781510670587

In this paper, a novel 6-Dimensional (3 positions and 3 orientations) pose estimation system using indirect Time-of-Flight (ToF) region-scanning LiDAR is proposed for long-distance space object recognition. Specifically, the targeted space object detection algorithm with an IR amplitude image, and 6D pose estimation algorithm with high-resolution 3D data are performed with the region-scanning LiDAR. The proposed system is verified with a self-collected dataset of space objects in space simulation environments. The proposed pose estimation algorithm shows a maximum position error of 1% and orientation error of 5 degrees in robust experiment conditions. Moreover, the proposed system outperformed the other conventional space object 6D pose estimation system with previous depth sensors in terms of the detection rates and 6D pose accuracy at long distances.

关键词： 6D object pose estimation Time-of-Flight (ToF) High-Resolution LiDAR space object detection

来源：评论

学校读者我要写书评

暂无评论

Bistatic detection of LEO Satellites from Very Long Distances Using LOFAR Radio Telescope

Bistatic Detection of LEO Satellites from Very Long Distance...

引用

International Radar Symposium (IRS)

作者： Jedrzejewski, Konrad Pozoga, Mariusz Modrzewski, Andrzej Karwacki, Michal Kulpa, Krzysztof Rothkaehl, Hanna Warsaw Univ Technol Inst Elect Syst Warsaw Poland Polish Acad Sci Space Res Ctr Warsaw Poland PIT Radwar SA Warsaw Poland

ISBN: (纸本)9798350371109;9788395602092

The paper presents the results of experimental investigations on the capabilities of detecting very-long-distance space objects moving in Low-Earth Orbit (LEO) based on bistatic measurements using the antenna array of a LOFAR (LOw-Frequency ARray) radio telescope and a non-cooperative radar illuminator operating in the VHF band. In this system, the signal transmitted by the non-cooperating radar is recorded using a relatively simple reference signal receiver located near the illuminator. The sensitive antenna array of the LOFAR radio telescope enables the reception of weak echo signals reflected from space objects orbiting in LEO, even those located extremely far from the illuminator and receiver in the bistatic sense. The detection of very distant and high-speed objects is made possible through the utilization of a dedicated object motion compensation method.

关键词： space situational awareness SSA space object detection bistatic radar LOFAR

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：