Space Object Recognition Research Articles

Stereoscopic artificial compound eyes for spatiotemporal perception in three-dimensional space.

Arthropods' eyes are effective biological vision systems for object tracking and wide field of view because of their structural uniqueness; however, unlike mammalian eyes, they can hardly acquire the depth information of a static object because of their monocular cues. Therefore, most arthropods rely on motion parallax to track the object in three-dimensional (3D) space. Uniquely, the praying mantis (Mantodea) uses both compound structured eyes and a form of stereopsis and is capable of achieving object recognition in 3D space. Here, by mimicking the vision system of the praying mantis using stereoscopically coupled artificial compound eyes, we demonstrated spatiotemporal object sensing and tracking in 3D space with a wide field of view. Furthermore, to achieve a fast response with minimal latency, data storage/transportation, and power consumption, we processed the visual information at the edge of the system using a synaptic device and a federated split learning algorithm. The designed and fabricated stereoscopic artificial compound eye provides energy-efficient and accurate spatiotemporal object sensing and optical flow tracking. It exhibits a root mean square error of 0.3 centimeter, consuming only approximately 4 millijoules for sensing and tracking. These results are more than 400 times lower than conventional complementary metal-oxide semiconductor-based imaging systems. Our biomimetic imager shows the potential of integrating nature's unique design using hardware and software codesigned technology toward capabilities of edge computing and sensing.

Remote Big Data Management and Visual Imagery Tools, Multisensor Fusion and Dynamic Routing Technologies, and 3D Space Mapping and Object Recognition Algorithms on Blockchain-based Metaverse Platforms

Remote Big Data Management and Visual Imagery Tools, Multisensor Fusion and Dynamic Routing Technologies, and 3D Space Mapping and Object Recognition Algorithms on Blockchain-based Metaverse Platforms

Localization and classification of space objects using EfficientDet detector for space situational awareness

Space situational awareness (SSA) systems play a significant role in space navigation missions. One of the most essential tasks of this system is to recognize space objects such as spacecrafts and debris for various purposes including active debris removal, on-orbit servicing, and satellite formation. The complexity of object recognition in space is due to several sensing conditions, including the variety of object sizes with high contrast, low signal-to-noise ratio, noisy backgrounds, and several orbital scenarios. Existing methods have targeted the classification of images containing space objects with complex backgrounds using various convolutional neural networks. These methods sometimes lose attention on the objects in these images, which leads to misclassification and low accuracy. This paper proposes a decision fusion method that involves training an EfficientDet model with an EfficientNet-v2 backbone to detect space objects. Furthermore, the detected objects were augmented by blurring and by adding noise, and were then passed into the EfficientNet-B4 model for training. The decisions from both models were fused to find the final category among 11 categories. The experiments were conducted by utilizing a recently developed space object dataset (SPARK) generated from realistic space simulation environments. The dataset consists of 11 categories of objects with 150,000 RGB images and 150,000 depth images. The proposed object detection solution yielded superior performance and its feasibility for use in real-world SSA systems was demonstrated. Results show significant improvement in accuracy (94%), and performance metric (1.9223%) for object classification and in mean precision (78.45%) and mean recall (92.00%) for object detection.

Resolving the neural mechanism of core object recognition in space and time: A computational approach

The underlying mechanism of object recognition- a fundamental brain ability- has been investigated in various studies. However, balancing between the speed and accuracy of recognition is less explored. Most of the computational models of object recognition are not potentially able to explain the recognition time and, thus, only focus on the recognition accuracy because of two reasons: lack of a temporal representation mechanism for sensory processing and using non-biological classifiers for decision-making processing. Here, we proposed a hierarchical temporal model of object recognition using a spiking deep neural network coupled to a biologically plausible decision-making model for explaining both recognition time and accuracy. We showed that the response dynamics of the proposed model can resemble those of the brain. Firstly, in an object recognition task, the model can mimic human’s and monkey’s recognition time as well as accuracy. Secondly, the model can replicate different speed-accuracy trade-off regimes as observed in the literature. More importantly, we demonstrated that temporal representation of different abstraction levels (superordinate, midlevel, and subordinate) in the proposed model matched the brain representation dynamics observed in previous studies. We conclude that the accumulation of spikes, generated by a hierarchical feedforward spiking structure, to reach abound can well explain not even the dynamics of making a decision, but also the representations dynamics for different abstraction levels.

Evaluating interhemispheric connectivity during midline object recognition using EEG.

Functional integration between two hemispheres is crucial for perceptual binding to occur when visual stimuli are presented in the midline of the visual field. Mima and colleagues (2001) showed using EEG that midline object recognition was associated with task-related decrease in alpha band power (alpha desynchronisation) and a transient increase in interhemispheric coherence. Our objective in the current study was to replicate the results of Mima et al. and to further evaluate interhemispheric effective connectivity during midline object recognition in source space. We recruited 11 healthy adult volunteers and recorded EEG from 64 channels while they performed a midline object recognition task. Task-related power and coherence were estimated in sensor and source spaces. Further, effective connectivity was evaluated using Granger causality. While we were able to replicate the alpha desynchronisation associated with midline object recognition, we could not replicate the coherence results of Mima et al. The data-driven approach that we employed in our study localised the source of alpha desynchronisation over the left occipito-temporal region. In the alpha band, we further observed significant increase in imaginary part of coherency between bilateral occipito-temporal regions during object recognition. Finally, Granger causality analysis between the left and right occipito-temporal regions provided an insight that even though there is bidirectional interaction, the left occipito-temporal region may be crucial for integrating the information necessary for object recognition. The significance of the current study lies in using high-density EEG and applying more appropriate and robust measures of connectivity as well as statistical analysis to validate and enhance our current knowledge on the neural basis of midline object recognition.

ARSRNet: accurate space object recognition using optical cross section curves.

Limited by the conditions and performance of ground-based optical observations, it is difficult for us to obtain a plethora of optical cross section (OCS) data for some space objects (SOs). Unevenly distributed OCS data and unclear labels will affect the performance of SOs recognition based on neural networks. Furthermore, when we need to identify a new SO or SO category using deep neural network, the trained network model may no longer be applicable. We need to retrain the network with new training data. In order to alleviate these problems and improve the generalization and training convergence speed of SOs recognition networks, a novel, to the best of our knowledge, neural network model, ARSRNet, is proposed in this paper. The ARSRNet can identify SOs and their attitude accurately using only a small quantity of training OCS data and without clear labels. And the proposed network is able to adapt to new recognition tasks. Meanwhile, we propose an AdamRprop network optimization algorithm to accelerate network training and improve recognition accuracy. Experimental results show that the recognition accuracy of ARSRNet reaches 90.60% on the test OCS dataset. Compared with mainstream network optimization algorithms, the proposed AdamRprop is more appropriate for ARSRNet and can accelerate the convergence of ARSRNet.

Segmenting Star Images with Complex Backgrounds Based on Correlation between Objects and 1D Gaussian Morphology

Space object recognition in high Earth orbits (between 2000 km and 36,000 km) is affected by moonlight and clouds, resulting in some bright or saturated image areas and uneven image backgrounds. It is difficult to separate dim objects from complex backgrounds with gray thresholding methods alone. In this paper, we present a segmentation method of star images with complex backgrounds based on correlation between space objects and one-dimensional (1D) Gaussian morphology, and the focus is shifted from gray thresholding to correlation thresholding. We build 1D Gaussian functions with five consecutive column data of an image as a group based on minimum mean square error rules, and the correlation coefficients between the column data and functions are used to extract objects and stars. Then, lateral correlation is repeated around the identified objects and stars to ensure their complete outlines, and false alarms are removed by setting two values, the standard deviation and the ratio of mean square error and variance. We analyze the selection process of each thresholding, and experimental results demonstrate that our proposed correlation segmentation method has obvious advantages in complex backgrounds, which is attractive for object detection and tracking on a cloudy and bright moonlit night.

Single Space Object Image Denoising and Super-Resolution Reconstructing Using Deep Convolutional Networks

Space object recognition is the basis of space attack and defense confrontation. High-quality space object images are very important for space object recognition. Because of the large number of cosmic rays in the space environment and the inadequacy of optical lenses and detectors on satellites to support high-resolution imaging, most of the images obtained are blurred and contain a lot of cosmic-ray noise. So, denoising methods and super-resolution methods are two effective ways to reconstruct high-quality space object images. However, most super-resolution methods could only reconstruct the lost details of low spatial resolution images, but could not remove noise. On the other hand, most denoising methods especially cosmic-ray denoising methods could not reconstruct high-resolution details. So in this paper, a deep convolutional neural network (CNN)-based single space object image denoising and super-resolution reconstruction method is presented. The noise is removed and the lost details of the low spatial resolution image are well reconstructed based on one very deep CNN-based network, which combines global residual learning and local residual learning. Based on a dataset of satellite images, experimental results demonstrate the feasibility of our proposed method in enhancing the spatial resolution and removing the noise of the space objects images.

非负矩阵分解在空间目标图像识别中的应用

非负矩阵分解在空间目标图像识别中的应用

Space collision probability computation based on on-board optical cues

This paper presents a novel on-board space collision analysis method for space situational awareness. The framework is developed under the following assumptions: 1) A satellite can be equipped with on-board sensors for space object recognition. 2) No a–priori knowledge of the space objects is provided. A space object size and relative state estimation method is firstly proposed, wherein optical cues acquired from onboard sensors are utilized to achieve the estimation. Then, the unscented transform approach is employed to calculate the probability density function (PDF) of collision probability based on the estimate information. Monte Carlo simulations and an experimental test demonstrate that the proposed approach can achieve high-precision on-board collision probability estimation with an error less than 3%.

Convolutional neural network-based encoding and decoding of visual object recognition in space and time

Representations learned by deep convolutional neural networks (CNNs) for object recognition are a widely investigated model of the processing hierarchy in the human visual system. Using functional magnetic resonance imaging, CNN representations of visual stimuli have previously been shown to correspond to processing stages in the ventral and dorsal streams of the visual system. Whether this correspondence between models and brain signals also holds for activity acquired at high temporal resolution has been explored less exhaustively. Here, we addressed this question by combining CNN-based encoding models with magnetoencephalography (MEG). Human participants passively viewed 1,000 images of objects while MEG signals were acquired. We modelled their high temporal resolution source-reconstructed cortical activity with CNNs, and observed a feed-forward sweep across the visual hierarchy between 75 and 200 ms after stimulus onset. This spatiotemporal cascade was captured by the network layer representations, where the increasingly abstract stimulus representation in the hierarchical network model was reflected in different parts of the visual cortex, following the visual ventral stream. We further validated the accuracy of our encoding model by decoding stimulus identity in a left-out validation set of viewed objects, achieving state-of-the-art decoding accuracy.

Image quality analysis and improvement of Ladar reflective tomography for space object recognition

Some problems in the application of Ladar reflective tomography for space object recognition are studied in this work. An analytic target model is adopted to investigate the image reconstruction properties with limited relative angle range, which are useful to verify the target shape from the incomplete image, analyze the shadowing effect of the target and design the satellite payloads against recognition via reflective tomography approach. We proposed an iterative maximum likelihood method basing on Bayesian theory, which can effectively compress the pulse width and greatly improve the image resolution of incoherent LRT system without loss of signal to noise ratio.

New control technology for visible light reflection characteristics of space target

For geosynchronous( GEO) orbit space object's detection and recognition is mainly dependent on analysis,identification and judgment of reflected sunlight photometric curve. Based on analysis of visible light reflection characteristics calculation model of space target surface diffuse reflection properties,a study on different shape reflector light reflection characteristics was carried out. Moreover,a different shape reflector method was proposed to control space target light reflection characteristics. Taking an example of a GEO satellite,simulate and analyze the effects of cone,sphere and cylinder as reflection control body on satellite reflection characteristics by using Monte Carlo non-sequential light tracing method. The results show that the method can change the satellite original visible light reflection curve,reduce the probability of being detected and recognized,and improve the invisibility of space object.

Multi-view space object recognition and pose estimation based on kernel regression

The application of high-performance imaging sensors in space-based space surveillance systems makes it possible to recognize space objects and estimate their poses using vision-based methods. In this paper, we proposed a kernel regression-based method for joint multi-view space object recognition and pose estimation. We built a new simulated satellite image dataset named BUAA-SID 1.5 to test our method using different image representations. We evaluated our method for recognition-only tasks, pose estimation-only tasks, and joint recognition and pose estimation tasks. Experimental results show that our method outperforms the state-of-the-arts in space object recognition, and can recognize space objects and estimate their poses effectively and robustly against noise and lighting conditions.

Resolving human object recognition in space and time

A comprehensive picture of object processing in the human brain requires combining both spatial and temporal information about brain activity. Here, we acquired human magnetoencephalography (MEG) and functional magnetic resonance imaging (fMRI) responses to 92 object images. Multivariate pattern classification applied to MEG revealed the time course of object processing: whereas individual images were discriminated by visual representations early, ordinate and superordinate category levels emerged relatively later. Using representational similarity analysis, we combine human fMRI and MEG to show content-specific correspondence between early MEG responses and primary visual cortex (V1), and later MEG responses and inferior temporal (IT) cortex. We identified transient and persistent neural activities during object processing, with sources in V1 and IT., Finally, human MEG signals were correlated to single-unit responses in monkey IT. Together, our findings provide an integrated space- and time-resolved view of human object categorization during the first few hundred milliseconds of vision.

基于TracePro的空间目标光学散射特性建模与仿真

For supplying space object detection and recognition with reference data,space object optical scattering characteristics model is established using TracePro.Combining with structure,material,background and orbit,the optical scattering characteristics of the space-based infrared system is simulated by creating solid model,defining material,defining light source and tracing rays in TracePro.The simulation results show that the object spectral irradiance is coincident with that of the sun.The object equivalent reflectivity is coincident with that of the GaAs in specular direction.As the object rotates,the object equivalent reflectivity turns to agree with that of the cladding material,and then keeps stable.The research could provide reference idea for the next work.

Full-viewpoint 3D Space Object Recognition Based on Kernel Locality Preserving Projections

Space object recognition plays an important role in spatial exploitation and surveillance, followed by two main problems: lacking of data and drastic changes in viewpoints. In this article, firstly, we build a three-dimensional (3D) satellites dataset named BUAA Satellite Image Dataset (BUAA-SID 1.0) to supply data for 3D space object research. Then, based on the dataset, we propose to recognize full-viewpoint 3D space objects based on kernel locality preserving projections (KLPP). To obtain more accurate and separable description of the objects, firstly, we build feature vectors employing moment invariants, Fourier descriptors, region covariance and histogram of oriented gradients. Then, we map the features into kernel space followed by dimensionality reduction using KLPP to obtain the submanifold of the features. At last, k-nearest neighbor (kNN) is used to accomplish the classification. Experimental results show that the proposed approach is more appropriate for space object recognition mainly considering changes of viewpoints. Encouraging recognition rate could be obtained based on images in BUAA-SID 1.0, and the highest recognition result could achieve 95.87%.

Scattering matrix application

The use of the scattering matrix leads to a formulation of radar reflectivity that is independent of radar polarization, target range, and target rotation about the line of sight. The resulting formulation is applied to the problem of object recognition in free space. Experimental results for right circular cones are presented.