False Stars Research Articles

RPNet: A Representation Learning-Based Star Identification Algorithm

A novel star identification network (RPNet) based on representation learning is proposed in this paper. Unlike other pattern-based stars identification algorithms, the RPNet does not require the creation of an elaborate pattern, nor does it need to search among patterns. Instead, a star pattern generator (SPG) in the RPNet helps in finding the best pattern that can distinguish different stars clearly. A star pattern classifier (SPC) in the RPNet is utilized to recognize the pattern generated before. The simulations show that the RPNet is extremely robust toward star position noise, star magnitude noise, and false stars. The performance on simulation images outperforms almost all other pattern-based stars identification algorithms. On average, it achieves an identification rate of 99.23% in simulated star images. The identification rate on real star images is higher than 98%. Moreover, the algorithm achieves this performance with lesser memory and faster speed compared to polygon algorithms.

Использование матриц смежности для визуализации больших графов

Exponential size growth of such graphs as social networks, Internet graphs, etc. requires new approaches to their visualization. Along with node-link diagram representations, adjacency matrices and various hybrid representations are increasingly used for large graphs visualizations. This survey discusses new approaches to the visualization of large graphs using adjacency matrices and gives examples of applications where these approaches are used. We describe various types of patterns arising when adjacency matrices corresponding to modern networks are ordered, and algorithms making it possible to reveal these patterns. In particular, the use of matrix ordering methods in conjunction with algorithms looking for such graph patterns as stars, false stars, chains, near-cliques, full cliques, bipartite cores and near-bipartite cores enable users to create understandable visualizations of graphs with millions of vertices and edges. Examples of hybrid visualizations using node-link diagrams for representing sparse parts of a graph and adjacency matrices for representing dense parts are also given. The hybrid methods are used to visualize co-authorship networks, deep neural networks, to compare networks of the human brain connectivity, etc.

A robust and fast star identification algorithm based on an ordered set of points pattern

A novel, robust, and fast star identification algorithm based on an OSP∗ pattern is presented in this paper. This new algorithm is appropriate for star sensors in the initial acquisition mode, in which no priori attitude information is available. The OSP pattern is constrained with the k-Nearest Neighbor algorithm with a specified order; hence, this pattern is very robust for star identification. In addition, a new hash search approach for accessing the star pattern database is proposed to improve identification speed. The proposed hash search approach is built on the identification error limit and quantified star angular distance. Analysis of numerical results shows that, with increases in the star centroid error, the ISR† of the proposed algorithm is high and stable, while the ISR values of the triangle algorithm, grid algorithm, and pyramid algorithm decrease rapidly. When the star centroid error is three pixels, the ISR of the proposed algorithm is 99.8%, which is 16% higher than the ISR values of the triangle algorithm and grid algorithm, and nearly 10% higher than that of the pyramid algorithm. Similarly, when the number of false stars is increased, the ISR of the proposed algorithm is high and stable, whereas the ISR values of the triangle algorithm, grid algorithm, and pyramid algorithm decrease rapidly. When there are three false stars in an image, the ISR of the proposed algorithm is 98.4%, which is 13% higher than the ISR values of the triangle algorithm and grid algorithm, and 3% higher than that of the pyramid algorithm. Furthermore, the time complexity of the proposed hash search approach is O(1), which is significantly faster than the time complexities of the linear search and k-vector search, which are O(n) and O(k), respectively. Compared with the traditional hash search approach, the proposed approach is much faster when using a compatible database.

Star Identification Based on Hash Map

A fast and robust star identification algorithm based on hash map is proposed in this paper. The algorithm is derived from the classical triangle feature matching algorithm but significantly improves the efficiency and robustness. By mapping each triangle feature to an integer and building a hash map of all the triangle features, the proposed algorithm is capable of matching a triangle feature with the computational complexity only O(8); based on this characteristics, a double-triangle verification method is proposed to improve the robustness of identification; moreover, an optimized strategy for triangle feature selection is proposed, which further improves efficiency of the algorithm. Experimental results on both simulation images and night sky images show that the proposed algorithm is quite robust to false stars, position noise and magnitude noise. The identification rate of the algorithm is higher than the grid algorithm and similar to the pyramid algorithm, while the identification speed is up to 100 times faster than the pyramid algorithm.

A triangle voting algorithm based on double feature constraints for star sensors

A novel autonomous star identification algorithm is presented in this study. In the proposed algorithm, each sensor star constructs multi-triangle with its bright neighbor stars and obtains its candidates by triangle voting process, in which the triangle is considered as the basic voting element. In order to accelerate the speed of this algorithm and reduce the required memory for star database, feature extraction is carried out to reduce the dimension of triangles and each triangle is described by its base and height. During the identification period, the voting scheme based on double feature constraints is proposed to implement triangle voting. This scheme guarantees that only the catalog star satisfying two features can vote for the sensor star, which improves the robustness towards false stars. The simulation and real star image test demonstrate that compared with the other two algorithms, the proposed algorithm is more robust towards position noise, magnitude noise and false stars.

False star detection and isolation during star tracking based on improved chi-square tests.

The star sensor is a precise attitude measurement device for a spacecraft. Star tracking is the main and key working mode for a star sensor. However, during star tracking, false stars become an inevitable interference for star sensor applications, which may result in declined measurement accuracy. A false star detection and isolation algorithm in star tracking based on improved chi-square tests is proposed in this paper. Two estimations are established based on a Kalman filter and a priori information, respectively. The false star detection is operated through adopting the global state chi-square test in a Kalman filter. The false star isolation is achieved using a local state chi-square test. Semi-physical experiments under different trajectories with various false stars are designed for verification. Experiment results show that various false stars can be detected and isolated from navigation stars during star tracking, and the attitude measurement accuracy is hardly influenced by false stars. The proposed algorithm is proved to have an excellent performance in terms of speed, stability, and robustness.

Robust polygon recognition method with similarity invariants applied to star identification

In the star identification process the goal is to recognize a star by using the celestial bodies in its vicinity as context. An additional requirement is to avoid having to perform an exhaustive scan of the star database. In this paper we present a novel approach to star identification using similarity invariants. More specifically, the proposed algorithm defines a polygon for each star, using the neighboring celestial bodies in the field of view as vertices. The mapping is insensitive to similarity transformation; that is, the image of the polygon under the transformation is not affected by rotation, scaling or translations. Each polygon is associated with an essentially unique complex number. We perform an exhaustive experimental validation of the proposed algorithm using synthetic data generated from the star catalog with uniformly-distributed positional noise introduced to each star. The star identification method that we present is proven to be robust, achieving a recognition rate of 99.68% when noise levels of up to ±424μ radians are introduced to the location of the stars. In our tests the proposed algorithm proves that if a polygon match is found, it always corresponds to the star under analysis; no mismatches are found. In its present form our method cannot identify polygons in cases where there exist missing or false stars in the analyzed images, in those situations it only indicates that no match was found.

A Discrete HMM-Based Feature Sequence Model Approach for Star Identification

This paper presents a novel star pattern recognition algorithm based on a discrete hidden Markov model (HMM) for autonomous spacecraft attitude determination in the lost-in-space mode. A two-layer structure is proposed to build an HMM-based star pattern for every guide star. The hidden layer describes the unique geometric distributions of stars in the field of view via the transitions among hidden states, and the observation layer uses discrete and rounded feature sequences consisting of the angular distances of star pairs as outputs. An HMM-based star pattern with maximum probabilities in generating the sequences obtained in the recognition process is uniquely matched. Experiments of synthesized and real star images are tested for the HMM algorithm. Results show that the proposed algorithm has a fast average identification time and is highly robust toward star positions, magnitude noise, and false stars. In the simulation, the HMM algorithm exhibits 99.65% identification rate with 0.4 pixel standard deviations of positional noises, 0.323-Mv magnitude noises, and 1.5-ms average identification time. Moreover, the identification rate is 99.11% and the average identification time is 1.2 ms when testing real star images.

A brightness-referenced star identification algorithm for APS star trackers.

Star trackers are currently the most accurate spacecraft attitude sensors. As a result, they are widely used in remote sensing satellites. Since traditional charge-coupled device (CCD)-based star trackers have a limited sensitivity range and dynamic range, the matching process for a star tracker is typically not very sensitive to star brightness. For active pixel sensor (APS) star trackers, the intensity of an imaged star is valuable information that can be used in star identification process. In this paper an improved brightness referenced star identification algorithm is presented. This algorithm utilizes the k-vector search theory and adds imaged stars' intensities to narrow the search scope and therefore increase the efficiency of the matching process. Based on different imaging conditions (slew, bright bodies, etc.) the developed matching algorithm operates in one of two identification modes: a three-star mode, and a four-star mode. If the reference bright stars (the stars brighter than three magnitude) show up, the algorithm runs the three-star mode and efficiency is further improved. The proposed method was compared with other two distinctive methods the pyramid and geometric voting methods. All three methods were tested with simulation data and actual in orbit data from the APS star tracker of ZY-3. Using a catalog composed of 1500 stars, the results show that without false stars the efficiency of this new method is 4∼5 times that of the pyramid method and 35∼37 times that of the geometric method.

A New Composite Algorithm for Identifying the Stars in the Star Tracker

Identifying the stars of field of view(FOV) is a difficult stage in determining the attitude through star tracker. Stars identification algorithms identify the stars of FOV, creating patterns from the stars and comparing them with the information saved in the memory. This research presents a new algorithm composed of pyramid, liebe and planer triangle algorithms. It has at least 4 stars as the stars of FOV and forms a triangle pattern. This algorithm is simulated with 16000 stars in MATLAB. Its accuracy, needed memory and searching time are studied and compared with those of pyramid, Liebe and triangular methods. Two errors, centroiding and in fake stars, are considered in the simulation for more approaching towards real status. According to the obtained results, the new algorithm can well identify the FOV among 140 stars in which only four stars are real and the rests are false. Its searching time is 8% lower, comparing to pyramid method, despite their similar accuracies. It compares the areas and polar moments of the four triangles with the information saved in the memory to meet a unique response. Besides its robustness and proper performance, it includes the false stars as well.

Highly Robust Lost-in-Space Algorithm Based on the Shortest Distance Transform

A robust and fast algorithm that solves the lost-in-space problem for star trackers is presented in this paper. The algorithm is based on an image-processing technique, the shortest distance transform, which transforms the camera image into a two-dimensional lookup table. The information from the database can then be efficiently inserted into this table to compare the camera image with the database. This approach results in an algorithm that is robust to false stars, distortions on star positions, and failure of registration of bright stars. As an example, the algorithm determines over 99% of camera images correctly when 400 false stars are added, distortions of 300 () arcseconds are present, and the brightest star is missing in the image. In case of incorrect determination, a very reliable criterion indicates that the determination step has to be repeated. The robustness of this algorithm can allow the use of star trackers in hostile environments. Furthermore, the algorithm is a valuable contribution to the expanding field of small satellite projects where the low-cost camera components are more prone to error and registration of false stars. Small satellites using this algorithm can acquire great functionality at low component costs.

Weighted ICP Algorithm for Alignment of Stars from Scanned Astronomical Photographic Plates

Given the coarse celestial coordinates of the centre of a plate scan and the field of view, we are looking for a mapping between the stars extracted from the image and the stars from a catalogue, where the stars from both sources are represented by their stellar magnitudes and coordinates, relatively to the image centre. In a previous work we demonstrated the application of Iterative Closest Point (ICP) algorithm for the alignment problem where stars were represented only by their geometrical coordinates. ICP leads to translation and rotation of the initial points - a correction required for one set of stars to fit over the other. This paper extends the previous work by demonstrating significant improvement of ICP by using the stellar magnitudes as point weights. The improvement consists of great decrease of the iteration count until convergence, which helps in the case of highly “misaligned” initial states. The essential aspects of the ICP method like noise tolerance of false or missing stars are still in charge.

Flower algorithm for star pattern recognition in space surveillance with star trackers

Using star tracker to perform space surveillance is a focal point of research in aerospace engineering. However, autonomous attitude determination with star trackers in missions is a challenging task, because of spacecraft attitude dynamics and false stars. We present a novel star pattern recognition algorithm to resolve these problems. The algorithm defines a star pattern, called a flower code, composed of angular distances and circular angles. Then, a three-step strategy is adopted to find the correspondence of the sensor pattern and the catalog pattern, including initial lookup table match, cyclic dynamic match, and validation. A number of experiments are carried out on simulated and real star images. The simulation results show that the proposed method provides improved performance, especially on robustness against false stars. Also, the results for real star images demonstrate the reliability of the method for ground-based measurements.

Fast Star-Pattern Recognition Using Planar Triangles

A new method for star identification using planar triangles is developed and compared to a standard angle method approach. The angle method creates angles between stars within the field of view of a star tracker and matches them to angles in a catalog to determine which stars they are. The new method creates triangles from the stars within the field of view of a star tracker, determines their area and polar moment, then matches the triangles to a catalog of triangles in a way that is faster and more successful than the angle method. Simulation results are provided that show the performance of the new method as well as its robustness with respect to false stars.

The Pyramid Star Identification Technique

ABSTRACT: A new highly robust algorithm, called Pyramid, is presented to identify the stars observed by star trackers in the general lost-in-space case, where no a priori estimate of pointing is available. At the heart of the method is the k-vector approach for accessing the star catalog, which provides a searchless means to obtain all cataloged stars from the whole sky that could possibly correspond to a particular measured pair, given the measured interstar angle and the measurement precision. The Pyramid logic is built on the identification of a four-star polygon structure—the Pyramid—which is associated with an almost certain star identification. Consequently, the Pyramid algorithm is capable of identifying and discarding even a high number of spikes (false stars). The method, which has already been tested in space, is demonstrated to be highly efficient, extremely robust, and fast. All of these features are supported by simulations and by a few ground test experimental results.

M-Type Vega-like Stars

We carried out a search for M-type Vega-like stars by correlating the IRAS Faint Source Catalog with Hipparcos-selected M-type stars. Three stars with apparent IRAS 25 μm excess emission are shown instead to be non–IR-excess stars from ground-based 11.7 and 17.9 μm photometry. Two stars previously suggested to have Vega-like mid-IR excess are also shown to be nonexcess stars. These results imply that other suggested mid–IR-excess stars in the literature may also be false excess stars. Detection threshold bias is apparently responsible for these bogus IR excesses. Sixty micron excess emission from a previously known M-type Vega-like star (GJ 803) is identified again.

Novel deblurring algorithms for images captured with CCD cameras

Weighted least-squares algorithms are proposed for reconstructing images from photon-limited data captured with CCD cameras. These cameras introduce blur and signal-independent noise, while forces of nature give rise to so-called background noise. The errors in the data are addressed by using a weighted least-squares formulation that emphasizes more reliable data values. The performance of the proposed algorithms is investigated by using synthetic and real data obtained from the Space Telescope Science Institute. The real data were captured with the wide-field/planetary camera aboard the original Hubble Space Telescope. Simulation studies with the synthetic data indicate that the proposed algorithms converge faster and have fewer “false stars” than a modified Richardson–Lucy algorithm.