Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userAutomated crater detection with human level performance
66
0
0.0
(
0
)
Ask ChatGPT about the research
No Arabic abstract
Crater cataloging is an important yet time-consuming part of geological mapping. We present an automated Crater Detection Algorithm (CDA) that is competitive with expert-human researchers and hundreds of times faster. The CDA uses multiple neural networks to process digital terrain model and thermal infra-red imagery to identify and locate craters across the surface of Mars. We use additional post-processing filters to refine and remove potential false crater detections, improving our precision and recall by 10% compared to Lee (2019). We now find 80% of known craters above 3km in diameter, and identify 7,000 potentially new craters (13% of the identified craters). The median differences between our catalog and other independent catalogs is 2-4% in location and diameter, in-line with other inter-catalog comparisons. The CDA has been used to process global terrain maps and infra-red imagery for Mars, and the software and generated global catalog are available at https://doi.org/10.5683/SP2/CFUNII.
rate research
Read More
Crater ellipticity determination is a complex and time consuming task that so far has evaded successful automation. We train a state of the art computer vision algorithm to identify craters in Lunar digital elevation maps and retrieve their sizes and 2D shapes. The computational backbone of the model is MaskRCNN, an instance segmentation general framework that detects craters in an image while simultaneously producing a mask for each crater that traces its outer rim. Our post-processing pipeline then finds the closest fitting ellipse to these masks, allowing us to retrieve the crater ellipticities. Our model is able to correctly identify 87% of known craters in the longitude range we hid from the network during training and validation (test set), while predicting thousands of additional craters not present in our training data. Manual validation of a subset of these new craters indicates that a majority of them are real, which we take as an indicator of the strength of our model in learning to identify craters, despite incomplete training data. The crater size, ellipticity, and depth distributions predicted by our model are consistent with human-generated results. The model allows us to perform a large scale search for differences in crater diameter and shape distributions between the lunar highlands and maria, and we exclude any such differences with a high statistical significance. The predicted test set catalogue and trained model are available here: https://github.com/malidib/Craters_MaskRCNN/.
The identification of impact craters on planetary surfaces provides important information about their geological history. Most studies have relied on individual analysts who map and identify craters and interpret crater statistics. However, little work has been done to determine how the counts vary as a function of technique, terrain, or between researchers. Furthermore, several novel internet-based projects ask volunteers with little to no training to identify craters, and it was unclear how their results compare against the typical professional researcher. To better understand the variation among experts and to compare with volunteers, eight professional researchers have identified impact features in two separate regions of the moon. Small craters (diameters ranging from 10 m to 500 m) were measured on a lunar mare region and larger craters (100s m to a few km in diameter) were measured on both lunar highlands and maria. Volunteer data were collected for the small craters on the mare. Our comparison shows that the level of agreement among experts depends on crater diameter, number of craters per diameter bin, and terrain type, with differences of up to $simpm45%$. We also found artifacts near the minimum crater diameter that was studied. These results indicate that caution must be used in most cases when interpreting small variations in crater size-frequency distributions and for craters $le10$ pixels across. Because of the natural variability found, projects that emphasize many people identifying craters on the same area and using a consensus result are likely to yield the most consistent and robust information.
Nuclear pleomorphism, defined herein as the extent of abnormalities in the overall appearance of tumor nuclei, is one of the components of the three-tiered breast cancer grading. Given that nuclear pleomorphism reflects a continuous spectrum of variation, we trained a deep neural network on a large variety of tumor regions from the collective knowledge of several pathologists, without constraining the network to the traditional three-category classification. We also motivate an additional approach in which we discuss the additional benefit of normal epithelium as baseline, following the routine clinical practice where pathologists are trained to score nuclear pleomorphism in tumor, having the normal breast epithelium for comparison. In multiple experiments, our fully-automated approach could achieve top pathologist-level performance in select regions of interest as well as at whole slide images, compared to ten and four pathologists, respectively.
The NASA Dawn mission has extensively examined the surface of asteroid Vesta, the second most massive body in the main belt. The high quality of the gathered data provides us with an unique opportunity to determine the surface and internal properties of one of the most important and intriguing main belt asteroids (MBAs). In this paper, we focus on the size frequency distributions (SFDs) of sub-kilometer impact craters observed at high spatial resolution on several selected young terrains on Vesta. These small crater populations offer an excellent opportunity to determine the nature of their asteroidal precursors (namely MBAs) at sizes that are not directly observable from ground-based telescopes (i.e., below ~100 m diameter). Moreover, unlike many other MBA surfaces observed by spacecraft thus far, the young terrains examined had crater spatial densities that were far from empirical saturation. Overall, we find that the cumulative power-law index (slope) of small crater SFDs on Vesta is fairly consistent with predictions derived from current collisional and dynamical models down to a projectile size of ~10 m diameter (Bottke et al., 2005a,b). The shape of the impactor SFD for small projectile sizes does not appear to have changed over the last several billions of years, and an argument can be made that the absolute number of small MBAs has remained roughly constant (within a factor of 2) over the same time period. The apparent steady state nature of the main belt population potentially provides us with a set of intriguing constraints that can be used to glean insights into the physical evolution of individual MBAs as well as the main belt as an ensemble.
The Geostationary Lightning Mapper (GLM) instrument onboard the GOES 16 and 17 satellites has been shown to be capable of detecting bolides (bright meteors) in Earths atmosphere. Due to its large, continuous field of view and immediate public data availability, GLM provides a unique opportunity to detect a large variety of bolides, including those in the 0.1 to 3 m diameter range and complements current ground-based bolide detection systems, which are typically sensitive to smaller events. We present a machine learning-based bolide detection and light curve generation pipeline being developed at NASA Ames Research Center as part of NASAs Asteroid Threat Assessment Project (ATAP). The ultimate goal is to generate a large catalog of calibrated bolide lightcurves to provide an unprecedented data set which will be used to inform meteor entry models on how incoming bodies interact with the Earths atmosphere and to infer the pre-entry properties of the impacting bodies. The data set will also be useful for other asteroidal studies. This paper reports on the progress of the first part of this ultimate goal, namely, the automated bolide detection pipeline. Development of the training set, ML model training and iterative improvements in detection performance are presented. The pipeline runs in an automated fashion and bolide lightcurves along with other measured properties are promptly published on a NASA hosted publicly accessible website, https://neo-bolide.ndc.nasa.gov.
suggested questions
Log in to be able to interact and post comments
comments
Fetching comments
Sign in to be able to follow your search criteria
