scholarly journals Detection of Small Impact Craters via Semantic Segmenting Lunar Point Clouds Using Deep Learning Network

2021 ◽  
Vol 13 (9) ◽  
pp. 1826
Author(s):  
Yifan Hu ◽  
Jun Xiao ◽  
Lupeng Liu ◽  
Long Zhang ◽  
Ying Wang
Keyword(s):  
Image Data ◽  
Landing Site ◽  
Semantic Segmentation ◽  
Point Clouds ◽  
Impact Craters ◽  
Detection Accuracy ◽  
The Moon ◽  
Small Impact ◽  
Crater Detection ◽  
Deep Learning Network

Impact craters refer to the most salient features on the moon surface. They are of huge significance for analyzing the moon topography, selecting the lunar landing site and other lunar exploration missions, etc. However, existing methods of impact crater detection have been largely implemented on the optical image data, thereby causing them to be sensitive to the sunlight. Thus, these methods can easily achieve unsatisfactory detection results. In this study, an original two-stage small crater detection method is proposed, which is sufficiently effective in addressing the sunlight effects. At the first stage of the proposed method, a semantic segmentation is conducted to detect small impact craters by fully exploiting the elevation information in the digital elevation map (DEM) data. Subsequently, at the second stage, the detection accuracy is improved under the special post-processing. As opposed to other methods based on DEM images, the proposed method, respectively, increases the new crusher percentage, recall and crusher level F1 by 4.89%, 5.42% and 0.67%.

scholarly journals Trajectory Determination of Chang’E-5 during Landing and Ascending

2021 ◽  
Vol 13 (23) ◽  
pp. 4837
Author(s):  
Peng Yang ◽  
Yong Huang ◽  
Peijia Li ◽  
Siyu Liu ◽  
Quan Shan ◽  
...  
Keyword(s):  
Image Data ◽  
Landing Site ◽  
Polynomial Functions ◽  
The Moon ◽  
B Spline ◽  
Lunar Reconnaissance Orbiter ◽  
Orbiter Image ◽  
Positioning Method ◽  
Trajectory Determination

Chang’E-5 (CE-5) is China’s first lunar sample return mission. This paper focuses on the trajectory determination of the CE-5 lander and ascender during the landing and ascending phases, and the positioning of the CE-5 lander on the Moon. Based on the kinematic statistical orbit determination method using B-spline and polynomial functions, the descent and ascent trajectories of the lander and ascender are determined by using ground-based radiometric ranging, Doppler and interferometry data. The results show that a B-spline function is suitable for a trajectory with complex maneuvers. For a smooth trajectory, B-spline and polynomial functions can reach almost the same solutions. The positioning of the CE-5 lander on the Moon is also investigated here. Using the kinematic statistical positioning method, the landing site of the lander is 43.0590°N, 51.9208°W with an elevation of −2480.26 m, which is less than 200 m different from the LRO (Lunar Reconnaissance Orbiter) image data.

scholarly journals Semi-Supervised Deep Learning for Lunar Crater Detection Using CE-2 DOM

2021 ◽  
Vol 13 (14) ◽  
pp. 2819
Author(s):  
Sudong Zang ◽  
Lingli Mu ◽  
Lina Xian ◽  
Wei Zhang
Keyword(s):  
Deep Learning ◽  
Landing Site ◽  
Training Data ◽  
The Moon ◽  
Detection Model ◽  
Processing Times ◽  
Crater Detection ◽  
High Resolution Imagery ◽  
Model Training ◽  
Digital Orthophoto

Lunar craters are very important for estimating the geological age of the Moon, studying the evolution of the Moon, and for landing site selection. Due to a lack of labeled samples, processing times due to high-resolution imagery, the small number of suitable detection models, and the influence of solar illumination, Crater Detection Algorithms (CDAs) based on Digital Orthophoto Maps (DOMs) have not yet been well-developed. In this paper, a large number of training data are labeled manually in the Highland and Maria regions, using the Chang’E-2 (CE-2) DOM; however, the labeled data cannot cover all kinds of crater types. To solve the problem of small crater detection, a new crater detection model (Crater R-CNN) is proposed, which can effectively extract the spatial and semantic information of craters from DOM data. As incomplete labeled samples are not conducive for model training, the Two-Teachers Self-training with Noise (TTSN) method is used to train the Crater R-CNN model, thus constructing a new model—called Crater R-CNN with TTSN—which can achieve state-of-the-art performance. To evaluate the accuracy of the model, three other detection models (Mask R-CNN, no-Mask R-CNN, and Crater R-CNN) based on semi-supervised deep learning were used to detect craters in the Highland and Maria regions. The results indicate that Crater R-CNN with TTSN achieved the highest precision (of 91.4% and 88.5%, respectively) in the Highland and Maria regions, even obtaining the highest recall and F1 score. Compared with Mask R-CNN, no-Mask R-CNN, and Crater R-CNN, Crater R-CNN with TTSN had strong robustness and better generalization ability for crater detection within 1 km in different terrains, making it possible to detect small craters with high accuracy when using DOM data.

scholarly journals HYBRID GEOREFERENCING, ENHANCEMENT AND CLASSIFICATION OF ULTRA-HIGH RESOLUTION UAV LIDAR AND IMAGE POINT CLOUDS FOR MONITORING APPLICATIONS

2020 ◽  
Vol V-2-2020 ◽  
pp. 727-734
Author(s):  
N. Haala ◽  
M. Kölle ◽  
M. Cramer ◽  
D. Laupheimer ◽  
G. Mandlburger ◽  
...  
Keyword(s):  
Point Cloud ◽  
Image Data ◽  
Semantic Segmentation ◽  
Point Clouds ◽  
Airborne Lidar ◽  
Image Texture ◽  
Image Point ◽  
Joint Orientation ◽  
3D Point Clouds ◽  
3D Data

Abstract. This paper presents a study on the potential of ultra-high accurate UAV-based 3D data capture by combining both imagery and LiDAR data. Our work is motivated by a project aiming at the monitoring of subsidence in an area of mixed use. Thus, it covers built-up regions in a village with a ship lock as the main object of interest as well as regions of agricultural use. In order to monitor potential subsidence in the order of 10 mm/year, we aim at sub-centimeter accuracies of the respective 3D point clouds. We show that hybrid georeferencing helps to increase the accuracy of the adjusted LiDAR point cloud by integrating results from photogrammetric block adjustment to improve the time-dependent trajectory corrections. As our main contribution, we demonstrate that joint orientation of laser scans and images in a hybrid adjustment framework significantly improves the relative and absolute height accuracies. By these means, accuracies corresponding to the GSD of the integrated imagery can be achieved. Image data can also help to enhance the LiDAR point clouds. As an example, integrating results from Multi-View Stereo potentially increases the point density from airborne LiDAR. Furthermore, image texture can support 3D point cloud classification. This semantic segmentation discussed in the final part of the paper is a prerequisite for further enhancement and analysis of the captured point cloud.

scholarly journals A Deep Learning Semantic Segmentation-Based Approach for Field-Level Sorghum Panicle Counting

2019 ◽  
Vol 11 (24) ◽  
pp. 2939 ◽  
Author(s):  
Lonesome Malambo ◽  
Sorin Popescu ◽  
Nian-Wei Ku ◽  
William Rooney ◽  
Tan Zhou ◽  
...  
Keyword(s):  
Image Analysis ◽  
Deep Learning ◽  
High Throughput ◽  
Precision Agriculture ◽  
Crop Improvement ◽  
Image Data ◽  
Semantic Segmentation ◽  
Detection Accuracy ◽  
Phenotypic Data ◽  
Field Level

Small unmanned aerial systems (UAS) have emerged as high-throughput platforms for the collection of high-resolution image data over large crop fields to support precision agriculture and plant breeding research. At the same time, the improved efficiency in image capture is leading to massive datasets, which pose analysis challenges in providing needed phenotypic data. To complement these high-throughput platforms, there is an increasing need in crop improvement to develop robust image analysis methods to analyze large amount of image data. Analysis approaches based on deep learning models are currently the most promising and show unparalleled performance in analyzing large image datasets. This study developed and applied an image analysis approach based on a SegNet deep learning semantic segmentation model to estimate sorghum panicles counts, which are critical phenotypic data in sorghum crop improvement, from UAS images over selected sorghum experimental plots. The SegNet model was trained to semantically segment UAS images into sorghum panicles, foliage and the exposed ground using 462, 250 × 250 labeled images, which was then applied to field orthomosaic to generate a field-level semantic segmentation. Individual panicle locations were obtained after post-processing the segmentation output to remove small objects and split merged panicles. A comparison between model panicle count estimates and manually digitized panicle locations in 60 randomly selected plots showed an overall detection accuracy of 94%. A per-plot panicle count comparison also showed high agreement between estimated and reference panicle counts (Spearman correlation ρ = 0.88, mean bias = 0.65). Misclassifications of panicles during the semantic segmentation step and mosaicking errors in the field orthomosaic contributed mainly to panicle detection errors. Overall, the approach based on deep learning semantic segmentation showed good promise and with a larger labeled dataset and extensive hyper-parameter tuning, should provide even more robust and effective characterization of sorghum panicle counts.

Fluctuation of recent large impact craters rate on Mars from automatic crater detection

2020 ◽  
Author(s):  
Anthony Lagain ◽  
Misha Kreslavsky ◽  
Gretchen Benedix ◽  
David Baratoux ◽  
Phil Bland ◽  
...  
Keyword(s):  
Impact Crater ◽  
Impact Craters ◽  
Asteroid Belt ◽  
Large Impact ◽  
The Moon ◽  
Crater Detection ◽  
Impact Flux ◽  
Ejecta Blanket ◽  
The Impact ◽  
Latitudinal Band

<p>Knowledge of collision rates through time and space is essential because meteoritic impact crater counting is the only way to determine the ages of surface geological units and processes on the solid bodies of our Solar System. All chronology models assume a constant size distribution of impactors and an exponential decay of the impact flux between 4 Ga and 2.5 Ga before the present followed by a constant rate over the last 2.5 Ga. These two assumptions are challenged by recent evidence for an increase of the impact flux on the Moon and the Earth and probably on Mars associated with a decoupling between the flux of small and large impactors over the last billion years. Here, using the results of an automatic crater detection algorithm, we investigate the evolution of the rate of formation of large impact craters (Dc ≥ 20km) on Mars and thus infer the evolution of the flux of large impactors (Di > 5km) from the size-frequency distribution of small craters superposed to the ejecta blankets of large ones.</p><p>The dating of large impact craters on Mars is limited by several factors such as the degradation of ejecta blankets and the retention rate of small craters superposed to their ejecta. We therefore focused on craters ≥20km in diameter exhibiting an ejecta blanket according to the crater database and located on a latitudinal band between ±35°. We then selected those whom their ejecta are not affected by volcanic/tectonic processes or by the formation of another large nearby impact crater. The final set includes 590 impact craters.</p><p>If one can argue the impact flux cannot be fully recorded for the last 4Ga due to resurfacing processes erasing progressively the ejecta blanket and large craters themselves, Hesperian and Noachian terrains within the 35° latitudinal band should nevertheless have retained all D≥20km craters over a portion of the Amazonian period. The CSFD of craters younger than 600Ma (113 craters) superposed to these terrains is consistent with the 600Ma isochron, supporting the fact that the entire population of craters ≥20km formed over the last 600 million years on this portion of the Martian surface has been counted completely. We therefore focused on the analysis of the impact rate evolution over this range of time from this crater sub-sample.</p><p>The formation of large impact craters is not homogeneously distributed over the time range investigated here. Our data suggest an inconsistency between the flux used to date each crater and the rate inferred from these datings, thus implying that the small and large body impact fluxes are decoupled from one another. We note also sharp peaks centered around 480, 280 and 100Ma. Preliminary statistical test show that 280Ma peak is marginally significant whereas the two others are too small to be statistically significant. This pattern would be consistent with other independent arguments for increased rate with similar intensity and timing on the Moon and Mars for which the causes are probably collisions and potentially formation of asteroid families within the main asteroid belt.</p>

scholarly journals A global rockfall map of the Moon powered by AI and Big Data

2020 ◽  
Author(s):  
Valentin Bickel ◽  
Jordan Aaron ◽  
Andrea Manconi ◽  
Simon Loew ◽  
Urs Mall
Keyword(s):  
Spatial Distribution ◽  
High Resolution ◽  
Lunar Surface ◽  
Image Data ◽  
Landing Site ◽  
Planetary Science ◽  
The Moon ◽  
Lunar Orbiter ◽  
Narrow Angle ◽  
Lunar Reconnaissance Orbiter

<p>Under certain conditions, meter to house-sized boulders fall, jump, and roll from topographic highs to topographic lows, a landslide type termed rockfall. On the Moon, these features have first been observed in Lunar Orbiter photographs taken during the pre-Apollo era. Understanding the drivers of lunar rockfall can provide unique information about the seismicity and erosional state of the lunar surface, however this requires high resolution mapping of the spatial distribution and size of these features. Currently, it is believed that lunar rockfalls are driven by moonquakes, impact-induced shaking, and thermal fatigue. Since the Lunar Orbiter and Apollo programs, NASA’s Lunar Reconnaissance Orbiter Narrow Angle Camera (NAC) returned more than 2 million high-resolution (NAC) images from the lunar surface. As the manual extraction of rockfall size and location from image data is time intensive, the vast majority of NAC images have not yet been analyzed, and the distribution and number of rockfalls on the Moon remains unknown. Demonstrating the potential of AI for planetary science applications, we deployed a Convolutional Neural Network in combination with Google Cloud’s advanced computing capabilities to scan through the entire NAC image archive. We identified 136,610 rockfalls between 85°N and 85°S and created the first global, consistent rockfall map of the Moon. This map enabled us to analyze the spatial distribution and density of rockfalls across lunar terranes and geomorphic regions, as well as across the near- and farside, and the northern and southern hemisphere. The derived global rockfall map might also allow for the identification and localization of recent seismic activity on or underneath the surface of the Moon and could inform landing site selection for future geophysical surface payloads of Artemis, CLPS, or other missions. The used CNN will soon be available as a tool on NASA JPL’s Moon Trek platform that is part of NASA’s Solar System Treks (trek.nasa.gov/moon/).</p>

Small impact craters in the polar regions of the Moon: Peculiarities of morphometric characteristics

2015 ◽  
Vol 49 (5) ◽  
pp. 295-302 ◽  
Author(s):  
A. A. Kokhanov ◽  
M. A. Kreslavsky ◽  
I. P. Karachevtseva
Keyword(s):  
Impact Craters ◽  
Polar Regions ◽  
The Moon ◽  
Morphometric Characteristics ◽  
Small Impact
Sign in / Sign up

Export Citation Format

Share Document