Planetary rover systems need to perform terrain segmentation to identify feasible driving areas and surround obstacles, which falls into the research area of semantic segmentation. Recently, deep learning (DL)-based methods were proposed and achieved great performance for semantic segmentation. However, due to the on-board processor platform’s strict comstraints on computational complexity and power consumption, existing DL approaches are almost impossible to be deployed on satellites under the burden of extensive computation and large model size. To fill this gap, this paper targeted studying effective and efficient Martian terrain segmentation solutions that are suitable for on-board satellites. In this article, we propose a lightweight ViT-based terrain segmentation method, namely, SegMarsViT. In the encoder part, the mobile vision transformer (MViT) block in the backbone extracts local–global spatial and captures multiscale contextual information concurrently. In the decoder part, the cross-scale feature fusion modules (CFF) further integrate hierarchical context information and the compact feature aggregation module (CFA) combines multi-level feature representation. Moreover, we evaluate the proposed method on three public datasets: AI4Mars, MSL-Seg, and S5Mars. Extensive experiments demonstrate that the proposed SegMarsViT was able to achieve 68.4%, 78.22%, and 67.28% mIoU on the AI4Mars-MSL, MSL-Seg, and S⁵Mars, respectively, under the speed of 69.52 FPS.

The limited resource of Geostationary Earth Orbit (GEO) is precious and most telecommunication, weather and navigational satellites are placed in this orbit. In order to guarantee the safety and health of active satellites, advanced surveillance and warning of unknown space targets such as space debris are crucial. However, space object detection still remains a very challenging problem because of the weak target characteristics and complex star background. To solve this problem, we conduct a deep-learning-based framework called PP-YOLOv2 for single-frame object detection and design a post-processing algorithm named CFS for further candidate filtration and supplement. First, we transform the label information and generate the according bounding boxes to train the PP-YOLOv2 detector to extract candidate coordinates for each frame. Then, the CFS technique is designed as an effective post-processing procedure to obtain the eventual prediction results. Experiments were conducted over a dataset from the Kelvins SpotGEO challenge, which demonstrate the effectiveness and the comparable detection performance of our proposed pipeline. Finally, the deployment results on NVIDIA Jetson Nano show that the proposed method has a competitive application prospect for a space target monitoring system.

Convolutional neural networks (CNNs) have achieved promising results in many tasks, and evaluating the model’s generalization ability based on the trained model and training data is paramount for practical applications. Although many measures for evaluating the generalization of CNN models have been proposed, the existing works are limited to small-scale or simplified model sets, which would result in poor accuracy and applicability of the derived methods. This study addresses these limitations by leveraging ResNet models as a case study to evaluate the model’s generalization ability. We utilized Intersection over Union (IoU) as a method to quantify the ratio of task-relevant features to assess model generalization. Class activation maps (CAMs) were used as a representation of the distribution of features learned by the model. To systematically investigate the generalization ability, we constructed a diverse model set based on the ResNet architecture. A total of 2000 CNN models were trained on the ImageNet subset by systematically changing commonly used hyperparameters. The results of our experiments revealed a strong correlation between the IoU-based evaluation method and the model’s generalization performance (Pearson correlation coefficient more than 0.8). We also performed extensive experiments to demonstrate the feasibility and robustness of the evaluation methods.

Identifying the surface topographic changes accurately plays a vital role in the task of planetary exploration. In this study, a deeply-supervised lightweight network (MViT-PCD) was proposed for monitoring dynamic surface changes using bi-temporal planetary images. The mobile vision transformer (MobileViT) was first introduced to make the most of the spatial information available. Subsequently, a multiscale feature differentiation and fusion (MFDF) block was adopted to improve the distinguishability of multi-level contextual information. Moreover, the strategy of information maximization (IM) was integrated to refine the model performance on the heterogeneous dataset. Then, experiments were conducted on the public Martian datasets. Compared with other state-of-the-art methods, MViT-PCD provides favorable performance, with the highest accuracy of 97.2% and 82.9%, respectively, under the speed of 43.4 FPS. Code is available at https://github.com/ lynn1023-max/MViT-PCD/.

Addressing the challenges in online multi-object tracking algorithms under complex scenarios, where the independence among feature extraction, object detection, and data association modules leads to both error accumulation and the difficulty of maintaining visual consistency for occluded objects, we have proposed an end-to-end multi-object tracking method based on hypergraph matching (JDTHM). Initially, a feature extraction and object detection module is introduced to achieve preliminary localization and description of the objects. Subsequently, a deep feature aggregation module is designed to extract temporal information from historical tracklets, amalgamating features from object detection and feature extraction to enhance the consistency between the current frame features and the tracklet features, thus preventing identity swaps and tracklet breaks caused by object detection loss or distortion. Finally, a data association module based on hypergraph matching is constructed, integrating with object detection and feature extraction into a unified network, transforming the data association problem into a hypergraph matching problem between the tracklet hypergraph and the detection hypergraph, thereby achieving end-to-end model optimization. The experimental results demonstrate that this method has yielded favorable qualitative and quantitative analysis results on three multi-object tracking datasets, thereby validating its effectiveness in enhancing the robustness and accuracy of multi-object tracking tasks.

The hyperspectral image compression scheme is a trade-off between the limited hardware resources of the on-board platform and the ever-growing resolution of the optical instruments. Predictive coding attracts researchers due to its low computational complexity and moderate memory requirements. We propose a near-lossless prediction-based compression scheme that removes spatial and spectral redundant information, thereby significantly reducing the size of hyperspectral images. This scheme predicts the target pixel’s value via a linear combination of previous pixels. The weight matrix of the predictor is iteratively updated using a recursive least squares filter with a loop quantizer. The optimal number of bands for prediction was analyzed experimentally. The results indicate that the proposed scheme outperforms state-of-the-art compression methods in terms of the compression ratio and quality retrieval.