A novel algorithm and hardware architecture for fast video-based shape reconstruction of space debris

Abstract

In order to enable the non-cooperative rendezvous, capture, and removal of large space debris, automatic recognition of the target is needed. Video-based techniques are the most suitable in the strict context of space missions, where low-energy consumption is fundamental, and sensors should be passive in order to avoid any possible damage to external objects as well as to the chaser satellite. This paper presents a novel fast shape-from-shading (SfS) algorithm and a field-programmable gate array (FPGA)-based system hardware architecture for video-based shape reconstruction of space debris. The FPGA-based architecture, equipped with a pair of cameras, includes a fast image pre-processing module, a core implementing a feature-based stereo-vision approach, and a processor that executes the novel SfS algorithm. Experimental results show the limited amount of logic resources needed to implement the proposed architecture, and the timing improvements with respect to other state-of-the-art SfS methods. The remaining resources available in the FPGA device can be exploited to integrate other vision-based techniques to improve the comprehension of debris model, allowing a fast evaluation of associated kinematics in order to select the most appropriate approach for capture of the target space debris.

High-Performance Embedded Computing in Space: Evaluation of Platforms for Vision-Based Navigation

“…Sparse 3-D point clouds recovered with stereo have been employed to estimate pose, for example in [46], where the 3-D points originate from natural keypoints, and [47], which determines image points as the intersections of detected line segments. To achieve more dense reconstruction, Carlo et al [48] suggest the combination of conventional feature-based stereo with shape from shading. It is noted that all stereo approaches still need to solve the problem of establishing correspondences between the 3-D data and the object model.…”

Hardware Implementation and Validation of 3D Underwater Shape Reconstruction Algorithm Using a Stereo-Catadioptric System

“…These methods, despite their good detection rate, necessitate big computation time since they use complex calculations (convolution with a Gaussian, gradient vectors, descriptors, etc.). Therefore, some authors propose different alternatives that permit to substitute global methods with a combination of standard processing methods such as those presented in [13,36,40] which permits to ensure good accuracy and run time.…”

“…For example, the capturing sensor constitutions and configurations present a great interest to facilitate the image processing algorithms. For underwater applications, with the use of passive sensors [2,36], the identification and recognition of object is tough and sometimes hopeless because of the underground homogenous texture and lack of illumination. Atmospheric conditions that affect the optical images at different acquisition times could be a potential source of error and they should be taken into consideration to avoid missed detection or false-offs.…”

“…In the other hand, when the configuration of the different sensors is less complex, the geometric modelisation (calibration) of the whole system becomes possible and more accurate. This allows us to define the properties of the epipolar geometry (calibration parameters) that could be used afterwards to enhance the different image processing, particularly to remove the outliers [5,6] facilitate the matching process using epipolar lines [5,36] and obviously for 3D triangulation [5,6]. Furthermore, methods such as SURF detector [6] or Harris corner detector [5] or Sobel edge detector [2] were used to detect most interesting points in the image.…”

An FPGA-Based Real-Time System for 3D Stereo Matching, Combining Absolute Differences and Census with Aggregation and Belief Propagation

“…This problem has been widely researched by the computer vision community and appears not only in stereo vision but in other image processing topics as well such as optical flow calculation [2]. The range of applications of 3D stereo vision cannot be underestimated, with new fields of application emerging continuously, such as in recent research on shape reconstruction of space debris [3].…”