Survey Registration for Long‐Term Natural Environment Monitoring

Abstract

This paper presents a survey registration framework to assist in the recurrent inspection of a natural environment. Our framework coarsely aligns surveys at the image‐level using visual simultaneous localization and mapping (SLAM), and it registers images at the pixel‐level using SIFT Flow, which enables rapid manual inspection. The variation in appearance of natural environments makes data association a primary challenge of this work. We discuss this and other challenges, including 1) alternative approaches for coarsely aligning surveys of a natural environment, 2) how to select which images to compare between two surveys, and 3) strategies to boost image registration accuracy. We evaluate each stage of our approach, emphasizing alignment accuracy and stability with respect to large seasonal variations. Our domain is lakeshore monitoring, in which an autonomous surface vessel surveyed a 1‐km lakeshore 33 times in 14 months. Our results show that our framework precisely aligns a significant number of images between surveys captured up to roughly three months apart, often across marked variation in appearance. Using these results, a human was able to spot several changes between surveys that would have otherwise gone unnoticed.

Image features for visual teach-and-repeat navigation in changing environments

“…Unlike SpG/CNN, which is designed for general use, the STAR/GRIEF combination is not meant to handle large viewpoint changes and one should be cautious when applying it for general long-term navigation and localisation. For example, [61] evaluated the performance of several image features in a scenario of lakeshore monitoring, where the on-board camera aims perpendicularly to the vehicle movement and thus, the viewpoint changes are significant. The authors of [61] concluded that in their scenario, the ORB feature, which is based on BRIEF, slightly outperformed the other features.…”

“…For example, [61] evaluated the performance of several image features in a scenario of lakeshore monitoring, where the on-board camera aims perpendicularly to the vehicle movement and thus, the viewpoint changes are significant. The authors of [61] concluded that in their scenario, the ORB feature, which is based on BRIEF, slightly outperformed the other features.…”

A survey on Visual-Based Localization: On the benefit of heterogeneous data

“…However, Valgren and Lilienthal [205] have shown that these representations are not well suited for similarity association across season cycles. GRIEF local descriptor [87] (derivatives of BRIEF [24]) or ORB feature [61] show better results for this task. The use of heterogeneous databases (i.e.…”

“…Desc. Used in VBL Pseudo Corners detector [129] Point [129,130] Hessian-affine [126] Point [71,2,100,173,4] FALoG [216] Point [49] SIFT [109] Point [188,181,103,223,224,140,220] RootSIFT [3] Point [125,4,173,199,200] SURF [18] Point [44,100,190,158,188,205] ORB [169] Point [61] BRIEF [24] Point [90,87] BRISK [94] Point [49,125,133]…”

MallARD: An Autonomous Aquatic Surface Vehicle for Inspection and Monitoring of Wet Nuclear Storage Facilities

“…To date, a range of ASVs have been developed [6][7][8][9][10], however, most are large vehicles that have been designed for operation in marine environments and lakes [7,8,11]. They rely on unsuitable sensor packages and lack the control accuracy required for operation in SFPs or wet silos.…”