Edge AI versus satellite remote sensing for the detection of environmental crimes: systematic review and multi-criteria analysis TOPSIS

Vol 57, 2025 - 339813
Poster
Favorite this paper
How to cite this paper?
Abstract

Automated detection of environmental crimes in the Amazon entails trade-offs among coverage, latency, deployment cost, scalability and robustness. The three approaches (Edge AI / IoT, satellite remote sensing with deep learning, and hybrid configurations) are compared through a PRISMA 2020 systematic review over four databases and a multicriteria TOPSIS analysis of 44 primary studies, published between 2020 and 2026, under five criteria and four weight scenarios. No approach dominates across the five criteria, and the ranking reverses with the scenario: under equal weights Edge AI leads (0.486), followed by hybrid (0.472) and satellite (0.405), with no detectable difference (Kruskal-Wallis, p = 0.111); satellite leads when effectiveness carries 40 % of the weight. The choice therefore depends on the decision maker's operational priority, and the evidence points to combination: the four hybrid studies retain edge latency (4.25) and recover satellite robustness (4.00), the two criteria where the approaches most oppose each other.

Share your ideas or questions with the authors!

Did you know that the greatest stimulus in scientific and cultural development is curiosity? Leave your questions or suggestions to the author!

Sign in to interact

Have a question or suggestion? Share your feedback with the authors!

Institutions
  • 1 Pontifícia Universidade Católica do Paraná
Track
  • MCD – Multicriteria Decision Support Methods
Keywords
Multicriteria analysis.
Automated detection of environmental crimes.
Systematic review.