Scene-based nonuniformity correction with homography transformation

2025-03-04

AI Summary

Uncooled focal plane arrays (UC-FPAs) mounted on UAVs suffer from bias/gain drift in agricultural remote sensing, necessitating frequent blackbody-based calibration—a logistical bottleneck in field operations. Method: This paper proposes a scene-driven non-uniformity correction (NUC) method that eliminates the need for physical calibration sources. It models UAV hovering motion as a homography transformation and leverages multi-frame geometric displacement constraints to formulate an end-to-end joint optimization framework that simultaneously estimates pixel-wise gain, bias, and the underlying true temperature image. The method integrates generalized Lucas-Kanade registration with alternating optimization based on maximum likelihood estimation and incorporates blind deconvolution principles to enhance robustness against motion blur and noise. Results: Simulation results demonstrate a Pearson correlation coefficient exceeding 0.9999998 and equivalent temperature recovery error below 0.01 °C—significantly outperforming conventional scene-based NUC methods.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Computer Vision: Scene Analysis & Understanding
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

Due to their affordable, low mass, and small dimensions, uncooled microbolometer-based thermal focal plane arrays (UC-FPAs) are useful for long-wave infrared (LWIR)imaging applications. However, in outdoor conditions typical in agricultural remote sensing, cameras based on UC-FPAs may suffer from drift in offset and gain. To tackle the persistent drift, the system requires continuous calibration. Our goal in this study was to eliminate this requirement via a computational schema. In a former study, we estimated unknown gain and offset values and thermographic images of an object from a sequence of pairs of successive images taken at two different blur levels.In the current work, we took on a similar problem using a sequence of shifted images, with relative shifts caused by realistic drone hovering modeled by homography transformation. This places our work in the realm of scene-based nonuniformity correction problems. We show that an object's thermographic values, as well as gain and offset, can be jointly estimated by relying on a few sets of shifted images. We use a minimum likelihood estimator, which is found using alternating minimization. Registration is done using a generalized Lucas-Kanade method. Simulations show promising accuracy with mean Pearson correlation of more than 0.9999998 between ground truth and restoration. Under ideal assumptions, this is equivalent to a mean restoration error of less than 0.01 Celsius degree.

Problem

• Addresses drift in offset and gain in UC-FPA cameras.
• Proposes scene-based nonuniformity correction using shifted images.
• Estimates thermographic values and calibration parameters jointly.

Innovation

• Scene-based nonuniformity correction using homography transformation
• Joint estimation of thermographic values, gain, and offset
• Alternating minimization for minimum likelihood estimator