Abstract:IR cameras are widely used for temperature measurements in various applications, including agriculture, medicine, and security. Low-cost IR cameras have the immense potential to replace expensive radiometric cameras in these applications; however, low-cost microbolometer-based IR cameras are prone to spatially variant nonuniformity and to drift in temperature measurements, which limit their usability in practical scenarios. To address these limitations, we propose a novel approach for simultaneous temperature estimation and nonuniformity correction (NUC) from multiple frames captured by low-cost microbolometer-based IR cameras. We leverage the camera's physical image-acquisition model and incorporate it into a deep-learning architecture termed kernel prediction network (KPN), which enables us to combine multiple frames despite imperfect registration between them. We also propose a novel offset block that incorporates the ambient temperature into the model and enables us to estimate the offset of the camera, which is a key factor in temperature estimation. Our findings demonstrate that the number of frames has a significant impact on the accuracy of the temperature estimation and NUC. Moreover, introduction of the offset block results in significantly improved performance compared to vanilla KPN. The method was tested on real data collected by a low-cost IR camera mounted on an unmanned aerial vehicle, showing only a small average error of $0.27-0.54^\circ C$ relative to costly scientific-grade radiometric cameras. Our method provides an accurate and efficient solution for simultaneous temperature estimation and NUC, which has important implications for a wide range of practical applications.

What problem does this paper attempt to address?

The paper aims to address the issues of non-uniformity and drift in temperature measurement using low-cost infrared (IR) cameras. Specifically, low-cost microbolometer-based IR cameras, while advantageous in terms of cost and power consumption, suffer from spatial variation non-uniformity and temperature measurement drift, which limits their usability in practical applications. To solve these problems, the paper proposes a novel method that simultaneously performs temperature estimation and non-uniformity correction (NUC) from multiple frames. This method leverages the physical imaging model of the camera and integrates it into a deep learning architecture called Kernel Prediction Network (KPN), enabling the combination of information even if the registration between multiple frames is imperfect. Additionally, the paper introduces a new offset block that incorporates ambient temperature into the model to estimate the camera's offset, which is a critical factor for temperature estimation. The study results indicate that the number of frames significantly impacts the accuracy of temperature estimation and non-uniformity correction. With the introduction of the offset block, there is a significant performance improvement compared to the baseline version of KPN. Testing on real datasets shows that the proposed method achieves an average error of only 0.27 to 0.54 degrees Celsius compared to expensive scientific-grade radiometric cameras. In summary, the method proposed in this paper provides an accurate and efficient solution for simultaneous temperature estimation and non-uniformity correction from multiple frames, which is of great significance for a wide range of practical applications such as agriculture, medicine, and security.

Simultaneous temperature estimation and nonuniformity correction from multiple frames

A Novel Infrared Focal Plane Non-Uniformity Correction Method Based on Co-Occurrence Filter and Adaptive Learning Rate

Estimating temperatures with low-cost infrared cameras using deep neural networks

An Efficient Temperature Calibration Method Based on the Improved Infrared Forward Model and Bayesian Inference

Scene-Based Nonuniformity Correction Method Using Principal Component Analysis for Infrared Focal Plane Arrays

Single-image-based Nonuniformity Correction of Uncooled Long-Wave Infrared Detectors: a Deep-Learning Approach.

Adaptive Neural Network Non-uniformity Correction Algorithm for Infrared Focal Plane Array Based on Bi-Exponential Edge-Preserving Smoother

A Non-Uniformity Correction Algorithm of Infrared Image Sequences Based on Constant-Statistics

A Soft Calibration Technique for Infrared Imaging Based on Physical Model of Thermal Radiation and Variational Bayesian Inference from Small Dataset

Nonuniformity correction based on focal plane array temperature in uncooled long-wave infrared cameras without a shutter

On-the-fly Extrinsic Calibration of Multimodal Sensing System for Fast 3D Thermographic Scanning

A Multi-Scale Non-Uniformity Correction Method Based on Wavelet Decomposition and Guided Filtering for Uncooled Long Wave Infrared Camera.

Temperature-Robust Learned Image Recovery for Shallow-Designed Imaging Systems

A Calibration Procedure for Field and UAV-Based Uncooled Thermal Infrared Instruments

Depth and Thermal Sensor Fusion to Enhance 3D Thermographic Reconstruction

A calibration and prediction method of the camera imaging parameters in variable temperature environment

Spatially Adaptive Column Fixed-Pattern Noise Correction in Infrared Imaging System Using 1D Horizontal Differential Statistics

Incremental Low Rank Noise Reduction for Robust Infrared Tracking of Body Temperature during Medical Imaging

Online Photometric Calibration of Automatic Gain Thermal Infrared Cameras

InfraNet: Accurate forehead temperature measurement framework for people in the wild with monocular thermal infrared camera

Factors Influencing Temperature Measurements from Miniaturized Thermal Infrared (TIR) Cameras: A Laboratory-Based Approach