Have a personal or library account? Click to login
Infrared Image Conversion from Grayscale to Temperature Using Linear Regression Cover

Infrared Image Conversion from Grayscale to Temperature Using Linear Regression

Open Access
|Dec 2025

Figures & Tables

Figure no. 1:

The infrared image with the legend of temperature
(Source: Matlab simulation)
The infrared image with the legend of temperature (Source: Matlab simulation)

Figure no. 2.a:

The evolution of the minimum and maximum temperature values
(Source: Matlab simulation)
The evolution of the minimum and maximum temperature values (Source: Matlab simulation)

Figure no. 2.b:

The evolution of the minimum temperature values without and with median filter
(Source: Matlab simulation)
The evolution of the minimum temperature values without and with median filter (Source: Matlab simulation)

Figure no. 3:

a. Legend of the infrared image; b. tempMin and tempMax directly from the IR image and with linear regression
(Source: Matlab simulation)
a. Legend of the infrared image; b. tempMin and tempMax directly from the IR image and with linear regression (Source: Matlab simulation)

Figure no. 4:

The temporal evolution of the minimum and maximum temperature values estimated using regression of the intermediate values from the IR image legend
(Source: Matlab simulation)
The temporal evolution of the minimum and maximum temperature values estimated using regression of the intermediate values from the IR image legend (Source: Matlab simulation)

Figure no. 5:

Graphical representation of the linear conversion from grayscale image to temperature
(Source: Matlab simulation)
Graphical representation of the linear conversion from grayscale image to temperature (Source: Matlab simulation)

Figure no. 6:

Top image: MSE for each frame; Bottom image: tempMin of each frame extracted from the bottom box of the legend of thermal image
(Source: Matlab simulation)
Top image: MSE for each frame; Bottom image: tempMin of each frame extracted from the bottom box of the legend of thermal image (Source: Matlab simulation)

Figure no. 7:

Top image: MSE for each frame; Bottom image: tempMin of each frame captured from the bottom box of the legend of thermal image, after median filtering
(Source: Matlab simulation)
Top image: MSE for each frame; Bottom image: tempMin of each frame captured from the bottom box of the legend of thermal image, after median filtering (Source: Matlab simulation)

Figure no. 8:

Minimum and maximum temperatures obtained with Method 1 with median filter and Method 2
(Source: Matlab simulation)
Minimum and maximum temperatures obtained with Method 1 with median filter and Method 2 (Source: Matlab simulation)

Figure no. 9:

The evolution of MSE (not normalized)
(Source: Matlab simulation)
The evolution of MSE (not normalized) (Source: Matlab simulation)

The mean values for MSE and R2

Method to determine the tempMin and tempMaxMSEvideoRvideo2R_{video}^2
Method 1 (ch. 3.1.a) without median filter0.05620.9668
Method 1 (ch. 3.1.a) with median filter0.04900.9710
Method 2 (ch. 3.1.b)0.02380.9861
DOI: https://doi.org/10.2478/bsaft-2025-0023 | Journal eISSN: 3100-5098 | Journal ISSN: 3100-508X
Language: English
Page range: 229 - 237
Published on: Dec 16, 2025
In partnership with: Paradigm Publishing Services
Publication frequency: 2 issues per year

© 2025 Cătălina NEGHINĂ, Annamaria SÂRBU, Mihai NEGHINĂ, published by Nicolae Balcescu Land Forces Academy
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 3.0 License.