Pathway Detection Using Fusion Polarization Features in Passive Millimeter-Wave Imaging

Pathway detection holds extreme significance for fields such as aircraft landing, autonomous driving, and terrain mapping. Passive millimeter-wave (PMMW) imaging technology, with its all-weather operational capability and excellent penetration through fog, and clouds, has shown great potential in pathway detection. However, most existing studies focus on the brightness temperature (TB) differences for detection, which is highly susceptible to interference from environmental radiation. Polarization features can effectively characterize the object and its environmental properties. This paper proposes a physics-based pathway detection method that utilizes feature-level fusion technology, fusing the polarization features of the first two Stokes parameters <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">T</i><italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">_I</i> and <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">T_Q</i>, as well as the degree of linear polarization (DoLP) and the angle of polarization (AoP). By introducing spatial similarity (SS), the method effectively excludes interference from non-pathway areas and improves the detection accuracy. After confirming the horizon position as the vanishing point of the pathway, the detection results are fused with a region growing algorithm to generate the final pathway extraction output. Experimental results demonstrate that this method can distinguish between pathway and non-pathway regions and accurately detect various types of pathways, such as rivers, roads and seas. Quantitative analysis shows that, compared to existing methods, the proposed method has significant advantages in terms of detection accuracy and robustness.