AbstractAsphalt thin beams (112mm x 12.7mm x 6.35mm, length x width x thickness) in a Bending Beam Rheometer (BBR) have been used in the asphalt industry as one of methods to predict low temperature cracking properties of asphalt mixtures. Given the research work dedicated to the asphalt industry, the potential benefits of testing smaller specimens (cheaper equipment, less material, faster conditioning, easier availability for quality control, etc.) are well recognized. However, the two main criticisms of using the BBR to test mixtures are raised: (a) the thickness of the beam is smaller than the maximum aggregate size; thus a single aggregate particle can affect the results of the test, and (b) such small specimens cannot represent the overall property of the mix. This paper is conducted to address the above issues by using a combination of imaging techniques, statistical analysis, and viscoelastic modeling. Three nominal maximum aggregate sizes (NMAS), 12.5mm (1/2”), 9.5mm (3/8”) and 6.25mm (1/4”), were prepared in a laboratory. Each one of the three NMAS specimens was cut into a six-faced block that was then scanned and analyzed its pixels using imaging techniques. Subsequently, asphalt blocks were trimmed into thin beams and tested in a BBR to obtain their stiffness and creep compliance values.