Imaging techniques for internal visualization of materials have become increasingly important in a wide range of industrial fields. Imaging techniques such as magnetic resonance imaging (MRI) and computed tomography (CT) employing X-rays have existed for quite some time. However, such classic techniques have physical limitations (for e.g., X-rays cannot penetrate high-density materials such as metals) because they were developed as medical applications. Therefore, alternative techniques for industrial fields should be explored. Photothermal imaging was explored from the perspective of visualization of the interior of metals as photothermal effects depend on the thermal properties of the materials. To establish photothermal imaging as a reliable technique, it is important to analyze the physical limitations and noise in the measurement. Therefore, in this study, the conditions leading to a minimum level of noise were explored using aluminum 6061 specimens of five surface roughness values under conditions of different pump beam frequencies and radii. In this work, surface roughness and pump beam frequency and radius were considered important parameters affecting photothermal imaging. Finally, comprehensive criteria to reduce signal noise in photothermal imaging were proposed.