This dataset comprises digital photographs taken at the Astrophysical Scattering Laboratory of the Department of Physics of the University of Helsinki, Finland. These photographs were taken in a darkened room, with targets illuminated by an Energetiq Laser-Driven Light Source (LDLD) EQ-77-QZ lamp module. We utilized three distinct stones borrowed from the 2024 course materials of the Geological Materials course at the University of Helsinki's Department of Geosciences and Geography. These stones vary in shape and reflectivity: one is regularly shaped with a smooth surface, another has an irregular shape but a smooth surface with high reflectivity, and the third is irregularly shaped with varying reflectivity across its surface. The stones were photographed from a distance of 230 cm at six different camera positions. At each camera position, the setting with the target was rotated. The images were gathered at 10-degree intervals, totaling 36 images for each stone at each camera position and 216 images per stone in total. The dataset consists of 648 unedited 16-bit Canon CR2 raw photos and the same 648 images cropped, color-graded, and processed into TIFF format. The TIFF data package is 625 MB, while the raw data package is 18.8 GB. Further details on data acquisition can be found in the accompanying PDF files. This dataset serves as a resource for testing and benchmarking various image processing and inverse method tasks, such as reconstructing object shapes using convolutional neural networks and inverse methods. It aims to aid researchers in image processing and related endeavors. Please note: Depending on your graphics editor, the program might automatically apply an adjustment of saturation, brightness, and contrast to the raw CR2 photos. In this case, the desired outcome of only the target being illuminated in the photo will be lost. Please ensure that your graphics editor does not apply automatic adjustments, or use the provided TIFF files.