This report provides a summary of insights and evidence from prior research published in the public domain on the topic of automated and semi-automated image-based inspection technologies focused on train components and parts. Extra effort has been made to consider factors such as changing lighting and weather conditions to represent realistic operational environments. One of the main performance limiting aspects of image-based inspection systems was reported to be related to adequate and uniform illumination. Based on several studies, incorporation of artificial lighting sources in an inspection system is a common solution to the lighting problem of image-based inspections. Furthermore, the accumulation of snow on train components or inspection cameras is another element that can hinder image-based inspections. Mitigating approaches for clearing snow such as the use of heaters on camera lenses are needed. Other factors requiring special attention in image-based inspections were reported to be calibration of cameras, lens correction for radial distortion and perspective effects, temporal synchronization of data streams from multiple cameras, and spatial alignment of images. Moreover, for any inspections performed solely by machines, an adequate database is needed for training the flaw detection algorithms. Also, noise in the data needs to be addressed when building and training robust algorithms. Finally, after a review of a limited number of images collected by existing cameras on railway lines, it was found that such data can be used to identify some broken or defective components underneath the moving train cars. However, currently the image-based inspections may not be viable for finding certain defects such as leaks.