Citizen science is an invaluable tool, with photographic observations and recorded measurements providing data on species across broader spatial and temporal scales than can be achieved through fieldwork alone. However, citizen-science images often lack size references, meaning length data often cannot be validated or extracted from photographs, limiting the amount of data available for analyses used to improve our understanding of species biology and inform management. Using Bayesian log-linear models, we tested whether full body lengths of school sharks (Galeorhinus galeus) could be estimated from ratios between pre-first dorsal (e.g., pre-branchial and head length) and eye length measurements extracted from images. Measured full body lengths fell within the prediction intervals of lengths estimated from ratios extracted from images. This approach provides a preliminary method for validating and extracting length measurements from citizen science observations of sharks, increasing the utility of public contributions to shark research.