The complexity of recently developed iron ore deposits is increasing in comparison with those developed only 20-30 years ago. The deposits previously considered as not economically viable, due to the presence of higher levels of gangue minerals, are now starting to be developed.To optimise processing/beneficiation procedures a detailed characterisation of such ores is needed, including mineral liberation, association and textural classification. Identification of different iron oxides and oxyhydroxides is already reliably performed by optical image analysis (OIA). Automated OIA identification of different gangue materials, particularly quartz, can be problematic though.The article demonstrates the capability of CSIRO OIA software Mineral4/Recognition4 to characterise low grade iron ores. Such characterisation includes identification of different types of goethite, hydrohematite and gangue materials such as quartz and kaolinite. XRD and XRF analysis results are compared with those from OIA. The article also discusses the peculiarities of textural classification for such ores. Correlation of these results and visual comparison shows that optical image analysis can be an effective tool for characterisation of low and medium grade iron ores. The work highlights issues regarding discrimination of aluminous goethite and gangue, micro and nano-porosity and effective density, for further study. CITATION:Donskoi, E, Manuel, J R, Hapugoda, S, Poliakov, A, Raynlyn, T, Austin, P and Peterson, M, 2019. Automated optical image analysis of lower grade iron ores, in Proceedings Iron Ore 2019, pp 291305 (The Australasian Institute of Mining and Metallurgy: Melbourne).