Lignocellulosic biomass is a renewable alternative to fossil resources. It presents as a complex matrix of cellulose, hemicelluloses and lignin that make it recalcitrant to conversion into bioproducts. To overcome this recalcitrance, a pretreatment step is thus required. One common industrial physico-chemical pretreatment is steam explosion, which is efficient to increase access to cellulose by degrading hemicelluloses while redistributing lignin. Since this process leads to important biomass ultrastructure modifications, it is useful to relate chemical properties and compositional data to morphological data in order to understand steam explosion effect for biomass recalcitrance. In our study, untreated and steam exploded (at different severities) samples of beechwood were observed using fluorescence macroscopy. Resulting images combine chemical data through fluorescence and morphological data of plant cells. Sections of the samples were exposed at 353 nm, which corresponds to lignin autofluorescence, and acquired images were analysed using an automated method developed under ImageJ software that allows quick, accurate and simultaneous quantification of cell walls fluorescence intensity, which gives information about chemical properties and the environment of lignin, and quantification of morphological parameters such as perimeter, area and circularity of cell lumens, which attest of the plant structure alterations. Results in fluorescence macroscopy show the impact of the steam explosion pretreatment as fluorescence intensity is visibly decreased with increasing severity. Morphological changes reveal shrinkage of cells and deformation of cell walls, which results in decrease of lumen area and loss of circular shape of lumens, respectively. Quantification of fluorescence and morphological parameters can be correlated to pretreatment severity and compositional analyses.