[Aim]  To use geometric morphometry to determine whether there are differences in morphology between the different color biotypes of Halyomorpha halys adults. [Methods]  A flatbed scanner (Microtek MRS-9600TFU2L) was used to capture images of the pronotum, forewing, and hindwing of adult H. halys. Data on landmarks and semi-landmarks were obtained by digitizing these and using the draw curves function in tpsDig2. After Procrustes fit in MorphoJ, centroid size and Procrustes coordinates were calculated and imported into SPSS for estimating size variation. After collating the data from different body-color types and genders, shape variation was analyzed using MorphoJ software. Finally, regression was performed on the procrustes coordinates and centroid size. The residuals were used to assess differences in shape independent of size, and for a cross-validation test. [Results]  Dark and light-colored biotypes differed significantly in the size of the pronotum (P_landmark=0.005, P_{semi-landmark}=0.003) and hindwings (P_landmark=0.045) of females and in the pronotum (P_landmark=0.042) of males. There was also significant sexual size dimorphism in the pronotum (P_light-color<0.050, P_dark-color<0.001), forewings (P_dark-color=0.020) and hindwings (P_{both color}<0.050) within the same biotype. There was sexual dimorphism in the shape of the pronotum (P<0.050)in the dark-colored biotype. Cross-validation indicated that 90% of males and 80% of females were correctly classified. Female morphology differed significantly between biotypes (P<0.050). Landmarks and semi-landmarks on the pronotum correctly classified > 90% of specimens, whereas those from the forewings correctly classified > 76%. When biotype and sex were analyzed as independent factors, cross-validation of pronotum data indicated that > 90% of all females had been correctly classified. Data from forewing landmarks correctly classified 100% of females, and 91.67% of males from the dark-color biotype. Controlling for allometric effects further improved the results, except for hindwing data. [Conclusion] Geometric morphometrics can be used to determine the relationship between body-color biotype and local morphological variation in H. halys, which, in turn, can be used to analyze sexual dimorphism in size and shape in this species.