Body core temperature is generally used as a proxy to detect health or thermoregulations problems. In summer, sows can face thermal stress, with negative impacts on their welfare and performances. In practice, body core temperature is evaluated by using rectal thermometer but this method may be stressful for the animal, with possible subsequent effects on the measurement quality. Thermal imaging camera represents a non-invasive alternative for measuring animal temperature. The purpose of this study was to compare gestating sows’ body temperatures measured with a rectal thermometer and with a thermal imaging camera in different ambient situations (14 to 32 °C). Individual temperatures were taken on 40 sows (8 primiparous), housed in two rooms, every morning (8am) during 12 days with a rectal thermometer and with a thermal imaging camera (model FLIR, E60; extracting software FLIRTOOL) on a shaved dorsal cutaneous area (2×12 cm2). The fixed effects of parity (primiparous vs multiparous), room (1 or 2), sow’s posture during measurement (lying vs standing) on body temperatures (rectal and cutaneous) were analysed on R applying a linear mixed-effects model taking into account the random effect of the sows. Correlation coefficients between the sows temperatures from one side and the room temperature, live body weight, backfat thickness on the other side were calculated using Pearson correlation tests. Primiparous sows had higher rectal temperature than multiparous sows (38.3 vs 37.9±0.61 °C, P=0.007). Rectal temperature was also correlated positively with room temperature (R2=0.34, P<0.001) and negatively with sow body weight (R2=0.38, P<0.001). The sows lying down had higher cutaneous temperature than the sows standing (35.3 vs 31.8±3.19 °C, P<0.001). Cutaneous temperature was positively correlated with the room temperature (R2=0.77, P<0.001) and rectal temperatures (R2=0.48, P<0.001). To conclude, there is a relationship to explore between sow’s rectal and cutaneous body temperatures. Few factors seem to influence these temperatures and therefore, they should be taken into account in the next step of this work aiming to develop a model predicting rectal temperature with infrared temperature and other non-invasive measurements.