OPS-SAT mission is a key element for operational validation in flight (or IOD/IOV) of the DeepCube service of Deep Learning at the edge that our company is preparing with the support of the GSTP programme (GSTP Make, ESA-CNES) and the R&D performed in the CORTEX project (Permanent Open Call EOEP-4, ESA/Phi-Lab, https://esacortexproject.agenium-space.com). The goal of the proposed experiment is to execute on-board the inference of the simplified models defined in those projects. Chosen use case is forest detection (segmentation problem). Our simplified DNN was pre trained on S2 images over Slovenia, these images were processed to simulate OPS SAT images. Then real OPS SAT images were used with the transfer learning method. The Binary Deep Neural Network using convolution was ported on the Cyclone V card using our own custom IP. A lot of efforts have been put to fit the software within the HW constraints (in terms of memory/storage resources) and many tests have been performed with the support of the OPS SAT team to be able to send our package on board. Final results of the experiments are still to be retrieved.