Recently,　the　field　of　artificial　intelligence　has　seen　many　advances　thanks　to　deep　learning　and　image　processing.　It　is　now　possible　to　recognize　images　or　even　find　objects　inside　an　image　with　a　standard　GPU.　Image　processing　is　a　recent　science　that　aims　to　provide　specialists　from　different　areas,　as　to　the　general　public,　tools　for　manipulating　these　digital　data　from　the　real　world.　The　detection　of　moving　objects　is　a　crucial　step　for　systems　based　on　image　processing.　The　movements　detected　by　the　classic　algorithms　are　not　necessarily　interesting　for　a　thorough　information　search,　and　the　need　to　distinguish　the　coherent　movements　of　parasitic　movements　exists　in　most　cases.　In　this　paper　we　are　going　to　use　a　simply　webcam　and　YOLO　algorithm　for　this　implementation.　The　YOLOv3　(Version　3)　model　makes　predictions　with　a　single　network　evaluation,　making　this　method　extremely　fast,　running　in　real　time　with　a　capable　GPU.　From　there　we＇ll　use　OpenCV,　Python,　and　deep　learning　to　apply　the　YOLOv3　object　to　images　and　apply　YOLOv3　to　video　streams.　©　2021　IEEE.