A new image processing method for cell detection in leaves of Arabidopsis thaliana is presented. Using complementary image processing techniques, we introduce a good way to obtain the original cell contour shapes, surpassing the limitations given by factors like noise, stomata, blurred edges, and non-uniform illumination. Preliminary results show this process minimizes considerably the time of cell detection in comparison with the traditional biology methods that include a tedious freehand path, and produces matching percentages of true borders over 80%. Experimental results are shown.