The use of camera vision and real-time computation as plant identification tool has become an active research. Application of both methods can not be separated from a pretension that the computation must produce correct results within the specified time interval where the truth of the calculation depends not only on logical truth but also on the time in which results are produced. The purpose of this study is to identify plants with fractal dimension analysis and the application of parallel computing. Fractal dimension analysis results showed that each plant has a typical Fractal dimension value. Filtration process with a small window size is accompanied by the use of multiple processors indicates that the image processing in parallel show results much faster than processing with a single processor.