The mathematical notation is well known and used throughout the world. Humanity has evolved from simple methods to represent accounts to the current formal notation capable of modeling complex problems. In addition, mathematical equations are a universal language in the scientific world, and many resources such as science and engineering technology, medical field also not an exception containing mathematics have been created during the last decades. However, to efficiently access all that information, scientific documents must be digitized or produced directly in electronic formats.
Although most people are able to understand and produce mathematical information, introducing mathematical equations into electronic devices requires learning special notations or using editors. The proposed methodology is focused on developing a method to recognize intricate handwritten mathematical equations. For pre-processing, Gray conversion and Weiner filtering are used. Segmentation is performed using the morphological operations, which increase the efficiency of the subsequent image of equation. Finally Neural Network based template matching technique is used to recognize the image of handwritten mathematical equation.