Recognition a Hand Drawn Chemical Structure Diagrams Using the Discrete B-Spline Wavelet Transform

Chemical structures are a suitable way to represent the chemical equations perfectly in 2D space. But sometimes a hands drawn structures have some complicated when one take them as a document image and then recognized it to its full meaning to be accepted in machine data mining techniques so far. The wavelets with Spline are very steady and commonly symmetric or anti-symmetric. B-Spline has the preferable parataxis properties over all different types of wavelets in order L-1. In this paper a unified framework was built to include the organic and inorganic expressions. A suitable way was presented to classify hand drawn chemical structures using the B-Spline wavelet transform as a tool for image classification. In empirical valuation one can show that an enforcement of this method exceed the open source system available. The proposed framework achieved in Test-5 with 84.7% data accuracy for recognition the handwritten chemical expression database. Also with 77.8% classification accuracy using discrete B-Spline wavelet transforms.