A Tally System Based on CNN and Block Chain

In order to design a reliable and convenient tally method, the emergence of a tally system solves the problem of high database maintenance costs, such as management of the site factory, telecom, power companies, etc. The novel phases, constituted of the Convolutional Neural Network (CNN) and the block chain, extracts character characteristics through the VGG network training character classification model. The principal component analysis(PCA) algorithm accelerates feature comparison efficiency by reducing the dimensions of the feature matrix. In terms of data storage, the consistency of the Practical Byzantine Fault Tolerance(PBFT) algorithm ensures the accuracy of the data. The experiment demonstrates that practicability of the system can catch the requirements of the automatic bookkeeping method in the construction site.