Accurate budget estimation is an important prerequisite to guide the project. The traditional method using a linear estimation model can not accurately reflect the contribution of each component to the budget estimation of the entire system, leading to poor estimating results. This paper proposes an accurate project budget estimation model based on chaotic post-processing SVM-PCA (Support Vector Machine-principle Component Analysis). On basis of SVM model, the model filters redundant information in the system to ensure the input information data contribution rate. Then after output the data, chaotic post-processing method is adopted to smooth irregular characteristics of the data, in order to ensure the accuracy of the budget estimating model. Finally, five projects in a group of 10 categories elements are used to conduct estimating budget experiments. Experimental results show that the project budget estimation model based on chaotic post-processing SVM-PCA can accurately estimate the core consumes of each project, therefore has great value in engineering.
Non-Model Based Method for an Automation of 3D Acquisition and Post-Processing