Design And Implementation Of Multiversion Procurement Contract Management System Based On Image Matching

With the development of the enterprises and the progress of Internet technology,enterprises in daily work of the electronic official document is becoming more and more multifarious,contract as one of the core of enterprise files,in the role of enterprise file management is self-evident,so an effective contract management system can not only speed up the processing efficiency of daily work,employees to reduce the error rate,but also can promote the development of enterprises,improve enterprise competitiveness.However,at present,many enterprises still manage the contract documents manually.The management cost is high,the efficiency is low,and it is difficult to find the contract.Many companies also consider using full-text retrieval technology to upgrade the contract management system.However,due to the textual similarity of the contract,manual comparison is still needed after retrieval,which is time-consuming and of low accuracy.Therefore,it is of practical significance to apply image matching technology to the contract management system.In view of the current situation of enterprise procurement contract management,this project started from the actual needs of a railway manufacturing company in wuhan,and upgraded its ERP system.Based on the procurement module,the overall design and development of the system were carried out.Finally,the multi-version procurement contract management system was completed.Firstly,the development of the basic business functions of procurement was completed,including procurement demand analysis,procurement application,supplier management,contract application,etc.Then,the management function of contract documents was added according to the difficulties in procurement contract management.For contract file management module,the main use of full-text retrieval technology and image matching technology,this paper compares the current popular image matching algorithm,aiming at the shortcomings of the SIFT algorithm in contract matching scenarios are analyzed in detail and put forward the improvement method,verified the effectiveness of the algorithm is improved by experiment.At the same time,the possible noise problem of text image was analyzed.The local binarization method was used to solve the uneven illumination problem of image.The image was skew corrected by Hough transformation method and Canny edge detection algorithm.After testing,version of the procurement contract management system can meet the demand of daily purchasing business of the company,contract management capabilities have greatly improved the staff work efficiency,system front-end interface simple and comfortable,better user experience,at the same time because the system uses the modular structure design,has good reusability,maintainability and extensibility.