基于特征智能识别的业扩报装申请材料校验技术研究

目前供电公司开展业扩报装业务需人工校验客户申请材料,占用供电公司及客户大量时间,并且存在一定的失误率。为提高申请材料校验速率与服务质量,推进业扩报装前端服务数字化水平,提出了基于图像、号码数字化特征识别以及公用系统信息对比校验的申请资料校验系统,其中通过改进的LBP算法识别图像数字化特征、卷积神经网络识别号码数字化特征,4SF特征相似度匹配算法实现特征对比。使用核心申请材料身份证校验验证了系统的有效性和可靠性。研究内容为促进业扩报装申请材料校验数字化发展,提供了有效理论与实践指导。

Research on Verification Technology of Business Expansion and Installation Based on Feature Intelligent Recognition

At present,power supply companies need to manually verify customer application materials for the business expansion,occupying a large amount of time for power supply companies and customers,and there is a certain error rate.In order to improve the rate and service quality of verification of application materials,and promote the digitalization of frontend services in the business expansion,this paper proposes an application data verification system based on image and number digitized feature recognition and public system information comparison and verification.In the system,an improved LBP algorithm is used to identify image digitized features,a convolutional neural network is used to identify number digitized features,and 4SF feature similarity matching algorithm achieving feature comparison.The verification of core application materials ID card is used to verify the validity and reliability of the system.Effective theoretical and practical guidance for promoting the digital development of materials verification in the business expansion is provided in this paper.