Intelligent Image Management Platform

Fields of Application

Finance, Treasury, Healthcare, Transportation

Solution Overview

The Intelligent Image Management Platform aims to build a bank-wide content management infrastructure, formulate unified standards for unstructured data management, achieve unified and secure management of massive unstructured data, efficient and convenient access to that data, achieve full integration and sharing of information, establish a unified view of the bank-wide content, and complete planning and governance on unstructured data.

Solution Highlights

1. Distributed storage to solve data disaster recovery and backup problems
The distributed storage system can automatically generate one (or more) copies when storing unstructured data. Storage nodes can be expanded horizontally and vertically, solving the problems of disaster recovery and backup of unstructured data. Through logical grouping on the underlying storage, multiple ODSs are configured in the same group to achieve soft RAID, which enhances the concurrent processing performance of the system and the fault-tolerant recovery capability of the data.

2. OCR technology for rapidly text information capture
OCR technology swiftly captures text information from target objects using image input devices such as scanners and mobile phone cameras. This technology is not dependent on network connection, and can be used on various systems and platforms, including PC, servers (Windows, Linux), and mobile terminals(iOS, Android), which is convenient and fast.

3. Intelligent double recording for key link traceability
Intelligent double recording components are essential for business operations and risk control in industries such as finance and insurance. The platform collects audio-visual data and electronic data through technical means such as audio and video recording, digital data, as well as recording and preserving key links in the sales process of financial and insurance products, so as to realize that sales behaviors can be played back, important information can be inquired, and problem responsibility can be confirmed.

4. Separation of front and back-end technologies
In the deployment of the intelligent imaging platform application architecture, front-end and back-end services are separated based on business functions, allowing for more flexible integration and business combinations.

Solution Architecture

Overall System Architecture

Business Value

The construction of the intelligent image content management platform aims to build a bank-wide content management infrastructure, formulate unified standards for unstructured data management, achieve unified and secure management of unstructured data, efficient and convenient access, achieve full integration and sharing of information, and finally establish a unified view of the bank-wide content to complete unstructured data planning and governance.

Application Scenarios

Scenario 1: Image acquisition
The intelligent image management platform utilizes the Agent client method for image acquisition, replacing the OCX control method. This approach enables cross-platform integration of image control, eliminating reliance on the IE browser. By using the Agent client, the platform avoids issues related to browser memory leaks that can cause system crashes.

Scenario 2: Image storage
The intelligent image management platform employs a microservices architecture (SpringCloud + Docker) for deployment. It uses distributed storage methods, including object storage, distributed storage, the distributed database SequoiaDB, file-based storage such as FastDFS, SDS, and NAS. This approach enhances the system's ability to handle concurrent processing and data fault tolerance, effectively ensuring data security. The platform can be configured according to the business volume in the early stage. With the growth of business volume, the processing capacity of the host can be expanded horizontally by adding hardware, so as to achieve "investment in stages and batches, flexible investment".
 

Successful Cases