| I&T Solution |
AI OCR Based Smart Lifts & Escalators Design Approval Platform
(REF: S-1216) |
| Trial Project |
|
| Solution Feature |
- Instead of traditional OCR algorithms, the system will be equipped with state-of-the-art and well recognized deep-learning-based OCR engines, trained with millions of human writing and machine-printed images in both Chinese and English characters. Generally, deep learning-based OCR engines are 15% more accurate than traditional pixel comparison based OCR algorithms.
- The system will be utilizing NLP based-data field mapping technologies to extract specific data from the submitted reports, with various layouts and expressions, into a structured format for subsequent handling.
- The system will be able to provide a confidence level/score for each approved/disapproved application, based on predefined approval logic, rules and calculations. The system should be able to compare new applications with past approval records and generate a confidence score for operator/engineer reference.
- The system will be able to provide report functions and design some report templates, provide data visualization and include workflow functions.
- The system will be able to generate standard approval documents for each application.
|
| Trial Application and Expected Outcome |
- Smart Lifts & Escalators Design Approval Platform is designed, developed and implemented with Optical Character Recognition (OCR), Artificial Intelligence (AI), and Big Data technologies.
- OCR technology is used to facilitate specific data extraction from the submitted design documents, mostly in the form of scanned images of documents.
- AI and Big Data technologies are used to enhance the accuracy of the OCR function of the platform, the platform also provides an automatic approval function for vetting of the submitted technical documents.
- The platform is also well integrated and developed to facilitate e-submission of type approval applications such as receiving documents for equipment details and specifications, comparing data, verifying compliance, issuing approval, and up-keeping records.
- The OCR function and automatic approval function could achieve high accuracy and useability when the platform development is finished.
|
| Info on I&T Solution Provider |
| Solution Provider | : | Datax Limited | | Address | : | Unit 708, Data Technology Hub, 5 Chun Cheong Street, Tseung Kwan O, Hong Kong | | Contact Person | : | Aric Fung |
| Position | : | Account Director | | Tel | : | 9143 8473 | | Email | : |
aric.fung@datax.io | | Webpage | : | https://datax.io |
|
