Applying the AI-OCR technology base, it helps automate business processes through effective document classification of image format documents and data capture of Key-Value recognition for key items within a format, and saves significant time and cost of manual work.
WHAT
· Which type of image format document used for work should the Key-Value recognition data value be converted into a DB or should be used? · Which types of image format documents used for work need to be classified?
WORK
· Which of the image format documents to be classified or converted into a DB of Key-Value recognition data values is repeatedly hand-processed?
HOW
· Automate manual tasks to improve work efficiency with Image Document Data Capture solutions
Additional development of data capture engines can be supported for specific forms of customer requirements
Key recognition items
· Key-Value definition of key items to extract data for each format
Key-Value Recognition
· After analyzing documents on keywords and key-value definitions in the format,
area recognition and OCR processing for document classification and key-value through deep learning learning
Recognize optimal data capture
Delivering a deep network
· Using AI Deep Learning Network for Key-Value with Different Formats of Stinks and Pictures
· Minimize false positives by applying optimal thresholds
· Accuracy of 95% or more based on normal photographs (light reflection, rotation, cropping, shaking, etc.)
Fast Rate of Recognition
· Fast recognition speed with GPU when using server version: within 0.2 seconds
※ GPU : Graphics Processing Unit
OS
·Windows 10 Professional 64bit or higher ·Windows 2016/2019/2022 Server Standard 64bit or higher ·Linux
Server Specifications
CPU Version · Intel Xeon 2.3Ghz 4-Core or higher · Memory 16GB or higher · HDD 300GB or higher GPU Version · Intel Xeon 2.3Ghz 4-Core or higher · Memory 16GB or higher · HDD 300GB or higher ※ GPU: At least 8GB of NVIDIA-series memory