Deep Learning based credible text extraction from the captured image.
There were not enough dataset available in Indic Languages for the training of our model, so we created a mechanism to genetare the synthetic text for the laguage. Generating systematic logs, updating the data on realtime. Enabling fingerprint feature on the android device. Word level extraction is done for text classification. Feature to assign the officer to a case and even add annotation volunteers to it. Increase employability: There is a facility of adding the volunteers as part time workers for data annotation. Auto-Scalable using Kubernetes.
Technologies used
Discussion