NCAI

Projects

judge, hammer, judgement

  • Automatic similar case retrieval
  • Legal text summarisation
  • Automatic information extraction from judgements
  • Knowledge-based and indexed database
  • information anonymisation on public facing portals of elite courts

  • Attending to customers in service sector such as banks, hospitals, stores, and hotels etc. by greeting, provision of information, taking orders, giving suggestions, leading tours and entertaining with music and dance
  • Initial screening of patients at hospitals to minimise human interaction and prevent contagious illness

An automated solution for

    • Computer-vision based inspection and grading of raw agricultural produce
    • Increased efficiency and enhanced capacity of food processing units

  • Classify buses, trucks, vans, SUVs, cars, bicycles, motorcycles, and loaders in real time with a mean accuracy of 90%
  • Robust vehicle counting and classification techniques under varying lighting conditions

Developed primarily for the Government of Saudi Arabia for Hajj and Umrah, the computer-vision based system

  • Detects if any person is moving against the direction of the crowd
  • Offers unsupervised statistical modelling of dominant motion in structured scenes
  • Can be useful to prevent accidents in dense crowds

A joint project with Business School at University of Salford and Manchester Metropolitan University, United Kingdom, it can predict bank failure in 1139 banks in G7 countries and Australia with 91% accuracy.
A data set comprising of 59 ratios and variables and 12529 samples from the year 2003 was employed to train the AI system to predict the failures.

Currently deployed at a shopping mall in Islamabad, the system offers

  • Real-time automatic counting of persons in crowds
  • Results independent of camera perspective and scale