Soon, AI-based attendance system at Prayagraj Municipal Corporation

Prayagraj Municipal Corporation (PMC) has initiated steps to have a unique mobile application-based facial recognition attendance system for all its staff and officers soon.

The smart Artificial Intelligence (AI)-based facial recognition attendance management solution would help in automatic identification and verification of an employee from live environment through comparison of selected facial features of the image taken from an already existing database, said Uttam Kumar Verma, environment engineer of Prayagraj Municipal Corporation who is overseeing the project.

The proposed facial recognition-based component would be available on Android/iOS platform to be used on smartphones exclusively, he added.

PMC presently has 5,160 sanitation workers working for it and this attendance system would be covering them in the first phase and subsequently it would be implemented in all departments of the municipal corporation, he said.

The application would have the capacity to be able to maintain a minimum peak load of transactions of 6,000 staff members (Official Login IDs: 40; Supervisor Login IDs: 150; Staff: 5810) supported by a minimum of 190 logins as per hierarchy.

“Presently we have a system wherein the sanitation supervisor takes the attendance of the sanitation workers out in the field, and the zonal officers concerned too keep a tab on them randomly. However, this new application would allow us to not just take the attendance of the workers in the morning while reporting for duty and evening when their duty ends but also randomly during the day on real-time basis. This would make our functioning more effective,” said Verma.

He said that post the tendering process, the agreement with the selected agency has been penned and presently the collation of requisite data is underway for the proposed database. “We expected the system to get implemented within the next 45 days,” he added.

How it works

The new smart attendance application would ensure that login access for a department is only available to the nodal officer assigned to it and no other staff, but in case of absenteeism provision would exist for the nodal officer to delegate or assign the responsibility to another staff as authorized. It would also have False Rejection Rate (FRR) of mere 0.001% and False Acceptance Rate (FAR) of less than 0.0001%. The system would come equipped with video/static face detection system to capture single and multiple presence at a time along with face anti-spoofing features to eliminate face recognition spoofing using photo/video to prevent an employee from using a photograph or a video to mark his attendance with the help of a colleague while not actually being present at the scene. The application would even possess constant self-learning of employee/worker using Artificial Intelligence (AI) technology to identify and capture the physical changes in the user over time whenever it is required, officials shared.