One of the most exciting uses of artificial intelligence is computer vision, also known as machine vision. Many innovations are built on algorithms that understand images, including moving videos and pictures. Smart industrial machinery, autonomous vehicles and filters for your smartphone that enhance the beauty of the images you share on social media include smart, self-driving cars as well as smart industrial machinery.
Recent years have seen attempts to imitate human senses using computing. While there is some success in vision, speech, and listening areas, it is very limited in the savory or olfactory area.
This treatise will discuss the use of Computer Vision to ensure post-Covid human productivity.
It is not a new concept to work from home, but it has become more prominent with the new environment.
WFH allows remote workers to also manage business operations remotely.
The work from home option was initially seen as a temporary solution. We now have multiple Covid-19 variants. This means that employees can choose to work remotely, at home, or remotely.
We are forced to work from home post-Covid. This presents different challenges for senior managers of organizations. Human surveillance is not sufficient and inefficient in situations where employees must be present at the desk to support the business process.
To track productivity and protect data, it is essential to have an automated monitoring system in place.
Tracking tools should be able to provide data for all models. Therefore, a single tool that supports multiple models is essential.
Remote management of some things that were previously done in the office will be required.
* Floor Walking
* Health of your PC
* Compliance audit
Login time and lost productive hours
Tech Mahindra recognized the need for a computer vision-powered monitoring system that generates data to analyze desktop usage.
KORNEA is an application that runs on the user’s computer/desktop and monitors only login activity. It also supports applications that are aided by Computer Vision algorithms.
You have a need?
Figure 2 Key Features
Below the hood:
Two aspects allow KORNEA’s core function to be performed after installation.
Registration:
- Face scanning will use deep neural network and Caffe model
- Face-scanned images are processed with ML to extract the encodings from them
- The encodings from all face images are combined to create a pickle file.
Authorization:
- A new Face image is compared to the pickle file and a confidence score in face similarities is calculated.
- The confidence % value is then compared to a predefined threshold match percentage to determine if the face matches the picklefile and authorize it.
The solution has been designed to protect sensitive data and ensure security.
The tool provides near-real-time reporting that allows for better decision-making by Ops teams.
These statistics show employees’ productivity as well as the health data of hosting machines and their compliance with current IT landscape.
This solution is possible when it is integrated with ITSM and WFM systems, QMS, CCT systems, and key stroke level data. Prediction models can also be built using historical trend analysis.
The operations team can eliminate the manual labor required to monitor, report and analyze data and instead use automated data classification and automation to automatically classify and classify data.
Key Benefits:
* Risk mitigation – The report provides a detailed assessment of IT security compliance and any potential risks.
* Performs an objective and thorough evaluation of the system’s status. This eliminates potential security risks and improves data quality. It also increases system performance.
* Preventive measures and security tightening.
* Cost efficiency – Reduces the need to audit compliance manually
* Customer focus – Provides assurance that compliance obligations and controls are being met as desired
* Cloud Ready – KORNEA Data collection can be set up to either Cloud or On-Premise, as requested by clients. We are able to work with any cloud provider like AWS/GCP or Azure.
We are committed to protecting the privacy of our employees. KORNEA solutions can be seen as intrusive tools to monitor employee activity and hold employees accountable for their non-compliance. You can check the Amazon site by clicking here, transparency being the key.
KORNEA’s goal is to increase efficiency and build trust. The implementors have visibility and insight into the data being captured and reported. KORNEA can be configured so that data can be classified to a level of “Productive” or “Non-Productive.”
Any granular details that are not within the scope of the solution (e.g. screen captures or URLs) can be blocked from being captured and reported.
KORNEA adheres to the standards, codes, and ethics of its customer organizations and thus ensures that we have the trust and confidence of our employees. The fact that employees can be confident that their personal information will not be captured or shared with others, even their managers, is a strong endorsement of the fact that they can trust that KORNEA respects the ethics and codes of the customer organization.
KORNEA has proven to be extremely efficient for back-office teams in several of its implementations. Back-office operations, particularly back office, lack the means to monitor and capture agent efficiency. Most of the capture is manual or advisor-led. Therefore, there are always insufficient insights into non-productive work activities.
This results in a lack of transparency regarding employee time usage, a lack of knowledge about productive hours, and a missed demarcation of application usage.
KORNEA is used across multiple customers to provide insight on the time spent on accessed apps. This allows us visibility on log-in hours and productive hours.
The inbuilt Time Track & Engagement feature provides advanced insight into the time spent on each activity, identifying patterns and tracking outliers. We were able to increase productivity by up to 7% and login hours by an additional 10%.
The future is bright
KORNEA’s core engine, a Deep Neural Network, is a Deep Neural Network that learns from samples and matches with an incoming picture with high accuracy. Our algorithm is accurate to more than 99 percent.
Face match can be used in a variety of ways. For example, we are often asked to create a report on duplicates and sort through photos.
Kornea has the ability to expose the API for citizen-development, which can help save quite a bit on consulting and development costs for customer organizations. Visit Gramhir to find out more information
Computer Vision’s future is bright ….., as you can see.
Author’s Biography –
Diptojeet Mokhopadhyay, a Digital Process Automation specialist with over 23 years of experience in end-to-end delivery across the globe, is Diptojeet Mukhopadhyay. His experience includes Customer management, Practice &CoE Management and Technical Delivery Management. Diptojeet has deep expertise in DPA transformation leveraging leading tools and technologies. Diptojeet has been involved in numerous consulting projects around the world for leading brands from Australia, North America, UK and the Middle East. He is also a Togaf 9 Certified Enterprise Architect, with additional certifications in UiPath and Appian.