IIIT-B upcoming projects: The International Institute of Information Technology, Bengaluru (IIIT-B) announced 11 visionary projects at the Bengaluru Tech Summit this year. These spanned across the EdTech, MedTech and AgriTech sectors, hoping to modify the way these fields function. Some of these projects have been envisioned against a backdrop of the international coronavirus pandemic, whilst other folks are hoping to address extended-standing issues in these fields. To give much more insight into these initiatives taken by the esteemed institution, IIIT-B Director Professor Sadagopan talked to The Spuzz Online’s Bulbul Dhawan. Here are edited excerpts from the interaction:
Drones seemed to be a massive concentrate at the Bengaluru Tech Summit discussion. Could you inform us much more about their utility in a variety of fields?
Drones are autonomous automobiles, which we mainly appear at as a new wave of AI and robots. Historically they had been viewed to transport issues and folks, and as a result, the applications for drones as viewed by the western planet had been healthcare and logistics. For the Indian context, we believed of other applications, like precision agriculture. The greatest issue with agriculture in India are the incredibly little farms. And we wondered if for them also, we could undertake image processing to assess the well being of the crops and possibility of any illness and if we could take the action and spray pesticides to include illnesses amongst these crops.
With remote sensing, India has played a top function, but regrettably, it does not function for agriculture. In this field, the photos need to have to be processed truly speedily so that action can be taken inside a day or two to safeguard the crops.
Another application is in the region of site visitors accidents. India has the biggest quantity of road accidents in the planet, and about 1.25 lakh deaths take location due to road accidents each and every year in the nation. Mostly, deaths happen mainly because the blood flow is not contained in time. So we looked at drones that would attain the accident spots reasonably rapid and include the blood loss, so that the nation would be in a position to save quite a few much more lives.
In a way, the IIIT-B is focusing on drones in the Indian context for the bigger social issues one of a kind to the nation.
Do you feel drones are disruptive innovation?
I feel drones are disruptive innovations in a couple of techniques. One is, historically, avionics and aeronautics had been noticed as technologies that would need to have an airport and big infrastructure, and which would be for a comparatively smaller sized quantity of folks. Looking at the earlier days of automotive, driven by a driver equipped with specific abilities, needing to go to the service incredibly regularly, we see that it took some time just before it reached a stage exactly where most folks can drive and have spaces at residence to park the vehicles. Essentially, we reached a stage of mass deployment. Similarly, drones can also be looked at planes for the popular folks in a way.
Another point is that trucks are the mainstays of logistics. Due to their size and the truth that an person generally does not demand truck-worth of something, they generally begin and finish at a certain location exactly where an person has to go and drop their issues or arrange for that to be performed. At this popular point, the truck businesses pack all the products of unique people in containers, load them, transport them, unpack them and then distribute them. Instead of that, drones appear at generating these processes point-to-point, a facility at the moment offered to only a handful of folks.
To me personally, a much more satisfying outcome is that it would lead to a new-age transportation of goods from individual to individual, particularly more than a quick distance. If you appear inside a city, a lot of little issues retain taking place, like the transportation of documents or little products for which drones can be significantly much more effective.
The cause why we couldn’t feel of drones for all these years was that their coordination is significantly tougher. But now, we have the technologies in which we are employed to routing millions of issues like YouTube videos, which has produced it less difficult to apply that variety of technologies to handle drones from a report point.
How is AgriTech developing and what is India’s function in it?
We have not looked at agriculture for all these years, and even for quite a few youngsters, it is not cool. Then we spent some time with senior agricultural scientists, who told us about challenges, challenges and possibilities in agriculture and it was thoughts-boggling. There are so quite a few possibilities for science and technologies in agriculture, and each and every aspect of it, be it climate forecasting, the suitable harvesting technologies, the provide chain and the continuous monitoring.
Three of the IIIT-B alumni are operating a begin-up and they are functioning on drastically escalating the yield and the margins that agriculturists get. The quantity of wastage of fruits and vegetables in India is big due to the disconnect in between the provide and the demand and they are attempting to function on somehow bridging this gap with the provide chain technologies brought to the farmer level. With that, the farmer would be in a position to straight connect with the buyer, in a group of shoppers if not individually.
Do you feel the step taken by your former students is a step ahead of the Centre’s e-NAM initiative?
Connecting farmers to shoppers is one particular portion of the project. You have to make the farmer take the worth out of it. The web was sort of offered in the early 1990s, but it took a whilst just before Google came and folks could search the web. Similarly, Amazon had to come so that folks could switch to on line buying and Facebook had to come for social media. Building the web was not enough.
This appears at letting farmers program ahead for 3 years or 5 years, significantly like corporates do, and if they do it, can they also access bank credit not day-to-day but also more than a period of time, and can forward trading can be undertaken in this as effectively.
Also study | From ventilators to VR-primarily based rehabilitation technique, IIIT-B unveils projects to revolutionise healthcare sector
What stages of improvement are these 11 projects in?
These projects are nonetheless in improvement and I feel it would be unfair for us to claim that we have solved the issues, and we are just beginning. We hope that in the subsequent couple of years we can begin delivering.
How did the group feel of creating a platform like BeIYo and what need to have does it hope to cater differently from the at the moment offered infrastructure?
In the existing scenario/infrastructure, all the health-related records are digitally developed by healthcare facilities in isolation. When the need to have arises, the concerned person/patient can access the very same in physical or digital type. This indicates that there is no technique to confirm the authenticity of such records in true-time. With this getting the existing situation, various circumstances of counterfeit COVID-19 test final results that is getting made by people for travel and other purposes have grow to be prevalent.
In an work to mend the existing scenario, the IIITB group has launched, BeIYo – India’s 1st COVID-19 blockchain platform jointly created by YoSync (a organization incubated at IIIT-B) and a Malaysian primarily based blockchain startup BelfricsBT in joint analysis with IIIT-B and funded by Mphasis.
Before creating BelYo, the group also developed an education blockchain DAAP (Decentralised application) exactly where educational institutes can generate graduation certificates and associated components. This is jointly owned by the student who can share his/her credentials to potential employers and can be verified in true-time. The solution, CrediBel is at the moment getting tested at IIIT-B.
When IIIT-B and Mphasis launched the COVID-19 challenge the very same idea was applied to health-related records exactly where the person/patient along with the healthcare facility co-owns the COVID-19 test certificate. This can be employed in the future for the vaccination certificate or any health-related record which the patient can share with any individual in true time. There is also the alternative to revoke the access, if expected. This option counters fake certificates and offers the patient one hundred% handle on his/her health-related records. BelYo also adheres to the prevailing privacy policies like GDPR and Sri Krishna Committee draft report on India’s Privacy policy.
Coincidentally, on 15th August, 2020, PM Narendra Modi launched National Digital Health Mission (NDHM) and as per the draft policy document, BelYo has currently created 2/4 key modules that are talked about. The IIIT-B group is confident about getting one particular of the service providers of NDHM in the future and has applied for the NDHM sandbox challenge.
How did the group function on coding the Margadarshi bot to ask concerns that are of the level of an IAS exam, provided the difficulty of IAS as effectively as the subjectivity of interviews?
IIITB connected with the India-for IAS Academy for the concerns and shared the very same with the BOT. The professionals shared their insights and thoughts and their expertise in the field more than the years. IIITB’s technologies permits stick to up query generation primarily based on the candidate’s answer, which is primarily based on Deep Neural Networks, particularly transformer primarily based models. In the future, there is hope to boost this additional with the information from the UPSC domain as effectively. IIITB is also in preparation to make the Agents much more expressive and genuine.
Could you clarify much more about the Bar Chart Digitise project?
This project aims to attain an automated reading of photos of charts, e.g. bar charts, and scatter plots. These charts are offered in image format predominantly, say in analysis articles, textbooks, and newsprint media. We are at the moment employing image processing tactics for extracting the information that has been employed for plotting the charts. We use machine finding out models (neural network models) for chart variety classification and text extraction. We also use optical character recognition for detecting text. Our future target is to recognize acceptable machine finding out algorithms or models to enhance our final results in information extraction. We have tested our approach for a number of datasets of chart photos, and are at the moment creating a tool which can take any user input of a chart image from the internet or scanned image, and extract the information table, re-produce the chart at larger resolution, redesign the chart, and so on. We have at the moment worked on digitizing photos of bar charts, such as seven sub-forms. Solving the issue space for popularly employed sub-forms of bar charts, e.g. stacked bars, grouped bars, and so on. is our novel contribution.
Charts are one particular of the simplest visual representations of information. There are a number of applications exactly where the entirely automated reading of charts will assistance, say in identifying crucial statistical trends in a number of charts generated for big datasets, for a number of runs of descriptive statistical evaluation, and so on. One of the crucial places exactly where personal computer vision of chart photos can assistance is in advertising STEM education for the visually impaired, and in incorporating information and facts from charts in documents for screen readers. However, the state-of-the-art is nonetheless in semi-automated options, owing to the vast design and style space in formatting charts employed in the plotting library, the higher quantity of accuracy necessary in reading charts, and the sparsity of pixel information and facts in the photos themselves. The sparsity in pixel information and facts aids in escalating information-ink ratio and is beneficial for human reading, but is counter-productive for an automated reading. Given these challenges, this region is nonetheless ripe for AI/ML options. Hence, our interest in the issue statement of chart image digitization.