Good day, Dr. Anuja George. I'm Arjun. I've had a look through your background and your professional record. I'm interested in hearing more about your work and how you see yourself fitting in here with our faculty. Ready to start?

Please start by introducing your professional journey in academia.

Hello, are you there?

Hello.

Please start by introducing your professional journey in academia.

Uh, yeah. Good evening, uh, myself, Doctor.

Thanks for sharing that!

Dr. Anuja George, thank you for joining. I see you obtained a Ph.D. from the National Institute of Technology Calicut and have certifications like the NPTEL Elite Certificate on Digital Circuits. You also seem proficient in tools and languages such as Verilog, Cadence, MATLAB, and Python. Let's begin by verifying a series of academic and technical skills. Please note, I will ask role-specific questions one skill at a time to understand your depth in each area. Starting with "Image Processing," can you describe a scenario or project where you applied image processing techniques? What specific tools or methodologies did you use, and what was the outcome?

Hello.

Hello.

Hello, Dr. George. I was just about to verify your expertise in Image Processing. Could you share a scenario or project where you applied image processing techniques, detailing the tools and methodologies you used, along with the outcome?

Yes, as part of my PhD I have done UH.

My thesis was on hardware efficient VLSI architectures for image processing image sensor networks. So as part of this what I have done is in the 1st chapter I have conducted.

The design and development of an hardware efficient DWT architecture which could be applied in which could be applied in image processing scenarios. Now as part of my second work I have done image processing status, I have designed and hardware efficient VLSI architecture.

Uh, for image processing, uh, image dehasing, uh and uh in my third work I have designed and developed a hardware efficient VLSA architecture for image denoising. That is, it will be estimating the noise and then denoising.

Then as part of my 4th work that is the last one, I have combined both the dehazing as well as the Dean noising that is a hardware efficient VLH. VLSI architecture for joint dehasing and denoising was the work that was learned as part of the thesis. So as part of all this I have used the matlab too. Then for the.

Design I have done with the Verilog HDR languages. I have synthesized the design using the genus synthesis tool.

And the outputs were verified to. The quality of the images were verified using MATLAB and the different metrics that were used for PS and ISS and SSIM, NIQ and C2000 parameters were used for the.

Analysis and.

Both the image quality verification as well as the hardware efficiency in terms of VLSI. Both this was the outcome of my research and have proved that without compromising much on the image quality, I could develop hardware efficient VLSI architectures which could be put in sensor nodes to address the issue. That is in presence of haze and noise there will be image degradation. The quality of the images will be great.

And to overcome this issue, I have developed a VLSI architectures that could be incorporated in the sensor notes so that the initial image processing of in the presence of haze and noise could be done within the node itself.

It's nice. MMM.

Understood. Moving to the next area: "Embedded & Communication." Could you briefly describe your experience with embedded systems or communication protocols? How did you approach integrating them into your projects or research?

Uh, yes, uh, in my, uh, in I have a guided.

Several prob.

So basically the batch for the previous batch I have added project named as a safe right. So in that project they have used an adeno board and the interfaces have been done accordingly. And the project was all about the image processing part was also part of the project in the project.

The helmet detector detection was done. Basically the riders often wear helmets, but they are not buckled properly, so the system captures the cameras and using several.

Object detection tools basically your loop they detect the presence of the camera. Sorry, they they detect the presence of the helmet. Apart from the helmet detection, detection, what was ensured in the depositors if the helmets were properly bucket, if the helmets were also properly buckled, only then the system gives a provision for the writer to start the bike. So it was a safe ride system with a helmet detection not only the detection but also it.

That the rider buckles the helmet properly for the safety and.

After only. In that case the system allows the, you know the rider to switch on the.

System or the cycle?

Thank you, Dr. George. Your explanation reflects a strong foundation in embedded systems and communication protocols, especially in integrating hardware like Arduino with image processing for real-world applications. The "Safe Ride" project demonstrates a creative and practical application of these technologies to improve safety. Now, moving on to teaching responsibilities: can you provide an example of how you’ve structured a theory or laboratory course, ensuring both clarity and engagement for students?

Uh, thank you for that. Uh.

I have a teaching expertise across the undergraduates, the postgraduates and event working professionals in the campus. So for the working professionals and the postgraduates, we have designed the courses and as part of the our our being an autonomous college, we have the permission to design the curriculum, design the syllabus. So as part of that, I have designed the syllabus and the.

In the syllabus, we have accommodated tools like the genus synthesis tools.

Of the cadence, so the student has done.

I have guided the student to do a large amount of work in approximate computing and using the genus tools. We have returned the verilog also. We approximate multipliers and then they are generated in the vibrato and then synthesized it in the genus So.

As I have done the syllabus revision and I have incorporated the tools that are currently used in the industry for the same.

Understood, Dr. George. Your approach of revising the syllabus to incorporate industry-relevant tools like the Genus Synthesis Tool and projects such as approximate computing demonstrates a structured and forward-thinking strategy. It’s clear you aim to prepare students for both academic and industrial challenges effectively. Next, let's discuss student evaluation and exam duties. Can you outline your approach to designing assessments that accurately measure student understanding, both theoretically and practically?

Umm, COB is, uh, evaluation is what I would like to prefer. So for whatever be the course that we designed, the particular course will have, uh, certain amount of Cus. So, uh, our target should be like if the students are able to attain the scenes properly. And for this we can have any sort of measurements, assignments, tutorials.

And sometimes signals I have.

Incorporated methods and my teaching.

Understood, Dr. George. You prefer an Outcome-Based Education (OBE) approach by aligning assessments to Course Outcomes (COs), using a mix of methods like assignments, tutorials, and signals for evaluation. This well-rounded approach helps gauge both theoretical understanding and practical application. Now, regarding guiding student projects and research: can you share a specific instance where you mentored students on a challenging project? What strategies did you employ to ensure their success?

Thank you for that question. I have mentored an N number of projects and recently the last year I have entered a project as already discussed a safe flight and the paper was presented in I triple conference and it is published in the I triple proceedings also so.

As part of the guiding, I always allow the students give them the freedom of thinking. We always look forward for innovative projects that can be solutions to real-world problems or challenges and we try our level best. We along with our students, we try our level best to bring them into the implementation level apart from the simulation alone.

We have set the standards like we always presented in some conferences so that we get some inputs from there and those inputs could be incorporated in the projects. That is our real idea of presenting our projects in the conference.

And we keep the criteria to.

And we always try our level best to get them published to set standards of how many meeting it is.

Thank you, Dr. George. Your commitment to guiding students towards innovative, real-world solutions and encouraging project implementation beyond simulations is commendable. Prioritizing conferences for feedback and ensuring project publication reflects a strong mentorship approach that prepares students for significant academic and professional contributions. Now, moving to clear communication and a structured teaching approach: how do you ensure your teaching materials and delivery methods are universally understood by a diverse student audience?

What'll be the codes that I deal with? Definitely the textbooks are unnecessary part of it. I always communicate with students the contents of the textbook. But apart from that, in a classroom there will be a number of students with different degree of understanding. So it is not always advisable to give the textbook and download the pages. So many a times it is quite important to deliver them.

Deliver the notes in a very simple manner so that students are belonging to you know.

With the different outlook, understanding would find it easy and.

To grasp so I put the content. Sometimes it is necessary put them in PPT, sometimes it is important to showcase things and videos, some kind of phenomenons or mechanisms. If they are delivered through videos, it becomes easy. So depending on the context and the complexity of delivery, I handle my sessions and I also.

Give the students the.

To ask questions.

Understood, Dr. George. Your approach to tailoring teaching materials—whether through simplified notes, PPTs, videos, or interactive questioning—demonstrates a thoughtful strategy to address varying levels of student understanding. This reflects a strong emphasis on clear communication and adaptability in teaching. Now, regarding your research credentials, could you elaborate on your research publications in reputed journals? What key contributions have you made in those works?

I have A to my credit I have got 5 uh international journals that are sei umm. I have published my works in journal of real-time image processing, getting back factor of.

Three, I have also published my works in IEEE Census journal with the impact factor of.

Four and in all these works, image processing is the base VLSI architectures for image processing is the so-called or my expertise and.

August 12th actually concentrates on image de hazing, image de noising or adjoint engaging organizing.

Thank you, Dr. George. Your record of publishing in high-impact journals like the Journal of Real-Time Image Processing and IEEE Sensors Journal, with a focus on VLSI architectures for image processing, highlights your significant academic contributions in advancing the field, particularly in areas like image dehazing and denoising. Now, let’s discuss any experience you have with industry projects or consultancy work. Can you share an example where you collaborated with industry professionals or provided consultancy, and what impact it had?

Yeah. As part of Cadence, we have interacted with a few professionals and those industry people have come to the campus to deliver sessions. And during the sessions our interactions about the real world challenges and the current trends in the industry. Keeping all this mind, I have incorporated few changes in the curriculum.

During the syllabus revision and I have presented the same as part of digital system DASA.

Delivered to the SO ECE students as part of the.

Eugene program and in that we have incorporated a course project which.

Thank

Makes use of these tools so that the bridge between the industry and the academic and produce. So even in the lower semester circles advisable according to major advisable to find the gaps and then try to, you know, bridge those gap between the Academy and the curriculum so that they gain the potentials.

To get easily accumulated with the industry by the end of their graduation.

Thank you, Dr. George. Your initiative to bridge the gap between academic curriculum and industry requirements by incorporating insights from professionals, revising syllabi, and introducing practical projects reflects a commendable effort to equip students with skills necessary for industry standards. This approach aligns well with the mission of fostering readiness among graduates for real-world challenges. With that, I have verified your skill set across the targeted areas. I will now proceed silently to the next step.

Sorry.

It's the meaning question.

Dr. George, I see that you've recently completed your Ph.D. at the National Institute of Technology, Calicut. Congratulations on this academic milestone. Let’s delve into your research. Could you highlight one of your significant publications or your thesis topic? I'd like to understand not only the research but also how you envision incorporating it into your teaching.

My uh, first work uh got published and I triple census journal uh, the first work was some hardware efficient BLSA architectures, sorry, hardware efficient FRWD architecture for image processing in visual sensor networks. So.

In that I have developed a hardware efficient DWT architecture which is called as FRWF that is practical filter.

Have proposed to VLSI architectures for the same for an MSN CBA filter. The first was the horizontal architecture and second one is a vertical affiliate architecture. Needs to architectures or DWT architectures that can be incorporated at the sensor nodes because they are hardware efficient and.

Doing performing, dehasing and denoising.

Kind of image processing applications in DWT domain would give better results when compared to the spatial domain. So that was intention of my first work which was published in the high impact factor iterations. The second was concentrated on developing hardware architecture for saturation based.

We know that Misha sensor modes which I kept with cameras, they for most of the applications like intruder detection or surveillance or wildlife monitoring, we are in need of high quality images. But in the presence of ASMI, it is impossible on the image degradation will definitely happen and the quality of images will decrease. So in this context, this is very important to deviate or denoise these images, OK and. The second problem is that these.

Image dehazing so in this.

Images, uh, the, the sensor notes, they are uh computationally, uh, less so we cannot incorporate very hardware hungry architectures in the sensor nodes. So it is quite important to develop a hardware architectures which can handle image processing while they are being in the sensor itself. So my second work concentrated on this and I have taken a saturation based concept in this. I've done several contributions.

I have tried to.

Incorporate the dark channel more.

OK, that was my second book. In the third book it was all about the noise estimation. So removing the address is quite necessary for image denoising, precise image denoising and most of the in most of the work constant thresholding is the method that is used for edge detection. But in the third book of mine I have proposed A polynomial repression based adaptive edge threshold deduction model and using this threshold model I have I.

Develop the edge detection and using that edge detection I do the noise estimation. So it becomes a precise noise estimation because the edge detection is an active 1 and it is also hardware efficient. The last work was the join the Dsing and denoising framework using the FRWF architecture. So in most of the works where DWT dominance incorporated we have 4 subbands, LLHHL, NHL subbands and most of the existing works.

Realising was done in all the three, sometimes that is LHL and HH, but in my work I concentrated on due to some edges of band and renoising was done only on the subband and in that to reduce the hardware hardware requirements I have also incorporated.

MCST Techniques.

And all these works have been published in High Impact.

Journals and while discussing about the VLSI design or I'll discuss about image processing. I motivate this tool when talking about these resources and my experience at the right calcit.

Thank you.

Thank you for providing that detailed walkthrough of your research work, Dr. George. It seems your work is deeply rooted in both theoretical innovation and practical application, particularly addressing the hardware constraints of sensor networks for image processing. Let me ask a follow-up. When incorporating these research insights into a teaching syllabus, for example, for a course on VLSI design or image processing, how would you balance the theoretical foundation with practical exposure for students? Could you outline what your approach might look like in the classroom?

Since I'm a person who has literally gone to different places of the BLSI design, it gives me an upper hand when delivering the classes. So when I talk about.

Definitely, uh.

Starting from.

Uh, description.

And yeah, while giving them, uh, you know, lectures on the BLS are the central and other aspects. I appreciate the students to come to the lab and at least to have a look at the blue literally using the tools. So it creates an understand the students. We know the workflow status. Yeah, quite easy for us to sometimes I create, I arrange a sessions from.

No industry people.

Come and take sessions on board.

So this creates an interest in the students.

Understood, Dr. George. It’s promising to see that you not only aim to build a solid theoretical foundation but also emphasize hands-on exposure and industry interaction to enhance student learning. A final question in this section—how do you tailor complex concepts, like hardware-efficient architectures or polynomial regression-based thresholding, to be accessible to undergraduate students who may not yet have a deep understanding of VLSI systems or image processing?

See, there are different techniques for hardware efficiency, so maybe we cannot deliver everything in a classroom section and it might be unwanted sometimes. But of course, when there is a requirement or if there is a curiosity from the student side, I definitely take up my time and discuss with them about how or why it is important. To have heart direction VLSI architectures, PLSA is a very promising domain.

In the current surround and that too when we are moving on to an era of edge computing, these LSA architectures which are hardware efficient as the soon. So it is quite important for the engineering students particularly with easy background to understand about how hardware efficient architectures can be designed and how why they are important and where they can be incorporated. What are the different applications involved.

That's such kind of information I definitely pass while being in the class. But uh Taru, understanding and detailed explanation of certain things may not be practically important in a classroom, and sometimes it seems to be unwanted for some students. So accordingly, depending on the demand and necessity, I handle this issue.

Thank you, Dr. George. Your balanced approach, integrating foundational knowledge with practical applicability and adapting to the students' needs, demonstrates a thoughtful teaching philosophy. I appreciate how you connect your research expertise with classroom instruction while remaining attuned to the contextual relevance for students. Let's proceed.

Professor, could you provide a detailed 3-year research roadmap focusing on how your work will drive high-impact outcomes and enhance institutional research rankings?

Uh, is it that, uh, with the VIT background I have to talk or, uh, are you talking about, uh, what are my, uh, goals set in research in the next few years?

Professor, I am asking specifically about your goals and strategies for the next three years in terms of research productivity. The focus should be on how your work could enhance institutional research rankings, such as through high-impact publications, external funding, or significant collaborations.

Thank you for the question and clarifying my.

Doubt.

I'm really impressed about the vids. Uh, you know, vits are repeated institute and it has a strong focus on the results, research excellence, excellence as well as the academic excellence. And it is my professional aspiration to do good research works. So I believe in, you know, the next three years I will be able to, uh, conquer or publish.

My works in high quality impact factor journals.

And I also would like to collaborate with the other interdisciplinary teams to bring about impactful projects. And if possible, I would like to market them. I don't know how it is feasible. And also I have a desire to do or actually file for a patent. So these three things are in mind high quality publications.

Then collaborative team works with interdisciplinary projects and.

Don't need to be baked.

Understood, Professor. Now, as a follow-up, how do you plan to mentor junior researchers effectively to ensure a consistent pipeline of high-quality publications supporting these goals?

Dedication is the most important thing in research, Uh.