Image commercially licensed from Unsplash

As the emphasis on manufacturing in the USA grows, the importance of efficient ‘Engineer to Order’ processes becomes paramount. Diving into this niche, I came across a distinguished paper by Narasimha Bhat. Impressed by the depth of his insights, I decided to reach out. Our discussions not only enlightened me on the intricacies of the ‘Engineer to Order’ domain but also led me to another of his intriguing research pursuits — analyzing compressed images, where he plays a pivotal role as a co-author. Graciously, Narasimha facilitated a dialogue with the entire research team from India, giving us a unique window into the future of document image processing. Dive in with me in this captivating conversation.

Introducing the Authors:

Kavita V. Horadi, Assistant Professor at the Department of ISE, Global Academy of Technology, Bangalore, India, is an emerging voice in the fields of Document Image Processing, Data Compression, Compressed Domain Processing, Document Analysis, Pattern Recognition, and Structure Analysis. With 8+ international papers in reputed journals and conferences under her belt, Kavita’s work is ushering in innovative techniques for comprehending document images.

Dr. Jagadeesh Pujari, Professor at the Department of ISE, SDM College of Engineering and Technology, Dharwad, India, stands as a beacon of knowledge. With more than 60 international papers to his name, Dr. Pujari’s extensive research in Image Processing, Pattern Recognition, and Machine Learning is well-recognized. Some of his pioneering papers in these areas have each amassed over 350 citations, reflecting their monumental significance in the academic world.

Narasimha Prasad Bhat, a seasoned professional with 15+ years in the Oil & Gas industry and an impressive 17+ years in SAP consulting, is presently a Supply Chain Architect at a global consulting firm in Brookfield WI, USA. Honoured with accolades like the ‘SAP Champion’,Globee’s Best IT Blogger’, and “The most admired Global Indians by Passion Vista” Narasimha’s multifaceted expertise is evident in his contributions. With five international papers to his name, he seamlessly bridges the gap between business and academia.

Q1: Kavita, your work dives deep into Document Analysis by detecting tables present in document images that are in compressed form. What inspired this direction of research?

Kavita V. Horadi: The genesis of this research lies in the sheer complexity of document images. Unlike standard images, documents encapsulate rich layers of information. My goal was to develop a system that could automatically, and accurately, recognize and process this information which is in a compressed version, especially the tables, which are vital for comprehending data.

Q2: Dr. Pujari, your studies often present a blend of Image Processing and Pattern Recognition. How does this synergy enhance the accuracy of document image analysis?

Dr. Jagadeesh Pujari: Image Processing and Pattern Recognition, when combined, create a powerful toolkit. While Image Processing helps in refining the raw image data, Pattern Recognition allows us to discern and categorize patterns within that data. For document images, this amalgamation ensures that not just the visual layout but also the content is analyzed with remarkable precision.

Q3: Narasimha, coming from a business background, especially in SAP consulting, how did your journey into this research initiative begin?

Narasimha Prasad Bhat: My journey into this research began when I wrote a paper on ‘Engineer to Order Manufacturing’. One of the chief challenges in the ETO space is analyzing the design knowledge encapsulated in digital documents. I approached Dr. Pujari with my insights, and that’s how our collaboration took off. I truly believe our design approach will benefit design teams in the ETO manufacturing process in business.

Q4: Kavita, can you shed light on the advantages of processing in the compressed domain?

Kavita V. Horadi: Absolutely. Processing in the compressed domain offers several benefits, primarily in storage and transmission. By eliminating the decompression stage, we can achieve faster and more efficient processing, making it especially suitable for large-scale document databases.

Q5: Dr. Pujari, your work emphasizes the incorporation of state-of-the-art techniques of object detection. How has this evolved over the years?

Dr. Jagadeesh Pujari: Over the years, object detection has evolved tremendously, driven by advancements in deep learning. By incorporating state-of-the-art techniques, we can pinpoint specific elements in document images, such as tables or graphics, with an accuracy that was previously unattainable.

Q6: Narasimha, with the convergence of pattern recognition and NLP, how do you envision the future of document image processing?

Narasimha Prasad Bhat: I foresee a future where document image processing will be intricately tied with NLP. As we move towards a more digital era, the ability to process text and non-text content efficiently will become paramount. By merging pattern recognition and NLP, we can achieve a holistic understanding of document images.

Q7: For all of you, where do you see the trajectory of your research heading in the next five years?

Kavita V. Horadi: I aim to delve deeper into the complexities of varying document images, particularly those combining printed and handwritten text content present in complex document images which are stored in compressed form.

Dr. Jagadeesh Pujari: My focus will be on further refining the deep learning frameworks and exploring their applicability across a broader spectrum of image types.

Narasimha Prasad Bhat: I plan to bridge the chasm between business needs and technological solutions even more, ensuring that our research has tangible impacts on industries.

Q8: Narasimha, considering the rapid technological advances, how do you believe your work in Document Image Processing will shape the future?

Narasimha Prasad Bhat: As we propel into a world increasingly reliant on digitization, Document Image Processing is poised to be a linchpin. Today, vast amounts of critical information are locked within unstructured document images. Our research aims to unlock this potential, making information more accessible and actionable. Imagine a future where businesses can instantly extract, analyze, and act upon data from a plethora of documents without manual intervention. This not only streamlines operations but also catalyzes informed decision-making. As industries strive for efficiency and agility, our work will be instrumental in redefining how businesses comprehend and utilize their data.

Conclusion:

This enlightening conversation with Kavita V. Horadi, Dr. Jagadeesh Pujari, and Narasimha Prasad Bhat not only offers insights into their groundbreaking research but also underscores the potential of Document Image Processing and Analysis. As we advance technologically, their research will undoubtedly play a pivotal role in shaping the way we understand and process document images. Here are the Google author profiles of these esteemed researchers

Kavita V Horadi: https://scholar.google.co.in/citations?user=gsxLeicAAAAJ&hl=en

Dr Jagadeesh Pujari: https://scholar.google.co.in/citations?user=9I3qnK4AAAAJ&hl=en

Narasimha Prasad Bhat:

https://scholar.google.com/citations?user=mEMs5dMAAAAJ&hl=en&oi=ao