Charlie Bouman, Showalter Professor in the Elmore Family School of Electrical and Computer Engineering and Weldon School of Biomedical Engineering, recently was named the 2021 recipient of the Claude Shannon-Harry Nyquist Technical Achievement Award. He is only the third person at Purdue to receive the honor.

Charles Bouman

Presented by the Institute of Electrical and Electronics Engineers (IEEE) Signal Processing Society (SPS), the award recognizes recipients “on the basis of outstanding technical contributions to theory and/or practice in technical areas within the scope of the society, as demonstrated by publications, patents, or recognized impact on the field,” according to the IEEE website. Bouman was selected by a panel of judges “for fundamental contributions to X-ray computed tomography and computational imaging.”

He is considered a pioneer in the field of computational imaging, which blends statistical signal processing, applied mathematics, physical modeling, and applications to create images from raw sensor data. Bouman’s technologies are used in printers, CT scanners, cell phones and security and scientific applications.

In the 1990s, he created the first algorithm (Resolution Synthesis) that has been used for image scaling in millions of HP inkjet printers for more than 20 years. His hardware-friendly de-screening is in all low-end HP multifunction printers (MFP), and he introduced the use of discrete hidden Markov models (HMM) on trees for the processing of images in applications such as image segmentation.

He spearheaded the application of model-based iterative reconstruction (MBIR) in CT images that resulted in a major advance in medical imaging technology. In 2011, GE Healthcare released “Veo”, the first commercially available MBIR reconstruction for medical CT. The device, which was touted to “reduce X-ray dosage by 78%,” was based on Bouman’s technology breakthrough.

He collaborated with the Air Force Research Laboratory to bring signal processing to the materials science field and general inverse problems, resulting in MBIR for transmission electron microscopy (TEM) imaging, TIMBIR-4D, and digital holographic MBIR. In addition, he introduced the Sparse Matrix Transform, which will enable on-board computation in the NASA NACHOS CubeSat to observe atmospheric chemistry.

Bouman introduced a novel mathematical framework known as Plug-and-Play for integrating machine learning models with more traditional physic-based sensor models. His 2016 TCI paper on this topic won the 2020 SIAM Imaging Sciences Best Paper Prize for its mathematical contributions. This method was used to make accurate through-the-screen fingerprint detection possible from ultrasonic measurements on the new Samsung Galaxy S20.

A fellow of the National Academy of Inventors, Bouman holds 60 U.S. patents. In 2015, he formed High Performance Imaging, LLC, which develops technology for high-performance computation of images, such as the patented super-voxel MBIR algorithm.

He co-founded the Computational Imaging Conference as part of the Information Systems & Technology (IS&T) Electronic Imaging Symposium, which he has chaired for the last 20 years. He also has served as vice president of publications and general co-chair of the Electronic Imaging Symposium.

Within SPS, he has served in a number of capacities. He is a fellow of the American Institute for Medical and Biological Engineering (AIMBE), IEEE, IS&T, and the Society of Photo-Optical Instrumentation Engineers (SPIE). He has supervised 44 PhD students, currently supervises 12, and developed a rich imaging science curriculum at Purdue.

Founded as IEEE’s first society in 1948, SPS is the world’s premier association for signal processing engineers and industry professionals. Its membership includes more than 19,000 signal processing engineers, academics, industry professionals and students who are all part of a dynamic global community – spanning 100 countries worldwide.