# David Casasent

David Paul Casasent (1942-2015) was an American electrical engineer and computer scientist, professor emeritus at Carnegie Mellon University, known for optical signal processing and pattern recognition, and president of SPIE and INNS.

David Paul Casasent (December 8, 1942 - November 16, 2015) was an American electrical engineer and computer scientist. He served as a professor emeritus in the Department of Electrical and Computer Engineering at [Carnegie Mellon University](https://www.wikiprompt.org/wiki/carnegie-mellon-university), where his research focused on optical image and signal processing. Casasent was president of the International Society for Optical Engineering (SPIE) in 1993 and of the International Neural Network Society in 1999.

Born in Washington, D.C., Casasent earned his B.S., M.S., and Ph.D. degrees from the [University of Illinois](https://www.wikiprompt.org/wiki/university-of-toronto) in 1964, 1965, and 1969, respectively. He joined Carnegie Mellon University as a faculty member in 1969 and was named the George Westinghouse Chair in 1980. He retired in 2009 and died near Pittsburgh, Pennsylvania, on November 16, 2015, survived by his wife and children.

## Optical Pattern Recognition

Casasent's early work, often in collaboration with Demetri Psaltis, established foundational techniques in optical correlation for pattern recognition. In 1976, they published papers on scale-invariant optical correlation using Mellin transforms, demonstrating how optical systems could recognize patterns regardless of scale. Their 1976 paper in Applied Optics extended this to position, rotation, and scale invariant correlation, a significant advance for the field. These methods leveraged the parallel processing capabilities of optics, which were seen as a precursor to modern computational approaches in [machine learning](https://www.wikiprompt.org/wiki/machine-learning).

His later research integrated [neural networks](https://www.wikiprompt.org/wiki/neural-network) with optical processing. A 1995 paper with Leonard M. Neiberg, published in Neural Networks, addressed classifier and shift-invariant automatic target recognition, combining optical correlators with neural network architectures. This work bridged classical optical signal processing and emerging computational intelligence methods.

## Academic Career and Leadership

Casasent spent his entire academic career at Carnegie Mellon University, from 1969 until his retirement in 2009. As the George Westinghouse Chair, he mentored numerous graduate students and contributed to the university's reputation in electrical and computer engineering. His teaching and research spanned optics, signal processing, and pattern recognition, influencing both theoretical developments and practical applications.

His leadership roles extended beyond his university. As SPIE president in 1993 and International Neural Network Society president in 1999, he helped shape professional communities in optics and neural computation. He received the SPIE President's Award in 1996 and was a fellow of SPIE, the IEEE, and the Optical Society of America. He also served as associate editor of the journal Optical Engineering.

## Selected Publications

Casasent authored and edited several influential works. His 1973 book "Electronic Circuits" (Quantum Publishers) served as a textbook. He edited "Optical Data Processing: Applications" (Springer-Verlag, 1978) and "Optical Pattern Recognition" (SPIE Proceedings Vol. 201, 1979), which collected key research in the field.

His journal articles, many with Psaltis, include seminal papers on optical transforms. The 1977 Proceedings of the IEEE paper "New optical transforms for pattern recognition" summarized advances in using Mellin and other transforms for invariant recognition. The 1979 Applied Optics paper on optical residue arithmetic proposed correlation-based approaches to arithmetic operations, showing the breadth of optical computing applications.

Later work with John-Scott Smokelin and Anqi Ye on wavelet and Gabor transforms for detection (Optical Engineering, 1992) connected optical methods to [signal processing](https://www.wikiprompt.org/wiki/data-augmentation) techniques that remain relevant in modern imaging and [artificial intelligence](https://www.wikiprompt.org/wiki/artificial-intelligence) research.

## Legacy

Casasent's contributions lie at the intersection of optics, signal processing, and neural networks. His invariant correlation techniques influenced subsequent work in computer vision and pattern recognition. By integrating optical computing with neural network concepts, he anticipated hybrid approaches that combine physical hardware with algorithmic learning. His leadership in professional societies helped standardize practices and foster collaboration across disciplines.

His work remains cited in fields such as optical computing, automatic target recognition, and [deep learning](https://www.wikiprompt.org/wiki/deep-learning) applications that draw on classical signal processing. Casasent's career exemplifies the transition from analog optical methods to digital and neural approaches, leaving a lasting impact on both theoretical and applied aspects of electrical engineering.

## External links

- [Wikipedia: David Casasent](https://en.wikipedia.org/wiki/David_Casasent)

---
Source: https://www.wikiprompt.org/wiki/david-casasent
License: CC BY-SA 4.0 (https://creativecommons.org/licenses/by-sa/4.0/)
Last updated: 2026-09-14T06:26:17.559749+00:00