Direct Optical Computation Of Linear Discriminants For Color Recognition

Caulfield, H. John; Mueller, P.

doi:10.1117/12.7973245

Cited by 12 publications

(13 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The processes we implement amount to nothing more than thresholding linear discriminants in the spectral domain. We showed two decades ago that linear discriminants can be implemented in optical hardware [6,7] and doing them with current electronic hardware is also easy. On the positive side, a software controlled smart camera is easily reprogrammed to perform new tasks.…”

Section: Introductionmentioning

confidence: 99%

Making a smarter color camera

Caulfield

2008

Image and Vision Computing

View full text Add to dashboard Cite

Section: Introductionmentioning

confidence: 99%

Making a smarter color camera

Caulfield

2008

Image and Vision Computing

View full text Add to dashboard Cite

“…Caulfield [13] proposed a method for color spectrum classification using an optical computing technique. Soon after that Jaaskelainen et al [14] proposed a realization of learning subspace method by optical computing.…”

mentioning

confidence: 99%

Subspace approach in spectral color science

Parkkinen

Laamanen

Hauta-Kasari

2015

Advances in Independent Component Analysis and Learning Machines

View full text Add to dashboard Cite

“…This dot product can be calculated either explicitly by MDCs or implicitly MOCs. In previous applications of MOC, the ILS/LD model was either implicitly assumed or explicitly applied [1,[9][10][11][12].…”

Section: Implementation Of Generalized Multivariate Optical Computatimentioning

confidence: 99%

“…This agreement is remarkable considering that two of the assumptions implicit in employing Equation (9) to predict relative precision between OCs and DCs were violated. Equation (9) assumes that the measurement errors are identically distributed for all wavelengths exhibiting a mean of 0 and a finite standard deviation that is independent of signal intensity. The measurement errors were independent of signal intensity and demonstrated the same distribution for all wavelengths.…”

Section: Spr Sensormentioning

confidence: 99%

“…In practice these devices are much simpler than pseudo-electronic optical computers and are consequently limited to specialized applications. For example, Caulfield and Mueller employed a linear variable filter for predictive classification of colored glasses [9]. In 1985, Jansen and Harris employed optical computations to calculate the two-dimensional second moment of a laser beam waste to determine focus diameter during thermal lens experiments [10].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Generalization of multivariate optical computations as a method for improving the speed and precision of spectroscopic analyses

Boysworth

Banerji

Wilson

et al. 2008

Journal of Chemometrics

View full text Add to dashboard Cite

Multivariate optical computations (MOCs) offer improved analytical precision and increased speed of analysis via synchronous data collection and numerical computation with scanning spectroscopic systems. The improved precision originates in the redistribution of integration time from spurious channels to informative channels in an optimal manner for increasing the signal-to-noise ratio with multivariate analysis under the constraint of constant total analysis time. In this work, MOCs perform the multiplication and addition steps of spectral processing by adjusting the integration parameters of the optical detector or adjusting the scanning profile of the tunable optical filter. Improvement in the precision of analysis is achieved via the implicit optimization of the analytically useful signal-to-noise ratio. The speed improvements are realized through simpler data post-processing, which reduces the computation time required after data collection. Alternatively, the analysis time may be significantly truncated while still seeing an improvement in the precision of analysis, relative to competing methods. Surface plasmon resonance (SPR) spectroscopic sensors and visible reflectance spectroscopic imaging were used as test beds for assessing the performance of MOCs. MOCs were shown to reduce the standard deviation of prediction by 15% compared to digital data collection and analysis with the SPR and up to 45% for the imaging applications. Similarly, a 30% decrease in the total analysis time was realized while still seeing precision improvements.

show abstract

Direct Optical Computation Of Linear Discriminants For Color Recognition

Cited by 12 publications

References 0 publications

Making a smarter color camera

Making a smarter color camera

Subspace approach in spectral color science

Generalization of multivariate optical computations as a method for improving the speed and precision of spectroscopic analyses

Contact Info

Product

Resources

About