In this work, we present a teaching methodology using active-learning techniques in the course “Devices and Instrumentation” of the Erasmus Mundus Master’s Degree in “Color in Informatics and Media Technology” (CIMET). A part of the course “Devices and Instrumentation” of this Master’s is dedicated to the study of image sensors and methods to evaluate their image quality. The teaching methodology that we present consists of incorporating practical activities during the traditional lectures. One of the innovative aspects of this teaching methodology is that students apply the concepts and methods studied in class to real devices. For this, students use their own digital cameras, webcams, or cellphone cameras in class. These activities provide students a better understanding of the theoretical subject given in class and encourage the active participation of students.
We have modified the Fuzzy C-Means algorithm for an application related to segmentation of hyperspectral images.
Classical fuzzy c-means algorithm uses Euclidean distance for computing sample membership to each cluster. We have
introduced a different distance metric, Spectral Similarity Value (SSV), in order to have a more convenient similarity
measure for reflectance information. SSV distance metric considers both magnitude difference (by the use of Euclidean
distance) and spectral shape (by the use of Pearson correlation). Experiments confirmed that the introduction of this
metric improves the quality of hyperspectral image segmentation, creating spectrally more dense clusters and increasing
the number of correctly classified pixels.
The profit of low-cost, multispectral imaging systems in estimating spectral power distributions has been widely studied. There are various mathematical methods available (PCA, Wiener's estimation method, spline interpolation, MDST, among others) which permit the accurate reconstruction of a spectrum from the response of a small set of sensors. One important issue in this task is the influence of noise, its propagation through mathematical transformations and how the selection of the sensors of the multispectral system, combined with the spectral estimation algorithm chosen, may reduce its influence. We report here on four different spectral recovery methods that reconstruct skylight spectra from the responses of three Gaussian sensors (the spectral profile of which is a Gaussian curve). The sensors are searched for using a simulated annealing algorithm, and they are optimized so that they give the best possible spectral and colorimetric reconstructions, even in the presence of noise. We show here how the accuracy of the reconstructions is influenced by the recovery method chosen.
Understanding the properties of time-varying illumination spectra is of importance in all applications where dynamical color changes due to changes in illumination characteristics have to be analyzed or synthesized. Examples are (dynamical) color constancy and the creation of realistic animations. In this article we show how group theoretical methods can be used to describe sequences of time changing illumination spectra with only few parameters. From the description we can also derive a differential equation that describes the illumination changes. We illustrate the method with investigations of black-body radiation and measured sequences of daylight spectra.
We study here the feasibility of a spectral daylight recovering algorithm using a linear model that takes advantage of the strong correlation among daylight curves. To test the algorithm we use the daylight eigenvectors obtained by a principal-value decomposition over 2600 daylight spectra recorded over a period of two years. A binary search, performed here, found the optimal spectral positions of a set of few narrow-band filters. We analyze, over the set of 2600 daylight curves, the algorithm accuracy when using three to six narrow filters, obtaining that such an daylight algorithm is not sufficiently accurate in comparison with similar linear models that recover objects spectral reflectances proposed in artificial-vision.
The objective of the study was to examine the influence of the mean luminance level on the detection thresholds for red-green chromatic gratings of three different spatial frequencies. Data for chromatic sinusoidal gratings with higher mean luminance levels (within the photopic level) than those ones used in previous works were reported. The study analyzed the transition luminance between the DeVries- Rose law and the Weber law regions, and considered the validity of the constant-flux hypothesis for the three spatial frequencies tested. The results suggest that the 'flux' would not be a critical factor in the processing of chromatic gratings in the low spatial frequency range.