Fig 6 - uploaded by Javier Hernández-Andrés
Content may be subject to copyright.
Examples of reconstructions of some standard CIE illuminants using four vectors of the global basis. (a) Illuminant A. GFC 0.999425. Solid curve: illuminant A; points: reconstruction. (b) Illuminant D 65. GFC 0.999657. Solid curve: illuminant D 65 ; points; reconstruction. (c) Illuminant F 2. GFC 0.997532. Solid curve: illuminant F 2 ; points; reconstruction. (d) Illuminant F 11. GFC 0.999904. Solid curve: illuminant F 11 ; points: reconstruction.
Source publication
By the principal-value decomposition process, we have obtained two linear bases for representing the spectral power distributions of illuminants, applicable for algorithms of color synthesis and analysis in artificial vision: one from experimental measurements of daylight and another combining both natural and artificial illuminants. The first basi...
Context in source publication
Similar publications
This paper focuses on the performance analysis of a camera-based vehicle-to-vehicle visible light communication system employing undersampled phase shift on-off keying modulation
under interference scenario. Two Nissan Qashqai front lights with daylight running light emitting diodes based lamps are used for communications. The bit error rate (BER)...
Soil color is usually described under natural daylight using Munsell charts. Daylight can vary but its effects on the Munsell notation are hardly known. Today, color-appearance models allow quantitative analyses of the perceived color under different lights. Using the CIECAM02 model, we studied the color changes in 238 Munsell chips and 229 soils u...
Citations
... al., selecting one measure for each of the groups they describe can be sufficient to asses the spectral similarity. 24 For that reason, we decided to use the Root Mean Square Error (RMSE), the Goodness-of-Fit Coefficient (GFC), 25 and the Spectral Angle Mapper (SAM) 26 metrics. On one hand, RMSE can range from 0 to 1 since the maximum value of our data is 1, indicating a value of RMSE equal to 0 perfect similarity. ...
Significance
Hyperspectral imaging sensors have rapidly advanced, aiding in tumor diagnostics for in-vivo brain tumors. Linescan cameras effectively distinguish between pathological and healthy tissue, while snapshot cameras offer a potential alternative to reduce acquisition time.
Aim
Our research compares linescan and snapshot hyperspectral cameras for in-vivo brain tissues and chromophores identification.
Approach
We compared a lines-can pushbroom camera and a snapshot camera using images from 10 patients with various pathologies. Objective comparisons were made using unnormalized and normalized data for healthy and pathological tissues. We utilized Interquartile Range (IQR) for the Spectral Angle Mapping (SAM), the Goodness-of-Fit Coefficient (GFC), and the Root Mean Square Error (RMSE) within the 659.95 to 951.42 nm range. Additionally, we assessed the ability of both cameras to capture tissue chromophores by analyzing absorbance from reflectance information.
Results
The SAM metric indicates reduced dispersion and high similarity between cameras for pathological samples, with a 9.68% IQR for normalized data compared to 2.38% for unnormalized data. This pattern is consistent across GFC and RMSE metrics, regardless of tissue type. Moreover, both cameras could identify absorption peaks of certain chromophores. For instance, using the absorbance measurements of the linescan camera we obtained SAM values below 0.235 for four peaks, regardless of the tissue and type of data under inspection. These peaks are: one for cytochrome b in its oxidised form at λ = 422 nm, two for HbO2 at λ = 542 nm and λ = 576 nm, and one for water at λ = 976 nm.
Conclusion
The spectral signatures of the cameras show more similarity with unnormalized data, likely due to snapshot sensor noise, resulting in noisier signatures post-normalization. Comparisons in this study suggest that snapshot cameras might be viable alternatives to linescan cameras for real-time brain tissues identification.
... Compared with th methods, the histograms of the proposed method are smooth and close to the truth. The goodness-of-fit coefficient (GFC) is useful for numerically evaluating hi distributions [24]. This measure is the correlation coefficient between the predic ...
... Compared with the other methods, the histograms of the proposed method are smooth and close to the ground truth. The goodness-of-fit coefficient (GFC) is useful for numerically evaluating histogram distributions [24]. This measure is the correlation coefficient between the predicted and target histogram curves. ...
This study considers a method for reconstructing a high dynamic range (HDR) original image from a single saturated low dynamic range (LDR) image of metallic objects. A deep neural network approach was adopted for the direct mapping of an 8-bit LDR image to HDR. An HDR image database was first constructed using a large number of various metallic objects with different shapes. Each captured HDR image was clipped to create a set of 8-bit LDR images. All pairs of HDR and LDR images were used to train and test the network. Subsequently, a convolutional neural network (CNN) was designed in the form of a deep U-Net-like architecture. The network consisted of an encoder, a decoder, and a skip connection to maintain high image resolution. The CNN algorithm was constructed using the learning functions in MATLAB. The entire network consisted of 32 layers and 85,900 learnable parameters. The performance of the proposed method was examined in experiments using a test image set. The proposed method was also compared with other methods and confirmed to be significantly superior in terms of reconstruction accuracy, histogram fitting, and psychological evaluation.
... With a CGFC mean value of 0.02 and a 90 th percentile value of 0.05, there is a significant correlation between estimated and ground truth spectra. Despite not meeting the 'good' spectral reproduction threshold of CGFC-value ≤ 0.01 on average, as defined by [61], the model still achieves a promising best CGFC value of 0.0026. Given the complexities of no-reference illuminant estimation, achieving perfect results across all scenes and conditions is improbable with a singular algorithm. ...
This paper introduces a novel framework for estimating the spectral power distribution of daylight illuminants in uncalibrated hyperspectral images, particularly beneficial for drone-based applications in agriculture and forestry. The proposed method uniquely combines image-dependent plausible spectra with a database of physically possible spectra, utilizing an image-independent principal component space (PCS) for estimations. This approach effectively narrows the search space in the spectral domain and employs a random walk methodology to generate spectral candidates, which are then intersected with a pre-trained PCS to predict the illuminant. We demonstrate superior performance compared to existing statistics-based methods across various metrics, validating the framework’s efficacy in accurately estimating illuminants and recovering reflectance values from radiance data. The method is validated within the spectral range of 382–1002 nm and shows potential for extension to broader spectral ranges.
... This function was initially used to assess the similarity of the light spectra. The GFC is an adaptation of the cosine similarity tailored for spectral evaluation without considering the sign [39], as defined by the following expression: ...
In this paper, a method for creating texture models with physical significance using a parametric equalizer (PEQ)-based approach is presented. The model is based on the frequency, amplitude, bandwidth, and noise ratio corresponding to texture profiles. Additionally, the method simplifies the modeling process by enabling the selective omission of imperceptible roughness layers based on human tactile thresholds. The effectiveness of the simplified model was validated through subjective evaluations comprising both absolute and relative assessments. The results confirmed that layers with imperceptible roughness could be excluded without compromising perceived similarity, streamlining the texture modeling process. Regression analysis revealed that PEQ parameters reflect physical interactions in tactile sensations, such as the relationship between texture roughness and vibration amplitude. This study contributes to haptic texture modeling by offering a method that efficiently reproduces actual textures and mirrors the fundamental physical principles of touch. The findings hold promise for applications in texture authoring and material selection, indicating potential advancements in the development of more intuitive and physically transparent texture models.
... The concentration vector that results in minimal distance from R λ is the corresponding solution for that pixel. The unmixing method has been implemented in Matlab© using constrained nonlinear optimization with the interior-point algorithm [33] and the cGFC distance metric [34] . This is a metric that is sensitive mostly to shape changes in the spectrum, and it has proven to work more effectively than RMSE in preliminary trials of the method. ...
The cleaning of re-polychromated plaster is a rather challenging task for Conservators. The optimal choice of solvents and its application instruments may depend on the original polychromy (binder and pigment) that aims to be revealed and the overlay polychromy (binder and pigment) as well. We present a new image-processing based method that uses hyperspectral imaging to obtain a non-invasive quantitative evaluation of cleaning procedures. It is based on the application of unmixing techniques for this problem. The method computes three quality indices: efficiency, destructivity, and alteration, that are based on the residual amounts of overlay pigment, the detection of excessive abrasion in the sample and the change in color that the cleaning produces. It produces as well pixel-by-pixel concentration maps and presence maps for plaster and overlay pigment, and it can self-evaluate by computing the error map comparing the modelled spectral reflectance with the measured spectral reflectance from the samples.
... Compared with the other methods, we can see that the histograms of the proposed method is smooth and close to the ground truth. The goodness-fitting coefficient (GFC) is useful for evaluating the histogram distributions [16] numerically. This measure is a kind of correlation coefficient between the predicted and target histogram curves. ...
... For the cost function, usually spectral metrics are used, such as Mean Square Error (MSE [ 17 ]) or the complement of the Goodnessof-Fit coefficient (cGFC [ 53 ]), related to the Spectral Angle Mapper metric (SAM). MSE is influenced by offset differences, and cGFC only accounts for shape differences. ...
... They ensure that all the factors would contribute equally to the metric value in an acceptable reflectance match. The tolerance values are based on experience and correspond with tolerances found in the literature for spectral estimation [ 53 ]. The metric contains a color difference term because color can be relevant for pigment identification, while both scale and shape differences in spectra are also accounted for. ...
Hyperspectral imaging has recently consolidated as a useful technique for pigment mapping and identification, although it is commonly supported by additional non-invasive analytical methods. Since it is relatively rare to find pure pigments in aged paintings, spectral unmixing can be helpful in facilitating pigment identification if suitable mixing models and endmember extraction procedures are chosen. In this study, a subtractive mixing model is assumed, and two approaches are compared for endmember extraction: one based on a linear mixture model, and the other, nonlinear and Deep-Learning based. Two spectral hyperspaces are used: the spectral reflectance (R hyperspace) and the -log(R) hyperspace, for which the subtractive model becomes additive. The performance of unmixing is evaluated by the similarity of the estimated reflectance to the measured data, and pigment identification accuracy. Two spectral ranges (400 to 1000 nm and 900 to 1700 nm) and two objects (a laboratory sample and an aged painting, both on copper) are tested. The main conclusion is that unmixing in the -log(R) hyperspace with a linear mixing model is better than for the non-linear model in R hyperspace, and that pigment identification is generally better in R hyperspace, improving by merging the results in both spectral ranges.
... The method for calculating standard daylights was not designed to be used for creating daylights with a correlated color temperature less than 4000 K. Although lower values CCTs are possible at sunrise and sunset (and are found in measured daylights [20]). Creating daylights for correlated color temperatures lower than 4000 K is made possible with our method. ...
... In the Granada set of daylights [20] there is a daylight spectrum that has a CCT of 3888 K (with respect to which we cannot calculate a standard daylight spectrum). In Figure 6 we plot the spectral power distribution of this measured daylight and the SPD calculated with our correction filter using Eq. ...
... Let us now consider how the CCTs for the 99 (complete set of) measured daylights in Granada [20]-which do not lie on the daylight locus-shift when they are filtered by a locus filter. First, let us measure how far these lights are from the daylight locus built using our new daylight Eq. (15). ...
... Taking into account all curves, the root mean squared error (RMSE) between the training and ECOSTRESS curves is equal to 0.14%. To further express the similarity between vectors x and y, one can use the goodness-of-fit coefficient (GFC) [65], which is defined as GFC = |x T y|/(||x|| 2 ||y|| 2 ). For the five pairs of curves in Fig. 4, the average GFC is equal to 0.999123, which qualifies as a good fit [66]. ...
This paper addresses the multispectral filter design problem for spectral ranges where a viewing subspace is not defined. The methodology of color filter design is extended to this case, which allows the optimization of custom filter transmittance that meets the physical constraints of available fabrication methods. Multispectral shortwave infrared filters are then designed for two scenarios: spectral reconstruction and false-color representation. The Monte Carlo method is used to verify the filter performance degradation due to deviations in fabrication. The results obtained indicate that the proposed method is useful for designing multispectral filters to be fabricated using generic processes without any additional constraints.
... The most common is to take the average JM distance computed for all pairs of classes. Deborah et al. [38] evaluated the performance of four different distance functions named Root Mean Square Error (RMSE), Goodness-of-Fit Coefficient (GFC) [39], Jeffrey divergence, and Levenshtein distance on both synthetic and real hyperspectral datasets to find a suitable distance measure for spectral image processing. They found that for the magnitude change, only RMSE followed by Jeffrey divergence performed in the desired way. ...
