Detection of grapes in real-life images is a serious task solved by researchers dealing with precision viticulture. In the case of white wine varieties, grape detectors based on SVMs classifiers, in combination with a HOG descriptor, have proven to be very efficient. Simplified versions of the detectors seem to be the best solution for practical applications. They offer the best known performance versus time-complexity ratio. As our research showed, a conversion of RGB images to grayscale format, which is implemented at an image preprocessing level, is ideal means for further improvement of performance of the detectors. In order to enhance the ratio, we explored relevance of the conversion in a context of a detector potential sensitivity to a rotation of berries. For this purpose, we proposed a modification of the conversion, and we designed an appropriate method for a tuning of such modified detectors. To evaluate the effect of the new parameter space on their performance, we developed a specialized visualization method. In order to provide accurate results, we formed new datasets for both tuning and evaluation of the detectors. Our effort resulted in a robust grape detector which is less sensitive to image distortion.
The resolving power of multicomponent spectral analysis and the computation reliability of the stability constants and molar absorptivities determined for five variously protonated anions of physostigmine salicylate by the SQUAD(84) and SPECFIT/32 programs has been examined with the use of simulated and experimental spectra containing overlapping spectral bands. The reliability of the dissociation constants of drug was proven with goodness-of-fit tests and by examining the influence of pre-selected noise level s(inst)(A) in synthetic spectra regarding the precision s(pK) and also accuracy of the estimated dissociation constants. Precision was examined as the linear regression model s(pK)=β(0)+β(1)s(inst)(A). In all cases the intercept β(0) was statistically insignificant. When an instrumental error s(inst)(A) is small and less than 0.5 mAU, the parameters' estimates are nearly the same as the bias ΔpK=pK(a,calc)-pK(a,true) is quite negligible. In all four dissociation constants the bias seems to be quite small even though for pK(a4) it is a little bit higher, i.e., +0.05 for s(inst)(A) about 1.0 mAU. In the interval of s(inst)(A) from 0.1 to 1.0 mAU all four dissociation constants pK(i) are accurate enough. Of the various regression diagnostics considered, the goodness-of-fit is the most efficient criterion of whether the parameters found adequately represent the data. The magnitude of instrumental error s(inst)(A) only slightly affects the shape of a Cattel's scree graph s(k)(A)=f(k) to determine the true number of light-absorbing species in the equilibrium mixture.
- MeSH
- Absorption MeSH
- Models, Chemical MeSH
- Physostigmine analysis chemistry MeSH
- Kinetics MeSH
- Hydrogen-Ion Concentration MeSH
- Computer Simulation MeSH
- Surface Properties MeSH
- Protons MeSH
- Regression Analysis MeSH
- Reproducibility of Results MeSH
- Software MeSH
- Spectrophotometry methods MeSH
- Thermodynamics MeSH
- Titrimetry methods MeSH
- Publication type
- Journal Article MeSH
- Research Support, Non-U.S. Gov't MeSH
