G. Polder

SPECTRAL IMAGE ANALYSIS FOR MEASURING RIPENESS OF TOMATOES

The present invention provides transport apparatus for transporting and articulating an article into and out of a housing having limited access, and includes a lifting beam having a curved shape to which is attached a lifting assembly. A hoist device is connected to the lifting assembly for raising and lowering the entire transport apparatus and also includes a worm-drive assembly for moving the lifting assembly relative to the lifting beam to thereby articulate an article about a first axis. In addition, one end of the lifting beam includes a rotational mechanism which is connected to a second worm-drive assembly. Operation of the second worm-drive assembly allows the article to be rotated relative to the lifting beam which allows the article to be rotated relative to a second axis. The combined operations of hoisting and movement about the two axes allows the article to be articulated in any direction for insertion or removal from a housing having limited access.

Tomato sorting using independent component analysis on spectral images

Independent Component Analysis is one of the most widely used methods for blind source separation. In this paper we use this technique to estimate the most important compounds which play a role in the ripening of tomatoes. Spectral images of tomatoes were analyzed. Two main independent components were found. These components resemble the actual absorption spectra of lycopene and chlorophyll. Concentration images of these compounds show increase of one compound and decrease of the other during ripening. The method can be implemented in an unsupervised real time sorting machine, using the total compound concentrations and the spatial distribution of the concentrations as criteria.

Calibration and characterization of spectral imaging systems

Spectral image sensors provide images with a large umber of contiguous spectral channels per pixel. This paper describes the calibration of spectrograph based spectral imaging systems. The relation between pixel position and measured wavelength was determined using three different wavelength calibration sources. Results indicate that for spectral calibration a source with very small peaks,such as a HgAr source, is preferred to arrow band filters. A second order polynomial model gives a better fit than a linear model for the pixel to wavelength mapping. The signal to noise ratio (SNR)is determined per wavelength. In the blue part of the spectrum,the SNR was lower than in the green and red part.This is due to a decreased quantum efficiency of the CCD,a smaller transmission coefficient of the spectrograph,as well as poor performance of the illuminant. Increasing the amount of blue light,using additional Fluorescent tube with special coating increased the SNR considerably. Furthermore, the spatial and spectral resolution of the system are determined.These can be used to choose appropriate binning factors to decrease the image size without losing information.

Visualization of spectral images

Spectral image sensors provide images with a large number of contiguous spectral channels per pixel. Visualization of these huge data sets is not a straightforward issue. There are three principal ways in which spectral data can be presented; as spectra, as image and in feature space. This paper describes several visualization methods and their suitability in the different steps in the research cycle. Combinations of the three presentation methods and dynamic interaction between them, adds significant to the usability. Examples of some software implementations are given. Also the application of volume visualization methods to display spectral images is shown to be valuable.

Measuring onion cultivars with image analysis using inflection points

SummaryThe suitability of image analysis was studied to measure bulb characteristics for varietal testing of onions (Allium cepa L.). Eighteen genotypes were used, which covered a whole range of onion shapes, including some quite identical ones. The characteristic height and diameter were measured both by image analysis and by hand. The methods gave comparable results with regard to precision and discriminating power. The discriminating power of the image analysis method could be increased by adding a new characteristic: the relative position of the diameter.For determining the height of the bulb with image analysis two methods are described to locate the top of a bulb. One method is to use the location of the inflection point as the top of the bulb. For most bulb shapes, this point coincides with the top of the bulb as indicated by a crop expert. For ovate shaped onions, however, the inflection point is below the top of the bulb as indicated by the crop expert. The other method is to use the location of the intercept of the tangent at the inflection point with the length axis of the bulb as the top of the bulb. This point is always above the top as indicated by the crop expert. Moreover it seems to perform slightly worse for the discrimination of cultivars. Therefore the location of the inflection point as the top of the bulb is preferred.

Identification of mushroom cultivars using image analysis

The application of image analysis for variety testing of mushrooms [Agaricus bisporus (Lange) Imbach] to grant plant breeders rights was investigated. Images of 230 mushrooms were recorded under controlled lighting conditions. Image analysis techniques were used to increase the number of morphological characters that can be measured. The measurements of length, width, and a range of more or less complex shape descriptors, determined by means of image analysis, were statistically analyzed. Image analysis exhibited a significant improvement over previous methods which used only manual assessments.

Application of image analysis for variety testing of mushroom

SummaryThe application of image analysis for variety testing of mushrooms to grant plant breeders rights was investigated. The measurements of length, width and a range of several other more or less complex shape descriptors determined by means of image analysis were statistically analyzed. A total of 460 observations were used in this experiment. Distinction between 80% of the cultivars in this experiment could be made, at a 1% significance level, using only 4 characters of 44 characters tested. Image analysis methods are a faster and more accurate manner to distinguish cultivars than the standard visual and manual ones presently in use.

Comparison of multispectral images across the Internet

Comparison in the RGB domain is not suitable for precise color matching, due to the strong dependency of this domain on factors like spectral power distribution of the light source and object geometry. We have studied the use of multispectral or hyperspectral images for color matching, since it can be proven that hyperspectral images can be made independent of the light source and object geometry. Hyperspectral images have the disadvantages that they are large compared to regular RGB-imags, which makes it infeasible to use them for image matching across the Internet. For red roses, it is possible to reduce the large number of bands of the spectral images to only three bands, the same numbers of an RGB-image, using Principal Component Analysis, while maintaining 99 percent of the original variation. The obtained PCA-images of the roses can be matched using for example histogram cross correlation. From the principal coordinates plot, obtained from the histogram similarity matrices of twenty images of red roses, the discriminating power seems to be better for normalized spectral images than for color constant spectral images and RGB-images, the latter being recorded under highly optimized standard conditions.