|
|
vol. 11, nr. 1 (2008)
|
Image analysis of apple tissue cells after mechanical deformation
|
|
|
Justyna Cybulska1, Artur Zdunek1, Roman Konskyy2, Krystyna Konstankiewicz1
|
|
| (get PDF )
|
|
|
1 Institute of Agrophysics, Polish Academy of Sciences, ul. Doświadczalna 4, 20-290 Lublin |
|
|
2 Karpenko Physico-Mechanical Institute of the National Academy of Sciences of Ukraine, ul. Naukova 5, Lwów 79601 |
|
|
vol. 11 (2008), nr. 1,
pp. 57-69
|
|
|
abstract:
Texture of apples depends on geometrical dimensions of the cellular mechanical skeleton. For quantitative measurements of the dimensions of cells an image analysis procedure should be developed which is proper to a certain imaging system. In this paper two different image analysis methods are applied: image segmentation and visual texture analysis (VTA) to images obtained by an optical microscope for which a special sample fixation procedure for apple tissue was developed. The image analysis was performed for thin sections of apple, cut by microtome from samples previously subjected to different strain levels: intact tissue, 5%, 10% and 15%. The experiment showed that segmentation of images from the optical microscope is possible after prior manual correction of the images of apple, while visual texture analysis (VTA) provides quantitative results for the size and shape of objects in the images automatically. The parameters obtained by the VTA correlate significantly with geometrical parameters obtained by the segmentation method. The highest correlation was obtained for product of the size of the horizontal and vertical linear structural element. Geometric dimensions of objects in the images increases with increase of the tissue strain level when observation plane is transversal to the direction of deformation. The effect can be affected by large amount of intercellular spaces in the apple, the volume of which decreases during deformation.
|
|
keywords:
segmentation, visual texture analysis, apple
|
|
original in:
English
|
|
|
|
|
|
