Characterization of vineyard's canopy through fuzzy clustering and svm over color images

In this work we propose an image acquisition and processing methodology (framework) developed for performance in-field grapes and leaves detection and quantification, based on a six step methodology: 1) image segmentation through Fuzzy C-Means with Gustafson Kessel (FCM-GK) clustering; 2) obtaining of FCM-GK outputs (centroids) for acting as seeding for K-Means clustering; 3) Identification of the clusters generated by K-Means using a Support Vector Machine (SVM) classifier. 4) Performance of morphological operations over the grapes and leaves clusters in order to fill holes and to eliminate small pixels clusters; 5)Creation of a mosaic image by Scale-Invariant Feature Transform (SIFT) in order to avoid overlapping between images; 6) Calculation of the areas of leaves and grapes and finding of the centroids in the grape bunches. Image data are collected using a colour camera fixed to a mobile platform. This platform was developed to give a stabilized surface to guarantee that the images were acquired parallel to de vineyard rows. In this way, the platform avoids the distortion of the images that lead to poor estimation of the areas. Our preliminary results are promissory, although they still have shown that it is necessary to implement a camera stabilization system to avoid undesired camera movements, and also a parallel processing procedure in order to speed up the mosaicking process.

Canopy fruit location can affect olive oil quality in Arbequina hedgerow orchards

The effect of location of fruit in canopies of hedgerow olive trees (Olea europaea L., cv. ‘Arbequina’) on quality of virgin oil was tested by analyzing oils extracted from different height layers and faces of 9 olive hedgerows (6 North-South oriented and 3 East-West). Although sensory attributes were not different other oil quality parameters may be significantly modified by fruit position. In some hedgerows, oils extracted from fruits harvested from higher layers exhibited significantly higher stability against oxidation, along with higher palmitic acid, linoleic acid and phenol contents, but lower oleic acid content. Oils extracted from fruits harvested from East and North facing hedgerows oriented North-South and East-West, respectively, exhibited higher oleic contents and lower saturated and polyunsaturated fatty acid contents. The mean phenol content of oils extracted from fruits from a North-South oriented hedgerow was significantly greater from one of the East-West oriented hedgerows. These findings may be relevant for the design of future olive hedgerows destined for olive oil production.

Canopy fruit location can affect olive oil quality in Arbequina hedgerow orchards.

En este trabajo se recogieron muestras de aceituna procedente de distintas alturas de setos cultivados en diferentes condiciones para evaluar el efecto en la calidad del aceite.

Canopy fruit location can affect olive oil quality in Arbequina hedgerow orchards.

The effect of location of fruit in canopies of hedgerow olive trees (Olea europaea L., cv. ‘Arbequina’) on quality of virgin oil was tested by analyzing oils extracted from different height layers and faces of nine olive hedgerows (6 North–South oriented and 3 East– West). Although sensory attributes were not different, other oil quality parameters may be significantly modified by fruit position. Oils extracted from fruits harvested from higher layers exhibited significantly higher stability against oxidation, along with higher palmitic acid, linoleic acid and phenol contents, but lower oleic acid content. Oils extracted from fruits harvested from East and North facing hedgerows oriented North–South and East–West, respectively, exhibited higher oleic contents and lower saturated and polyunsaturated fatty acid contents. The mean phenol content of oils extracted from fruits from a North–South oriented hedgerow was significantly greater from one of the East–West oriented hedgerows. These findings may be relevant for the design of future olive hedgerows destined for olive oil production.

Assessment of watercore development in apples with MRI: Effect offruit location in the canopy

tWatercore distribution inside apple fruit (block or radial), and its incidence (% of tissue) were relatedto the effect of solar radiation inside the canopy as measured by a set of low-cost irradiation sensors.221 samples were harvested in two seasons from the top and the bottom of the canopy and submittedto the non-invasive and non-destructive technique of magnetic resonance imaging (MRI) in order toobtain 20 inner tomography slices from each fruit and analyze the damaged areas using an interactive3D segmentation method. The number of fruit corresponding to each type of damage and the relevantpercentage were calculated and it was found that apples from the top of the tree were mainly of the radialtype (84%) and had more watercore (approx. 5% more) than apples from the bottom (65% radial). From theimage segmentation, the Euler number, a morphometric parameter, was extracted from the segmentedimages and related to the type of watercore symptoms. Apples with block watercore were grouped inEuler numbers between −400 and 400 with a small evolution. For apples with radial development, theEuler number was highly negative: up to −1439. Significant differences were also found regarding sugarcomposition, with higher fructose and total sugar contents in apples from the upper canopy, compared tothose in the lower canopy location. In the seasons studied (2011 and 2012), significantly higher sorbitoland lower sucrose and fructose contents were found in watercore-affected tissue compared to the healthytissue of affected apples and also compared to healthy apples.

Importance of canopy porosity into vineyard and the relationship with the grape maturity. Digital estimation method

In warm and dry climates, the use of porous systems should be required in order to allow a better leaf distribution inside the plant, causing more space in the clusters area and enhancing determined physiological processes so in the leaf (photosynthesis, v entilation, transpiration) as in berry (growth and maturation). Plant geometry indexes, yield and must composition have been studied in three different systems: sprawl with 12 shoots/m (S1); sprawl system with 18 shoots/m (S2) and vertical positioned syste m or VSP with 12 shoots/m (VSP1). Total leaf area increases as the crop load does, whoever surface area depends on to two factors: crop load and the training system (VSP vs. sprawl), which can provide differences in leaf exposure efficiencies. The main objective of this study was to validate digital photography measurements used to compare porosity differences among treatments and, as they affect plant microclimate and, therefore, yield and berry quality. Also, all previous studied indexes (LAI, SA, SFEr) tended to overestimate the relationship between exposed leaf surface and porosity of each treatment, but the use of digital method proved to be an effective tool in order to assess canopy porosity. Results showed that not positioned and free systems (sprawl) scored between 25- 50% more porosity in the clusters area than the fixed vertical system (VSP), which resulted in a better plant microclimate for test conditions, mainly by improving the exposure of internal clusters and internal canopy ventilation. On the other hand, higher crop load treatment (S2) showed a real increase in yield (16%) without any relevant change into must composition, even improving total anthocyanin content into berry during ripening

Importance of canopy porosity into vineyard and the relationship with the grape maturity. Digital estimation method

In warm and dry climates, the use of porous systems should be required in order to allow a better leaf distribution inside the plant, causing more space in the clusters area and enhancing determined physiological processes so in the leaf (photosynthesis, ventilation, transpiration) as in berry (growth and maturation). Plant geometry indexes, yield and must composition have been studied in three different systems: sprawl with 12 shoots/m (S1); sprawl system with 18 shoots/m (S2) and vertical positioned system or VSP with 12 shoots/m (VSP1). Total leaf area increases as the crop load does, whoever surface area depends on to two factors: crop load and the training system (VSP vs . sprawl), which can provide differences in leaf exposure efficiencies. The main objective of this study was to validate digital photography measurements used to compare porosity differences among treatments and, as they affect plant microclimate and, therefore, yield and berry quality. Also, all previous studied indexes (LAI, SA, SFEr) tended to overestimate the relationship between exposed leaf surface and porosity of each treatment, but the use of digital method proved to be an effective tool in order to assess canopy porosity. Results showed that not positioned and free systems (sprawl) scored between 25 - 50% more porosity in the clusters area than the fixed vertical system (VSP), which resulted in a better plant microclimate for test conditions, mainly by improving the exposure of internal clusters and internal canopy ventilation. On the other hand, higher crop load treatment (S2) showed a real increase in yield (16%) without any relevant change into must composition, even improving total anthocyanin content into berry during ripening

A study on the inclusion of forest canopy morphology data in numerical simulations for the purpose of wind resource assessment

A series of numerical simulations of the flow over a forest stand have been conducted using two different turbulence closure models along with various levels of canopy morphology data. Simulations have been validated against Stereoscopic Particle Image Velocimetry measurements from a wind tunnel study using one hundred architectural model trees, the porosities of which have been assessed using a photographic technique. It has been found that an accurate assessment of the porosity of the canopy, and specifically the variability with height, improves simulation quality regardless of the turbulence closure model used or the level of canopy geometry included. The observed flow field and recovery of the wake is in line with characteristic canopy flows published in the literature and it was found that the shear stress transport turbulence model was best able to capture this detail numerically.