MUCKE aims to mine a large volume of images, to structure them conceptually and to use this conceptual structuring in order to improve large-scale image retrieval. The last decade witnessed important progress concerning low-level image representations. However, there are a number problems which need to be solved in order to unleash the full potential of image mining in applications. The central problem with low-level representations is the mismatch between them and the human interpretation of image content. This problem can be instantiated, for instance, by the incapability of existing descriptors to capture spatial relationships between the concepts represented or by their incapability to convey an explanation of why two images are similar in a content-based image retrieval framework. We start by assessing existing local descriptors for image classification and by proposing to use co-occurrence matrices to better capture spatial relationships in images. The main focus in MUCKE is on cleaning large scale Web image corpora and on proposing image representations which are closer to the human interpretation of images. Consequently, we introduce methods which tackle these two problems and compare results to state of the art methods. Note: some aspects of this deliverable are withheld at this time as they are pending review. Please contact the authors for a preview.

http://ieeexplore.ieee.org/iel2/4524/12820/00592460.pdf

Image Processing

Suppressive composts represent a sustainable approach to combat soilborne plant pathogens and an alternative to the ineffective chemical fungicides used against those. Nevertheless, suppressiveness to plant pathogens and reliability of composts are often inconsistent with unpredictable effects. While suppressiveness is usually attributed to the compost's microorganisms, the mechanisms governing microbial recruitment by the roots and the composition of selected microbial communities are not fully elucidated. Herein, the purpose of the study was to evaluate the impact of a compost on tomato plant growth and its suppressiveness against Fusarium oxysporum f. sp. lycopersici (Foxl) and Verticillium dahliae (Vd). First, growth parameters of tomato plants grown in sterile peat-based substrates including 20 and 30% sterile compost (80P/20C-ST and 70P/30C-ST) or non-sterile compost (80P/20C and 70P/30C) were evaluated in a growth room experiment. Plant height, total leaf surface, and fresh and dry weight of plants grown in the non-sterile compost mixes were increased compared to the plants grown in the sterile compost substrates, indicating the plant growth promoting activity of the compost's microorganisms. Subsequently, compost's suppressiveness against Foxl and Vd was evaluated with pathogenicity experiments on tomato plants grown in 70P/30C-ST and 70P/30C substrates. Disease intensity was significantly less in plants grown in the non-sterile compost than in those grown in the sterile compost substrate; AUDPC was 2.3- and 1.4-fold less for Foxl and Vd, respectively. Moreover, fungal quantification in planta demonstrated reduced colonization in plants grown in the non-sterile mixture. To further investigate these findings, we characterized the culturable microbiome attracted by the roots compared to the unplanted compost. Bacteria and fungi isolated from unplanted compost and the rhizosphere of plants were sequence-identified. Community-level analysis revealed differential microbial communities between the compost and the rhizosphere, suggesting a clear effect of the plant in the microbiome assembly. Proteobacteria and Actinobacteria were highly enriched in the rhizosphere whereas Firmicutes were strongly represented in both compartments with Bacillus being the most abundant species. Our results shed light on the composition of a microbial consortium that could protect plants against the wilt pathogens of tomato and improve plant overall health.

/pdf/rhizosphere-microbiome-recruited-from-a-suppressive-compost-3jd4c4h7ac.pdf

Rhizosphere Microbiome Recruited from a Suppressive Compost Improves Plant Fitness and Increases Protection against Vascular Wilt Pathogens of Tomato.

Apple (Malus spp.) is a widely grown and valuable fruit crop. Leaf shape is important for flowering in apple and may also be an early indicator for other agriculturally valuable traits. We examined 9,000 leaves from 869 unique apple accessions using linear measurements and comprehensive morphometric techniques. We identified allometric variation as the result of differing length-to-width aspect ratios between accessions and species of apple. The allometric variation was due to variation in the width of the leaf blade, not the length. Aspect ratio was highly correlated with the first principal component (PC1) of morphometric variation quantified using elliptical Fourier descriptors (EFDs) and persistent homology (PH). While the primary source of variation was aspect ratio, subsequent PCs corresponded to complex shape variation not captured by linear measurements. After linking the morphometric information with over 122,000 genome-wide single nucleotide polymorphisms (SNPs), we found high SNP heritability values even at later PCs, indicating that comprehensive morphometrics can capture complex, heritable phenotypes. Thus, techniques such as EFDs and PH are capturing heritable biological variation that would be missed using linear measurements alone.

/pdf/morphometrics-reveals-complex-and-heritable-apple-leaf-4v7v34ui52.pdf

Morphometrics Reveals Complex and Heritable Apple Leaf Shapes.

Following recent advances in bioimaging, high-resolution 3D models of biological structures are now generated rapidly and at low-cost. To use this data to address evolutionary and ecological questions, an array of tools has been developed to conduct shape analysis and quantify topographic complexity. Here we focus particularly on shape techniques applied to irregular-shaped objects lacking clear homologous landmarks, and propose a new ‘alpha-shapes’ method for quantifying 3D shape complexity. We apply alpha-shapes to quantify shape complexity in the mammalian baculum as an example of a morphologically disparate structure. Micro- computed-tomography (μCT) scans of bacula were conducted. Bacula were binarised and converted into point clouds. Following application of a scaling factor to account for absolute size differences, a suite of alpha-shapes was fitted per specimen. An alpha shape is formed from a subcomplex of the Delaunay triangulation of a given set of points, and ranges in refinement from a very coarse mesh (approximating convex hulls) to a very fine fit. ‘Optimal’ alpha was defined as the refinement necessary in order for alpha-shape volume to equal CT voxel volume, and was taken as a metric of overall ‘complexity’. Our results show that alpha-shapes can be used to quantify interspecific variation in shape ‘complexity’ within biological structures of disparate geometry. The ‘stepped’ nature of alpha curves is informative with regards to the contribution of specific morphological features to overall ‘complexity’. Alpha-shapes agrees with other measures of complexity (dissection index, Dirichlet normal energy) in identifying ursid bacula as having low shape complexity. However, alpha-shapes estimates mustelid bacula as being most complex, contrasting with other shape metrics. 3D fractal dimension is identified as an inappropriate metric of complexity when applied to bacula. Alpha-shapes is used to calculate ‘optimal’ alpha refinement as a proxy for shape ‘complexity’ without identifying landmarks. The implementation of alpha-shapes is straightforward, and is automated to process large datasets quickly. We interpret alpha-shapes as being particularly sensitive to concavities in surface topology, potentially distinguishing it from other shape complexity metrics. Beyond genital shape, the alpha-shapes technique holds considerable promise for new applications across evolutionary, ecological and palaeoecological disciplines.

/pdf/alpha-shapes-determining-3d-shape-complexity-across-ur9hqhuljm.pdf

Alpha shapes: determining 3D shape complexity across morphologically diverse structures.

Image-based phenotyping methodologies are powerful tools to determine quality parameters for fruit breeders and processors. The fruit size and shape of American cranberry (Vaccinium macrocarpon L.) are particularly important characteristics that determine the harvests' processing value and potential end-use products (e.g., juice vs. sweetened dried cranberries). However, cranberry fruit size and shape attributes can be difficult and time consuming for breeders and processors to measure, especially when relying on manual measurements and visual ratings. Therefore, in this study, we implemented image-based phenotyping techniques for gathering data regarding basic cranberry fruit parameters such as length, width, length-to-width ratio, and eccentricity. Additionally, we applied a persistent homology algorithm to better characterize complex shape parameters. Using this high-throughput artificial vision approach, we characterized fruit from 351 progeny from a full-sib cranberry population over three field seasons. Using a covariate analysis to maximize the identification of well-supported quantitative trait loci (QTL), we found 252 single QTL in a 3-year period for cranberry fruit size and shape descriptors from which 20% were consistently found in all years. The present study highlights the potential for the identified QTL and the image-based methods to serve as a basis for future explorations of the genetic architecture of fruit size and shape in cranberry and other fruit crops.

/pdf/image-based-phenotyping-for-identification-of-qtl-4h08tymxdm.pdf

Image-based phenotyping for identification of QTL determining fruit shape and size in American cranberry (Vaccinium macrocarpon L.).

Recent studies have demonstrated that grapevine (Vitis spp.) leaf shape can be quantified using digital approaches which indicate phylogenetic signal in leaf shape, discernible patterns of developmental context within single leaves, and signatures of local environmental conditions. Here, we extend this work by quantifying intra-individual, intraspecific, and interspecific variation in leaf morphology in accessions of North American V. riparia and V. rupestris in a common environment. For each species at least four clonal replicates of multiple genotypes were grown in the Missouri Botanical Garden Kemper Center for Home Gardening. All leaves from a single shoot were harvested and scanned leaf images were used to conduct generalized Procrustes analysis, linear discriminant analysis, and elliptical fourier analysis. Leaf shapes displayed genotype-specific signatures and species distinctions consistent with taxonomic classifications. Leaf shape variation within genotypes and among clones was the result of pest and pathogen-induced leaf damage that alters leaf morphology. Significant trends in leaf damage caused by disease and infestation were non-random with respect to leaf position on the shoot. Digital morphometrics is a powerful tool for assessing leaf shape variation among species, genotypes, and clones under common conditions and suggests biotic factors such as pests and pathogens as important drivers influencing leaf shape.

/pdf/digital-morphometrics-of-two-north-american-grapevines-vitis-23hxd6ubpz.pdf

Digital Morphometrics of Two North American Grapevines (Vitis: Vitaceae) Quantifies Leaf Variation between Species, within Species, and among Individuals.

Chad Chapnick

Papers

Digital Morphometrics of Two North American Grapevines (Vitis: Vitaceae) Quantifies Leaf Variation between Species, within Species, and among Individuals.