Identification Of Olive Cultivars Research Articles

Deep learning networks for olive cultivar identification: A comprehensive analysis of convolutional neural networks

Deep learning networks, more specifically convolutional neural networks, have shown a notable distinction when it comes to computer vision problems. Their versatility spans various domains, where they are applied for tasks such as classification and regression, contingent primarily on the availability of a representative dataset. This work explores the feasibility of employing this approach in the domain of agriculture, particularly within the context of olive growing. The objective is to enhance and facilitate cultivar identification techniques by using images of olive tree leaves. To achieve this, a comparative analysis involving ten distinct convolutional networks (VGG16, VGG19, ResNet50, ResNet152-V2, Inception V3, Inception ResNetV2, XCeption, MobileNet, MobileNetV2, EfficientNetB7) was conducted, all initiated with transfer learning as a common starting point. Also, the impact of adjusting network hyperparameters and structural elements was explored. For the training and evaluation of the networks, a dedicated dataset was created and made available, consisting of approximately 4200 images from the four most representative categories of the region. The findings of this study provide compelling evidence that the majority of the examined methods offer a robust foundation for cultivar identification, ensuring a high level of accuracy. Notably, the first nine methods consistently attain accuracy rates surpassing 95%, with the top three methods achieving an impressive 98% accuracy (ResNet50, EfficientNetB7). In practical terms, out of approximately 2016 images, 1976 were accurately classified. These results signify a substantial advancement in olive cultivar identification through computer vision techniques.

Discrimination of 14 olive cultivars using morphological analysis and machine learning algorithms.

Traditional morphological analysis is a widely employed tool for the identification and discrimination of olive germplasm by using morphological markers which are monitored by subjective manual measurements that are labor intensive and time-consuming. Alternatively, an automated methodology can quantify the geometrical features of fruits, leaves and endocarps with high accuracy and efficiency in order to define their morphological characteristics. In this study, 24 characteristics for fruits, 16 for leaves and 25 for endocarps were determined and used in an automated way with basic classifiers combined with a meta-classsifier approach. This resulted to the discrimination of 14 olive cultivars utilizing data obtained from two consecutive olive growing periods. The cultivar classification algorithms were based on machine learning techniques. The 95% accuracy rate of the meta-classifier approach indicated that was an efficient tool to discriminate olive cultivars. The contribution of each morphological feature to cultivar discrimination was quantified, and the significance of each one was automatically detected in a quantitative way. The higher the contribution of each feature, the higher the significance for cultivar discrimination. The identification of most cultivars was guided by the features of both endocarps and fruits, while those of leaves were only efficient to identify the Kalamon cultivar. The combined use of morphological features of three olive organs might have an additive effect leading to higher capacity for discrimination of cultivars. The proposed methodology might be considered a phenomics tool for olive cultivar identification and discrimination in a wide range of applications including breeding.

Classification of olive cultivars by machine learning based on olive oil chemical composition.

Extra virgin olive oil traceability and authenticity are important quality indicators, and are currently the subject of exhaustive research, for developing methods to secure olive oil origin-related issues. The aim of this study was the development of a classification model capable of olive cultivar identification based on olive oil chemical composition. To achieve our aim, 385 samples of two Greek and three Italian olive cultivars were collected during two successive crop years from different locations in the coastline part of western Greece and southern Italy and analyzed for their chemical characteristics. Principal Component Analysis showed trends of differentiation among olive cultivars within or between the crop years. Artificial intelligence model of the XGBoost machine learning algorithm showed high performance in classifying the five olive cultivars from the pooled samples.

Characterization of Olive (Olea Europaea L.) Genetic Resources via PCR-Based Molecular Marker Systems

Olea europaea L., which is one of the ancient culture species cultivated in the Mediterranean area, has approximately 1200 cultivars. Its wild thype forms from O. europaea subsp. europaea var. sylvestris and its culture type forms from O. europaea subsp. europaea var. europaea. Olive cultivation is multiplied by grafting or cutting whereas wild type olive seedlings derived from seeds of O. europaea L. var. sylvestris. Because they have very big level heterozygosities, the genetic diversity of olive cultivars is quite high that they are predominantly allogamus. This genetic variability causes many confusions for identification of olive cultivars and since both oil quality and olive productivity are traits inherited to a variety, it is urgently needed to solve characterization and evaluation of olive genetic resources. The molecular marker systems are independent from environmental factors and effective technology to both detect homonymous and synonymous of cultivars and identify olive varieties. The current review is aimed to present molecular marker systems for identification and characterization of olive cultivars and emphasize their application for conservation of olive germplasm.

Computer vision-based dimension measurement system for olive identification

Olive tree is an important portion of the human history of Mediterranean nations. On the other hand, local varieties are important for each producer regions and even countries. So, olive cultivars are important for agricultural production for these people. The traditional pomological identifiers of the olive trees based on fruits and leaves, also morphometric analysis of size, additionally shape elliptic analysis of endocarp. Because of this reason, in this study, for the ‘Picual’ olive cultivar identification, a fast and easy method was presented. For this aim, ‘Picual’ olive leaf, fruit, and stone dimension measurement values obtained from real-time video images. ‘Picual’ olive fruit, stone, leaf samples evaluated by using real-time computer vision measurements. Regression analysis was applied to the data which was obtained from real-time video and caliper measurements. According to the regression coefficient results, the regression between caliper length measurement (OLLM) and computer vision video length measurement (OLLCV) found as 98.9%, also the regression between caliper width measurement (OLWM) and computer vision video width measurement (OLWCV) found as 97.9% at ‘Picual’ leaves, additionally, the regression between caliper length measurement (OFLM) and computer vision video length measurement (OFLCV) found as 98.5% the regression between caliper width measurement (OFWM) and computer vision video width measurement (OFWCV) found as 98.1 % at ‘Picual’ fruits, at last, the regression between caliper length measurement (OSLM) and computer vision video length measurement (OSLCV) found as 86.2%, the regression between caliper width measurement (OSWM) and computer vision video width measurement (OSWCV) found as 75.3% at ‘Picual’ stones.

The newly developed genomic-SSR markers uncover the genetic characteristics and relationships of olive accessions.

BackgroundOlive (Olea europaea L.) is an important oil and fruit crop worldwide, owning a rich germplasm with a large number of cultivars. Simple sequence repeats (SSRs) are excellent markers and have been used for the identification of olive cultivars. However, the limited number of SSR markers and the occurrence of confusion on the names of cultivars, as well as the possible appearance of clonal variation make it difficult to identify cultivars and interpret relationships among olive cultivars.MethodSSR markers were designed based on trinucleotide repeat sequences by screening the whole genome of olive, and the polymorphic SSR markers were developed that were applied to the identification of 53 olive accessions. The genetic characteristics and relationships of these olive accessions were evaluated based on the developed SSR markers.ResultsTwenty-one highly polymorphic genomic-SSR markers were developed, covering most chromosomes of olive. These SSR markers could well distinguish all 53 olive accessions, confirming their effectiveness. DNA fingerprints of the 53 olive accessions were constructed based on the 21 SSR markers. The dendrogram clearly divided the tested accessions into two main groups, which was also supported by the results of principal coordinate analysis. A total of 31 private alleles were detected in 15 olive accessions, which reflected the genetic diversity within 53 olive accessions to some extent. Six homonymy cases were also clarified by genetic analysis. These results suggest that the newly developed olive SSR markers are informative for the exploitation, preservation and breeding of olive.

Usefulness of a New Large Set of High Throughput EST-SNP Markers as a Tool for Olive Germplasm Collection Management.

Germplasm collections are basic tools for conservation, characterization, and efficient use of olive genetic resources. The identification of the olive cultivars maintained in the collections is an important ongoing task which has been performed by both, morphological and molecular markers. In the present study, based on the sequencing results of previous genomic projects, a new set of 1,043 EST-SNP markers has been identified. In order to evaluate its discrimination capacity and utility in diversity studies, this set of markers was used in a representative number of accessions from 20 different olive growing countries and maintained at the World Olive Germplasm Collection of IFAPA Centre ‘Alameda del Obispo’ (Córdoba, Spain), one of the world’s largest olive germplasm bank. Thus, the cultivated material included: cultivars belonging to previously defined core collections by means of SSR markers and agronomical traits, well known homonymy cases, possible redundancies previously identified in the collection, and recently introduced accessions. Marker stability was tested in repeated analyses of a selected number of accessions, as well as in different trees and accessions belonging to the same cultivar. In addition, 15 genotypes from a cross ‘Picual’ × ‘Arbequina’ cultivars from the IFAPA olive breeding program and a set of 89 wild genotypes were also included in the study. Our results indicate that, despite their relatively wide variability, the new set of EST-SNPs displayed lower levels of genetic diversity than SSRs in the set of olive core collections tested. However, the EST-SNP markers displayed consistent and reliable results from different plant material sources and plant propagation events. The EST-SNPs revealed a clear cut off between inter- and intra-cultivar variation in olive. Besides, they were able to reliably discriminate among different accessions, to detect possible homonymy cases as well as efficiently ascertain the presence of redundant germplasm in the collection. Additionally, these markers were highly transferable to the wild genotypes. These results, together with the low genotyping error rates and the easy and fully automated procedure used to get the genotyping data, validate the new set of EST-SNPs as possible markers of choice for olive cultivar identification.

A computer vision approach based on endocarp features for the identification of olive cultivars

The identification of olive cultivars is of utmost importance for a multitude of factors affecting both, the olive oil elaboration process and fair trade exchanges. The accurate varietal identification is a time consuming task that requires trained specialists or expensive and specific equipment. When applying the traditional method, a specialist assesses morphological features using the olive endocarp. A proposal to automate this identification method is presented in this paper. Endocarp images, acquired under three different perspectives, are processed to extract the same information that the specialist utilizes. Then, the partial least squares discriminant analysis classifier, with or without feature selection, has been tested on a set of 250 samples from 5 different varieties. Results show that the proposal is an alternative identification method which could also be used in the traditional one in order to assist the specialist in the determination of the variety.

Genotypic and phenotypic identification of olive cultivars from north-western Spain and characterization of their extra virgin olive oils in terms of fatty acid composition and minor compounds

Abstract Galicia (NW Spain) is emerging as a new olive-growing region. Galician oil producers are currently striving to recover old autochthonous cultivars with a view to obtaining high quality extra virgin olive oil (EVOO). In this work, a total of 32 trees were studied in order to established their identity and genetic relationships to the main cultivated material in the Iberian Peninsula. The analysis of 11 morphological features of the endocarp and 14 microsatellite markers allowed three different cultivars to be identified among the sampled trees. Comparison with the morphological and molecular profiles available in the World Olive Germplasm Bank of Cordoba (WOGBC) revealed that 24 trees (75%) were of the ‘Brava’ cultivar and 7 (22%) of the ‘Mansa’ cultivar. The other tree, labelled as Picuda, matched no specific cultivar in WOGBC. Characterizing the oils obtained from the studied cultivars revealed a high potential for producing high-quality EVOOs of specific origin.

Identification of olive cultivars using image processing techniques

As a country in an olive production zone, olive (Olea europaea L.) production is of great economic importance for Turkey. Turkey has 88 domestic and 28 foreign cultivars of olive, all of which are located in the Kemalpasa Production and Research Garden (Olive Gene Bank) at the Olive Research Station in Izmir (38°25'34.7628''N, 27°25'22.9872''E). The aim of this study was the identification of olive cultivars using image processing techniques. For this aim, images of olives, taken at 2896 x 1944 pixel and 300 dpi resolution, were captured using a DSLR camera, and evaluations of pixels were used for considering the pixel distribution and dimension measurements. For this aim, MATLAB v2012 and Image j software were used. In the light of the obtained data, analysis of variance and Duncan?s test were used to characterize the olive cultivars. As a result, all observed olive cultivars were identified at P < 0.05 significance level. Sari ulak, Gemlik, Edincik su, Memecik, Esek zeytini, Ayvalik, Kilis yaglik, Uslu, Cilli, and Domat olive cultivars were identified utilizing the validation process employing image processing and analysis techniques. Only the Erkence cultivar was not identified. Moreover, different classification techniques were applied to the olive stone color value data by the help of SPSS v22 and Clementine v12, which is a data mining software package from SPSS. In addition to the first results the Erkence cultivar was identified 69% with artificial neural networks.

OGDD (Olive Genetic Diversity Database): a microsatellite markers' genotypes database of worldwide olive trees for cultivar identification and virgin olive oil traceability.

Olive (Olea europaea), whose importance is mainly due to nutritional and health features, is one of the most economically significant oil-producing trees in the Mediterranean region. Unfortunately, the increasing market demand towards virgin olive oil could often result in its adulteration with less expensive oils, which is a serious problem for the public and quality control evaluators of virgin olive oil. Therefore, to avoid frauds, olive cultivar identification and virgin olive oil authentication have become a major issue for the producers and consumers of quality control in the olive chain. Presently, genetic traceability using SSR is the cost effective and powerful marker technique that can be employed to resolve such problems. However, to identify an unknown monovarietal virgin olive oil cultivar, a reference system has become necessary. Thus, an Olive Genetic Diversity Database (OGDD) (http://www.bioinfo-cbs.org/ogdd/) is presented in this work. It is a genetic, morphologic and chemical database of worldwide olive tree and oil having a double function. In fact, besides being a reference system generated for the identification of unkown olive or virgin olive oil cultivars based on their microsatellite allele size(s), it provides users additional morphological and chemical information for each identified cultivar. Currently, OGDD is designed to enable users to easily retrieve and visualize biologically important information (SSR markers, and olive tree and oil characteristics of about 200 cultivars worldwide) using a set of efficient query interfaces and analysis tools. It can be accessed through a web service from any modern programming language using a simple hypertext transfer protocol call. The web site is implemented in java, JavaScript, PHP, HTML and Apache with all major browsers supported.Database URL: http://www.bioinfo-cbs.org/ogdd/

ISSR Analysis for Determination of Genetic Diversity and Relationship in Eight Turkish Olive (Olea europaea L.) Cultivars

Olive cultivar identification is very important especially for fruit productivity and olive oil quality. Up to date, numerous techniques have been used for characterization of olive cultivars and detection of genetic variability. In the present study, 40 clones specific to Turkey (8 cultivars, each having 5 clones) were assessed for cultivar determination via inter simple sequence repeat (ISSR) marker systems using 10 ISSR primers. In total, 217 band profiles were obtained, 206 of which were polymorphic (94.9% polymorphism). The cultivars were segregated into two main clusters, each cluster being subdivided into two groups, while all the clones of a cultivar belong to the same main cluster. The only exception was the distribution of the clones of cultivar 'Tekir': 'Tekir 1', 'Tekir 2' and 'Tekir 3' on cluster III, while 'Tekir 4' and 'Tekir 5' were on cluster IV, therefore grouped into different main clusters. In the present study, ISSR analysis displayed a high level of genetic variability among Turkish olive cultivars, indicating a potential resource for the use of this germplasm in clonal selection programs.

ISSR Analysis for Determination of Genetic Diversity and Relationship in Eight Turkish Olive (Olea europaea L.) Cultivars

Olive cultivar identification is very important especially for fruit productivity and olive oil quality. Up to date, numerous techniques have been used for characterization of olive cultivars and detection of genetic variability. In the present study, 40 clones specific to Turkey (8 cultivars, each having 5 clones) were assessed for cultivar determination via inter simple sequence repeat (ISSR) marker systems using 10 ISSR primers. In total, 217 band profiles were obtained, 206 of which were polymorphic (94.9% polymorphism). The cultivars were segregated into two main clusters, each cluster being subdivided into two groups, while all the clones of a cultivar belong to the same main cluster. The only exception was the distribution of the clones of cultivar 'Tekir': 'Tekir 1', 'Tekir 2' and 'Tekir 3' on cluster III, while 'Tekir 4' and 'Tekir 5' were on cluster IV, therefore grouped into different main clusters. In the present study, ISSR analysis displayed a high level of genetic variability among Turkish olive cultivars, indicating a potential resource for the use of this germplasm in clonal selection programs.

11th Euro Fed Lipid Congress, New Strategies for a High Quality Future (27–30 October 2013, Antalya, Turkey)

11th Euro Fed Lipid Congress, New Strategies for a High Quality Future (27–30 October 2013, Antalya, Turkey)

Characterisation and identification of olive cultivars from North-eastern Algeria using molecular markers

SummaryIn the present study, ten microsatellite (simple sequence repeat; SSR) and seven single nucleotide polymorphism (SNP) markers were used to characterise a set of 14 cultivars of olive (Olea europaea L.) from North-eastern Algeria. The genetic profiles of these cultivars were compared with two Tunisian, one Moroccan, and two Algerian cultivars. A total of 77 alleles were amplified for the SSR markers. The mean values of the expected heterozygosity (He), the observed heterozygosity (Ho), and the polymorphic information content (PIC) values were 0.806, 0.851, and 0.752, respectively. A combination of three SSR loci (ssrOeUA-DCA9, ssrOeUA-DCA18, and UDO099-043) permitted the identification of all 14 olive accessions analysed. Although less polymorphic than the SSR markers, the SNP markers used in this study showed relatively high diversity values. The mean He value was 0.374. The same value was obtained for the Ho, and PIC values ranged from 0.373 (Cycl-1) to 0.117 (LS2). Eleven out of all 19 olive cultivars (63%) were identified by means of the SNP markers, confirming the value of the differences from SSR markers in terms of their efficiency for cultivar discrimination and stressing the need to develop more SNP markers. The results obtained in this work highlight the rich genetic diversity available in even a small set of olive germplasm from Algeria and confirm the utility of SSR and SNP markers for genotyping olive cultivars.

SNP-based markers for discriminating olive (Olea europaea L.) cultivars

A set of 11 polymorphic markers (1 cleaved amplified polymorphic sequence (CAPS), 2 sequence-characterized amplified regions (SCARs), and 8 single-nucleotide polymorphism (SNP)-derived markers) was obtained for olive cultivar identification by comparing DNA sequences from different accessions. Marker development was more efficient, using sequences from the database rather than cloning arbitrary DNA fragments. Analyses of the sequences of 3 genes from 11 diverse cultivars revealed an SNP frequency of 1 per 190 base pairs in exons and 1 per 149 base pairs in introns. Most mutations were silent or had little perceptible effect on the polypeptide encoded. The higher incidence of transversions (55%) suggests that methylation is not the major driving force for DNA base changes. Evidence of linkage disequilibrium in 2 pairs of markers has been detected. The set of predominantly SNP-based markers was used to genotype 65 olive samples obtained from Europe and Australia, and was able clearly to discriminate 77% of the cultivars. Samples, putatively of the same cultivar but derived from different sources, were revealed as identical, demonstrating the utility of these markers as tools for resolving nomenclature issues. Genotyping data were used for constructing a dendrogram by UPGMA cluster analysis using the simple matching similarity coefficient. Relationships between cultivars are discussed in relation to the route of olive's spread.

EFFICIENCY OF RAPD MARKERS FOR VARIETAL IDENTIFICATION IN OLIVE (OLEA EUROPAEA L.) GERMPLASM COLLECTIONS

The aim of this study was to evaluate the resolving power of RAPD markers for rapid and reliable identification of olive cultivars in germplasm collections. The D parameter (the probability that two randomly chosen cultivars have different banding patterns), used for that purpose, showed high values for most of the 21 primers used and its values ranged from 0.6114 (OPI-13) to 0.9762 (OPK-16) with a mean value of 0.8566. This parameter was used to select the five most discriminating primers: OPK-16, OPA-19, OPX-09, OPF-06 and OPZ-11. The joint confusion probability and the statistical number of undistinguishable pairs of cultivars were estimated for these primers (under independence hypothesis). The combination of three primers (OPK-16, OPA-19 and OPX-09) was found optimal for rapid discrimination of 103 cultivars with a very low value of cumulative confusion probability (1.72 x 10-5), leaving 0.0903 pairs of cultivars undistinguishable. This fact together with the efficiency of the most discriminating primers combination on an increasing number of cultivars proved the utility of RAPD markers for varietal identification of olive cultivars in collections and in nurseries.

Characterization and Identification of the Main Spanish and Italian Olive Cultivars by Simple-sequence-repeat Markers

Nine simple-sequence-repeat (SSR) primer pairs were assayed in 35 Spanish and Italian olive cultivars of commercial interest. All microsatellites were polymorphic, showing 5 to 13 alleles per locus (7.5 alleles per locus on average). The frequency of each alleles was generally low, with most of the alleles present at one or two cultivars. Heterozigosity ranged from 0.15 to 0.95; the discrimination power (PD) ranged from 0.30 to 0.93 (mean 0.79). The set of microsatellites analyzed discriminated all cultivars investigated. The combination of only three SSR primer pairs—UDO99-009+UDO99-043+UDO99-14—made possible the identification of all cultivars included in the study. Cluster analysis did not find differences between Spanish and Italian cultivars, but most of the cultivars from southern and central Spain grouped together. Hence, microsatellites markers are recommended for olive fingerprinting to generate a database for olive cultivar identification.

Characterization and Identification of Spanish Olive Germplasm by Means of RAPD Markers

Thirteen randomly amplified polymorphic DNA (RAPD) primers were assayed in 82 Spanish olive cultivars of economical interest. A total of 82 bands were scored giving an average of 6.3 bands per primer. A total of 4 (OPA-01) to 10 bands (OPA-19) was amplified, while the number of polymorphic fragments ranged from 2 (OPK-07) to 9 (OPA-19) with a mean of 5.7 polymorphic bands per primer. A total of 89% of the amplification products (73 bands) were polymorphic. The 13 primers yielded 184 banding patterns (14.9 per primer). The number of banding patterns per primer ranged from 4 (OPK-07) to 39 (OPA-19). Fifty-three unique banding patterns were found, the majority of them resulted from different combinations of polymorphic bands. The combination of only five primers OPA-19, OPF-06, OPX-01, OPX-03, and OPI-12, allowed identification of all the cultivars. Seventy-four cultivars (90%) were identified only by the combination of the first four primers. The addition of the fifth primer (OPI-12) was necessary for the identification of the eight remaining cultivars (10%). The ordination of the primers according to their practical discriminating capacity in this study was: OPA-19 &gt; OPF-06 &gt; OPX-01 &gt; OPX-03 &gt; OPI-12. Hence RAPD markers are recommended for olive fingerprinting in order to generate a database for olive cultivar identification.

Optimal Use of RAPD Markers for Identifying Varieties in Olive (Olea europaea L.) Germplasm Collections

The aim of this work was to study in depth the resolving power of RAPD markers for rapid and reliable identification of olive cultivars in germplasm collections. The D parameter (the probability that two randomly chosen cultivars have different banding patterns), used for that purpose, showed high values for most of the 21 primers tested and its values ranged from 0.6114 (OPI-13) to 0.9762 (OPK-16) with a mean value of 0.8566. This parameter was used to select the five most discriminating primers: OPK-16, OPA-19, OPX-09, OPF-06 and OPZ-11. The joint confusion probability and the statistical number of indistinguishable pairs of cultivars were estimated for these primers (under independence hypothesis). The combination of three primers (OPK-16, OPA-19 and OPX-09) was found optimal for rapid discrimination of 103 cultivars with a very low value of cumulative confusion probability (1.72 × 10-5), leaving 0.09 pairs of cultivars indistinguishable. This fact, together with the efficiency of the most discriminating primers combination on an increasing number of cultivars, evidenced the utility of RAPD markers for discrimination of olive cultivars in collections and in nurseries.