Identification of Oil Tea (Camellia oleifera C.Abel) Cultivars Using EfficientNet-B4 CNN Model with Attention Mechanism

Cultivar identification is a basic task in oil tea (Camellia oleifera C.Abel) breeding, quality analysis, and an adjustment in the industrial structure. However, because the differences in texture, shape, and color under different cultivars of oil tea are usually inconspicuous and subtle, the identification of oil tea cultivars can be a significant challenge. The main goal of this study is to propose an automatic and accurate method for identifying oil tea cultivars. In this study, a new deep learning model is built, called EfficientNet-B4-CBAM, to identify oil tea cultivars. First, 4725 images containing four cultivars were collected to build an oil tea cultivar identification dataset. EfficientNet-B4 was selected as the basic model of oil tea cultivar identification, and the Convolutional Block Attention Module (CBAM) was integrated into EfficientNet-B4 to build EfficientNet-B4-CBAM, thereby improving the focusing ability of the fruit areas and the information expression capability of the fruit areas. Finally, the cultivar identification capability of EfficientNet-B4-CBAM was tested on the testing dataset and compared with InceptionV3, VGG16, ResNet50, EfficientNet-B4, and EfficientNet-B4-SE. The experiment results showed that the EfficientNet-B4-CBAM model achieves an overall accuracy of 97.02% and a kappa coefficient of 0.96, which is higher than that of other methods used in comparative experiments. In addition, gradient-weighted class activation mapping network visualization also showed that EfficientNet-B4-CBAM can pay more attention to the fruit areas that play a key role in cultivar identification. This study provides new effective strategies and a theoretical basis for the application of deep learning technology in the identification of oil tea cultivars and provides technical support for the automatic identification and non-destructive testing of oil tea cultivars.

Identification of Cassava Varieties in Ex-Situ Collections and Global Farmer’s Fields: An Update from 1990 to 2020

The identification of cassava cultivars is important for understanding the crop’s production system, enabling crop improvement practitioners to design and deliver tailored solutions with which farmers can secure high yields and sustainable production. Across the lowland tropics today, a large number improved varieties and landraces of cassava are under cultivation, making it inefficient for breeders and geneticists to set improvement goals for the crop. The identification and characterization of cassava genotypes is currently based on either morphological characters or molecular features. The major aim of cultivar identification is to catalog the crop’s genetic diversity, but a consensus approach has still not been established. Of the two approaches to the identification of variety, morphological characters seem to account for most of the genetic variability reported in cassava. However, these characters must be treated with caution, as phenotypic changes can be due to environmental and climatic conditions as well as to the segregation of new highly heterozygous populations, thus, making the accurate identification of varieties difficult. The use of molecular markers has allowed researchers to establish accurate relationships between genotypes, and to measure and track their heterozygous status. Since the early 1990’s, molecular geneticists working with cassava have been developing and deploying DNA-based tools for the identification and characterization of landraces or improved varieties. Hence, in the last five years, economists and social scientists have adopted DNA-based variety identification to measure the adoption rates of varieties, and to support the legal protection of breeder’s rights. Despite the advances made in the deployment of molecular markers for cassava, multiple platform adoption, as well as their costs and variable throughput, has limited their use by practitioners of crop improvement of cassava. The post-genomic era has produced a large number of genome and transcriptome sequencing tools, and has increased our capacity to develop and deploy genome-based tools to account for the crop’s genetic variability by accurately measuring and tracking allele diversity. These technologies allow the creation of haplotype catalogs that can be widely shared across the cassava crop improvement community. Low-density genome-wide SNP markers might be the solution for the wide adoption of molecular tools for the identification of cultivars or varieties of cassava. In this review we survey the efforts made in the past 30 years to establish the tools for cultivar identification of cassava in farmer’s fields and gene banks. We also emphasize the need for a global picture of the genetic diversity of this crop, at its center of origin in South America.

Automated Grapevine Cultivar Identification via Leaf Imaging and Deep Convolutional Neural Networks: A Proof-of-Concept Study Employing Primary Iranian Varieties

Extending over millennia, grapevine cultivation encompasses several thousand cultivars. Cultivar (cultivated variety) identification is traditionally dealt by ampelography, requiring repeated observations by experts along the growth cycle of fruiting plants. For on-time evaluations, molecular genetics have been successfully performed, though in many instances, they are limited by the lack of referable data or the cost element. This paper presents a convolutional neural network (CNN) framework for automatic identification of grapevine cultivar by using leaf images in the visible spectrum (400–700 nm). The VGG16 architecture was modified by a global average pooling layer, dense layers, a batch normalization layer, and a dropout layer. Distinguishing the intricate visual features of diverse grapevine varieties, and recognizing them according to these features was conceivable by the obtained model. A five-fold cross-validation was performed to evaluate the uncertainty and predictive efficiency of the CNN model. The modified deep learning model was able to recognize different grapevine varieties with an average classification accuracy of over 99%. The obtained model offers a rapid, low-cost and high-throughput grapevine cultivar identification. The ambition of the obtained tool is not to substitute but complement ampelography and quantitative genetics, and in this way, assist cultivar identification services.

MFCIS: an automatic leaf-based identification pipeline for plant cultivars using deep learning and persistent homology

Recognizing plant cultivars reliably and efficiently can benefit plant breeders in terms of property rights protection and innovation of germplasm resources. Although leaf image-based methods have been widely adopted in plant species identification, they seldom have been applied in cultivar identification due to the high similarity of leaves among cultivars. Here, we propose an automatic leaf image-based cultivar identification pipeline called MFCIS (Multi-feature Combined Cultivar Identification System), which combines multiple leaf morphological features collected by persistent homology and a convolutional neural network (CNN). Persistent homology, a multiscale and robust method, was employed to extract the topological signatures of leaf shape, texture, and venation details. A CNN-based algorithm, the Xception network, was fine-tuned for extracting high-level leaf image features. For fruit species, we benchmarked the MFCIS pipeline on a sweet cherry (Prunus avium L.) leaf dataset with >5000 leaf images from 88 varieties or unreleased selections and achieved a mean accuracy of 83.52%. For annual crop species, we applied the MFCIS pipeline to a soybean (Glycine max L. Merr.) leaf dataset with 5000 leaf images of 100 cultivars or elite breeding lines collected at five growth periods. The identification models for each growth period were trained independently, and their results were combined using a score-level fusion strategy. The classification accuracy after score-level fusion was 91.4%, which is much higher than the accuracy when utilizing each growth period independently or mixing all growth periods. To facilitate the adoption of the proposed pipelines, we constructed a user-friendly web service, which is freely available at http://www.mfcis.online.

An efficient deep learning model for cultivar identification of a pistachio tree

PurposeThis paper proposes a novel deep learning based method towards the identification of a pistachio tree cultivar from its image.Design/methodology/approachThe investigated scope of this study includes Iranian commercial pistachios (Jumbo, Long, Round and Super long) trees. Effective use of high-resolution images with standard deep models is addressed in this study. A novel image patches extraction method is also used to boost the number of samples and dataset augmentation. In the proposed method, handcrafted ORB features are used to detect and extract patches which may contain identifiable information. An innovative algorithm is proposed for searching and extracting these patches. After extracting patches from initial images, a Convolutional Neural Network, named EfficientNet-B1, was fine-tuned on it. In the testing phase, several patches were extracted from the prompted image using the ORB-based method, and the results of their prediction were consolidated. In this method, patch prediction scores were in descending order, sorted by the highest score in a list, and finally, the average of a few list tops was calculated and the final decision was made.FindingsExamining the proposed method on the test images led to an achievement of a recognition rate of 97.2% accuracy. Investigation of decision-making in the test dataset could reveal that this method outperformed human experts.Originality/valueCultivar identification using deep learning methods, due to their high recognition speed, lack of specialist requirement, and independence from human decision-making error is considered as a breakthrough in horticultural science. Variety cultivars of pistachio trees possess variant characteristics or traits, accordingly recognising cultivars is crucial to reduce the costs, prevent damages and harvest the optimal yields.

Metabolomics for Biomarker Discovery: Key Signatory Metabolic Profiles for the Identification and Discrimination of Oat Cultivars

The first step in crop introduction—or breeding programmes—requires cultivar identification and characterisation. Rapid identification methods would therefore greatly improve registration, breeding, seed, trade and inspection processes. Metabolomics has proven to be indispensable in interrogating cellular biochemistry and phenotyping. Furthermore, metabolic fingerprints are chemical maps that can provide detailed insights into the molecular composition of a biological system under consideration. Here, metabolomics was applied to unravel differential metabolic profiles of various oat (Avena sativa) cultivars (Magnifico, Dunnart, Pallinup, Overberg and SWK001) and to identify signatory biomarkers for cultivar identification. The respective cultivars were grown under controlled conditions up to the 3-week maturity stage, and leaves and roots were harvested for each cultivar. Metabolites were extracted using 80% methanol, and extracts were analysed on an ultra-high performance liquid chromatography (UHPLC) system coupled to a quadrupole time-of-flight (qTOF) high-definition mass spectrometer analytical platform. The generated data were processed and analysed using multivariate statistical methods. Principal component analysis (PCA) models were computed for both leaf and root data, with PCA score plots indicating cultivar-related clustering of the samples and pointing to underlying differential metabolic profiles of these cultivars. Further multivariate analyses were performed to profile differential signatory markers, which included carboxylic acids, amino acids, fatty acids, phenolic compounds (hydroxycinnamic and hydroxybenzoic acids, and associated derivatives) and flavonoids, among the respective cultivars. Based on the key signatory metabolic markers, the cultivars were successfully distinguished from one another in profiles derived from both leaves and roots. The study demonstrates that metabolomics can be used as a rapid phenotyping tool for cultivar differentiation.

Genetic Diversity Assessment and Cultivar Identification of Cucumber (Cucumis sativus L.) Using the Fluidigm Single Nucleotide Polymorphism Assay

Genetic diversity analysis and cultivar identification were performed using a core set of single nucleotide polymorphisms (SNPs) in cucumber (Cucumis sativus L.). For the genetic diversity study, 280 cucumber accessions collected from four continents (Asia, Europe, America, and Africa) by the National Agrobiodiversity Center of the Rural Development Administration in South Korea and 20 Korean commercial F1 hybrids were genotyped using 151 Fluidigm SNP assay sets. The heterozygosity of the SNP loci per accession ranged from 4.76 to 82.76%, with an average of 32.1%. Population genetics analysis was performed using population structure analysis and hierarchical clustering (HC), which indicated that these accessions were classified mainly into four subpopulations or clusters according to their geographical origins. The subpopulations for Asian and European accessions were clearly distinguished from each other (FST value = 0.47), while the subpopulations for Korean F1 hybrids and Asian accessions were closely related (FST = 0.34). The highest differentiation was observed between American and European accessions (FST = 0.41). Nei’s genetic distance among the 280 accessions was 0.414 on average. In addition, 95 commercial F1 hybrids of three cultivar groups (Baekdadagi-, Gasi-, and Nakhap-types) were genotyped using 82 Fluidigm SNP assay sets for cultivar identification. These 82 SNPs differentiated all cultivars, except seven. The heterozygosity of the SNP loci per cultivar ranged from 12.20 to 69.14%, with an average of 34.2%. Principal component analysis and HC demonstrated that most cultivars were clustered based on their cultivar groups. The Baekdadagi- and Gasi-types were clearly distinguished, while the Nakhap-type was closely related to the Baekdadagi-type. Our results obtained using core Fluidigm SNP assay sets provide useful information for germplasm assessment and cultivar identification, which are essential for breeding and intellectual right protection in cucumber.

Apples before the fall: Does shape stability coincide with maturity?

Fruit shape is the result of the interaction between genetic, epigenetic, environmental factors and stochastic processes. As a core biological descriptor both for taxonomy and horticulture, the point at which shape stability is reached becomes paramount in apple cultivar identification, and authentication in commerce. Twelve apple cultivars were sampled at regular intervals from anthesis to harvest over two growing seasons. Linear and geometric morphometrics were analysed to establish if and when shape stabilised and whether fruit asymmetry influenced this. Shape stability was detected in seven cultivars, four asymmetric and three symmetric. The remaining five did not stabilise. Shape stability, as defined here, is cultivar-dependent, and when it occurs, it is late in the growing season. Geometric morphometrics detected stability more readily than linear, especially in symmetric cultivars. Key shape features are important in apple marketing, giving the distinctness and apparent uniformity between cultivars expected at point of sale.