Plant species identification represents an extraordinary challenge for machine learning due to visual interspecies similarities and large intraspecies variations. Furthermore, research literature reports that plant species identification usually lacks sufficiently large datasets for training classification models. In this paper, we address this problem with a model that simplifies the VGG-16 architecture, the N-VGG model. The idea behind N-VGG is to reduce experimentally observed overfitting on VGG-16 by using as few trainable parameters as possible. To do this, we substitute the flattening layer on the VGG architecture with a global average pooling layer. This reduces the size of the feature vector. In addition, we eliminate one of the two fully-connected layers and use a new hyper-parameter, N, to indicate the number of nodes on the remaining layer. To show the robustness of the N-VGG model, we conducted extensive experimentation. We trained N-VGG on five datasets for plant species identification. Four of these datasets are publicly available and have been widely used as benchmarks for plant identification models. For all datasets, we compare the accuracy of N-VGG to that of the VGG-16, Inception-v4, and EfficienNet-B3 models. The experimental results show that the N-VGG model achieved the best classification performance for all but one datasets, whereas all the models showed a remarkable performance for the remaining dataset. This evidence supports our initial idea that, for plant species classification, some accuracy might be lost due to overfitting and that having fewer trainable parameters helps in producing a more robust model.

Resumen: El desarrollo de sistemas de identificación de hojas de plantas es un reto actual que comprende numerosas aplicaciones que van desde alimentación, medicina, industria y medio ambiente. En la literatura actual, se han propuesto varias técnicas con el objetivo de identificar plantas en diversos campos de aplicación. Sin embargo, las técnicas actuales están restringidas al reconocimiento e identificación de tipos de plantas limitados, utilizando descriptores de características específicos. En este artículo, se realiza un análisis comparativo de diversos métodos de extracción de características (texturales, cromáticas y geométricas) y clasificaciÌon sobre conjuntos de plantas muy similares y disimiles entre sí. Doce conjuntos de hojas con características de forma similares son estudiados utilizando varios clasificadores. Se analiza el desempeño de diferentes combinaciones de características en cada conjunto. Los resultados obtenidos muestran que para incrementar el desempeño de los clasificadores estudiados, es necesaria una combinación de las diferentes técnicas de extracción de características, esta necesidad es mayor cuando se trabaja con conjuntos de hojas con características muy similares. Además, se muestra el mejor desempeño de un clasificador con otro. The development of vision systems for identifying plants by leaves is an important challenge which has numerous applications ranging from food, medicine, industry and environment. Recently, several techniques have been proposed in the literature in order to identify plants in various fields of application. However, current techniques are restricted to the recognition and identification of plants using specific descriptors. In this paper, is accomplished a comparative analysis using different methods of feature extraction (textural, chromatic and geometric) and different methods of classification. The experiments are executed on very similar plants. Twelve sets of leaves with similar shape characteristics are studied using several classifiers. The performance of different combinations of classifiers-descriptors are analyzed in detail for each set. The results show that a combination of different feature extraction techniques is necessary in order to improve the performance. This combination of descriptors is more necessary when the leaves have similar characteristics. Palabras clave: Clasificación, Descriptores, SVM, Conjuntos de Datos, Características, Keywords: Classification, Descriptors, SVM, Data Set

Recently, deep-learning techniques have become the foundations for many breakthroughs in the automated identification of plant diseases. In the agricultural sector, many recent visual-computer approaches use deep-learning models. In this approach, a novel predictive analytics methodology to identify Tobacco Mosaic Virus (TMV) and Pepper Huasteco Yellow Vein Virus (PHYVV) visual symptoms on Jalapeño pepper (Capsicum annuum L.) leaves by using image-processing and deep-learning classification models is presented. The proposed image-processing approach is based on the utilization of Normalized Red-Blue Vegetation Index (NRBVI) and Normalized Green-Blue Vegetation Index (NGBVI) as new RGB-based vegetation indices, and its subsequent Jet pallet colored version NRBVI-Jet NGBVI-Jet as pre-processing algorithms. Furthermore, four standard pre-trained deep-learning architectures, Visual Geometry Group-16 (VGG-16), Xception, Inception v3, and MobileNet v2, were implemented for classification purposes. The objective of this methodology was to find the most accurate combination of vegetation index pre-processing algorithms and pre-trained deep- learning classification models. Transfer learning was applied to fine tune the pre-trained deep- learning models and data augmentation was also applied to prevent the models from overfitting. The performance of the models was evaluated using Top-1 accuracy, precision, recall, and F1-score using test data. The results showed that the best model was an Xception-based model that uses the NGBVI dataset. This model reached an average Top-1 test accuracy of 98.3%. A complete analysis of the different vegetation index representations using models based on deep-learning architectures is presented along with the study of the learning curves of these deep-learning models during the training phase.