Image dataset of pomegranate fruits (Punica granatum) for various machine vision applications

Dataset - an essential aspect and the requirement for any of the machine learning project. Collection/creation of dataset in the agriculture domain is highly challenging task because the domain itself is uncertain. Main objective of the present paper is to create an image dataset of pomegranate fruits of different grades. Accordingly, we have considered ‘Ruby’ cultivar of pomegranate and sincerely constructed the dataset. Fruits belonging to three grades are considered. The images for each fruit are covered from all the three angles. The dataset created also contains the weights of the fruits. The dataset consists of 12 folders named after their effective quality grades. The usage of this dataset is already proved in the works carried out by the authors in their previous studies. This dataset is highly helpful for the data science engineer / machine learning programmer or machine learning expert working in the field of precision agriculture.

A hybrid optimization-based deep belief neural network for the classification of vegetation area in multi-spectral satellite image

Over the last few decades, multiple advances have been done for the classification of vegetation area through land cover, and land use. However, classification problem is one of the most complicated and contradicting problems that has received considerable attention. Therefore, to tackle this problem, this paper proposes a new Firefly-Harmony search based Deep Belief Neural Network method (FHS-DBN) for the classification of land cover, and land use. The segmentation process is done using Bayesian Fuzzy Clustering,and the feature matrix is developed. The feature matrix is given to the proposed FHS-DBN method that distinguishes the land coverfrom the land use in the multispectral satellite images, for analyzing the vegetation area. The proposed FHS-DBN method is designedby training the DBN using the FHS algorithm, which is developed by the combination of Firefly Algorithm (FA) and Harmony Search (HS) algorithm. The performance of the FHS-DBN model is evaluated using three metrics, such as Accuracy, True Positive Rate (TPR), and False Positive Rate (FPR). From the experimental analysis, it is concludedthat the proposed FHS-DBN model achieves ahigh classification accuracy of 0.9381, 0.9488, 0.9497, and 0.9477 usingIndian Pine, Salinas scene, Pavia Centre and university, and Pavia University scene dataset.

Disease Identification and Grading of Pomegranate Leaves Using Image Processing and Fuzzy Logic

Plant diseases cause major losses to several agricultural and horticultural crops around the World. Therefore, methods for proper diagnosis of diseases found in any parts of the plant body play a crucial role in disease management. In the past few decades, many methods and techniques of image processing and soft computing are applied on a number of plants to diagnose and treat variety of plant diseases. Hence, the present work is aimed to develop an automated system that results in three major outcomes for a leaf image. They are disease identification, disease grading and treatment advisory. The methodology begins with capturing of samples of healthy and diseased leaf images of Pomegranate plant. All the images are made to undergo pre-processing steps and different features are extracted and stored in the database. Analysis is done on the extracted features to determine those features that constitute a disease in the leaf. Later, a query image is taken and is tested to determine whether that image is healthy or diseased one. If the query image is found to be diseased, then the grade of the disease is determined. Finally, a treatment advisory module is built which ultimately helps agriculturists/farmers.