Applied Soft Computing 36 (2015) 45–56 Contents lists available at ScienceDirect Applied Soft Computing j ourna l h o mepage: www.elsevier.com/locate/asoc Fuzzy classification of pre-harvest tomatoes for ripeness estimation – An approach based on automatic rule learning using decision tree Nidhi Goel a,∗ , Priti Sehgal b a Department of Computer Science, University of Delhi, New Delhi, India b Keshav Mahavidyalaya, University of Delhi, New Delhi, India a r t i c l e i n f o Article history: Received 3 December 2014 Received in revised form 12 June 2015 Accepted 13 July 2015 Available online 23 July 2015 Keywords: Fuzzy rule based classification Tomato ripeness Decision trees Automatic rule learning Machine vision Image retrieval a b s t r a c t Tomato (Solanum lycopersicum) ripeness estimation is an important process that affects its quality eval- uation and marketing. However, the slow speed, subjectivity, time consumption associated with manual assessment has been forcing the agriculture industry to apply automation through robots. The vision sys- tem of harvesting robot is responsible for two-tasks. The first task is the recognition of object (tomato) and second is the classification of recognized objects (tomatoes). In this paper, Fuzzy Rule-Based Classi- fication approach (FRBCS) has been proposed to estimate the ripeness of tomatoes based on color. The two color depictions: red-green color difference and red-green color ratio are derived from extracted RGB color information. These are then compared as a criterion for classification. Fuzzy partitioning of the feature space into linguistic variables is done by means of a learning algorithm. A rule set is automatically generated from the derived feature set using Decision Trees. Mamdani fuzzy inference system is adopted for building the fuzzy rule based classification system that classifies the tomatoes into six maturity stages. Dataset used for experiments has been created using the real images that were collected from a farm. 70% of the total images were used for training and 30% images of the total were used for testing the dataset respectively. Training dataset is divided into six classes representing the six different stages of tomato ripeness. Experimental results showed the system achieved the ripeness classification accuracy of 94.29% using proposed FRBCS. © 2015 Elsevier B.V. All rights reserved. 1. Introduction India is an agriculture-based country. Agriculture plays a crucial role in the economy and is the prime source for country’s national income. Fundamental factor responsible for consistent marketing of crops is its quality. For many crops, the main indicator of quality is its ripeness. The consumer (wholesaler or retailer) observes the quality of fresh fruits and vegetables from their visual or external appearance. The visual appearance of the crop is used to judge its ripeness, which is measured, by color, size, and shape. Out of these three factors, color is the most important factor. It has high influ- ence on quality and consumers’ preference. For many agricultural products, certain colors are preferred and demand higher selling prices. This is true for apples [1], broccoli [2], and cranberries [3]. Color is one of the most commonly used feature to evaluate matu- rity for various fruits, vegetables like tomatoes [4], watermelons ∗ Corresponding author. Tel.: +91 9953803900. E-mail addresses: nidhi hansraj@yahoo.co.in (N. Goel), psehgal25.08@gmail.com (P. Sehgal). [5], bananas [6], and dates [7]. Harvesting of fruits and vegetables at proper stage of maturity is of paramount significance for attaining desirable quality. The level of maturity helps in estimation of shelf life, selection of storage methods, and selection of processing oper- ations for value addition. The maturity has been divided into two categories i.e. physiological maturity and horticultural maturity [8]. Horticultural (pre-harvest) maturity refers to the stage of develop- ment when a crop is ready for harvest. Physiological (post-harvest) maturity is the stage when a crop is capable of further development or ripening after it is harvested i.e. ready for eating or processing. Quality characteristics such as flavor, texture, and color are sus- tained when the fruit is harvested at an optimal stage of maturity. Therefore, ripeness monitoring and controlling has become a very important issue in crop industry. Tomato is one of the most important food crops in India, which is marketed all over the country. It is a climacteric fruit i.e. it con- tinues ripening even after it has been harvested. Quality of tomato is judged by its ripeness. Arbitrating the level of ripeness is feasi- ble by analyzing the color of the tomato surface [9]. A classification chart by USDA (United States Department of Agriculture) discrim- inates six stages of ripening based on color namely green, breaker, http://dx.doi.org/10.1016/j.asoc.2015.07.009 1568-4946/© 2015 Elsevier B.V. All rights reserved.