Application of Image Processing in Fruit and Vegetable Analysis: A Review

Abstract

Images are an important source of data and information in the agricultural sciences. The use of image-processing techniques has outstanding implications for the analysis of agricultural operations. Fruit and vegetable classification is one of the major applications that can be utilized in supermarkets to automatically detect the kinds of fruits or vegetables purchased by customers and to determine the appropriate price for the produce. Training on-site is the underlying prerequisite for this type of arrangement, which is generally caused by the users having little or no expert knowledge. We explored various methods used in addressing fruit and vegetable classification and in recognizing fruit disease problems. We surveyed imageprocessing approaches used for fruit disease detection, segmentation and classification. We also compared the performance of state-of-the-art methods under two scenarios, i.e., fruit and vegetable classification and fruit disease classification. The methods surveyed in this paper are able to distinguish among different kinds of fruits and their diseases that are very alike in color and texture.

Classificação de Espécies de Pimentas Empregando Análise de Imagens Digitais de Componentes Principais

“…Em [7], foram avaliados três métodos de classifi cação para discriminar Eucalyptus globulus e Eucalyptus nitens para infravermelho próximo (NIR) coletado das folhas. Dubey e Jalal [8], exploram vários métodos para a classifi cação de frutas e vegetal e o reconhecimento de doença em frutas. Bajwa [9] usaram a classificação baseada em PCA para classifi car os diferentes tipos de uni-camadas e multicamadas de nuvens.…”

Automatic Recognition of Ripening Tomatoes by Combining Multi-Feature Fusion with a Bi-Layer Classification Strategy for Harvesting Robots

“…The first proposed method utilized basic graphical information. Colour composition is the basis of an image, equally important for identification of all vegetables and fruits; therefore, colour-based models have been most commonly studied (e.g., [20,21,22,23,24,25,26,27,28]). In particular, Xu et al [22] used a colour-based model to analyse the fruit’s colour information, and red-blue (R-B) chromatic aberration information of images has been used for recognizing oranges hanging from a tree.…”

“…The Hough transform method, based on the idea of shape model, was used to simplify the process of target identification [32]. Texture can also be useful for identification, owing to its importance in visual perception; consequently, many scholars explored that venue (e.g., [27,33,34]). Recently, both the accuracy of fruit recognition and the performance of a detection system were significantly improved using machine learning methods [35].…”

“…Recently, both the accuracy of fruit recognition and the performance of a detection system were significantly improved using machine learning methods [35]. Application of machine learning methods to the recognition of fruits and vegetables has become very popular (e.g., [25,27,36,37,38,39]). In their review, Gongal et al [40] summarized machine learning methods for fruit recognition, including unsupervised classification, supervised classification, and soft computing methods, and discussed the advantages and drawbacks of different classifiers.…”

Apple disease classification using color, texture and shape features from images

“…Application of image processing in fruit and vegetable analysis is surveyed by Dubey and Jalal [3]. They also proposed a framework for fruits and vegetables recognition and classification [4,5].…”