Automatic detection of skin defects in citrus fruits using a multivariate image analysis approach

Advance in Citrus Postharvest Management: Diseases, Cold Storage and Quality Evaluation

“…In addition, species and cultivars of citrus present a high rate of unpredictability in texture and colour, which makes it difficult to develop a general unsupervised method able to perform this task. In this context, a novelty detection technique was performed by using unsupervised method, based on a multivariate image analysis strategy in combination with PCA for the detection of new unpredictable defects in oranges and mandarins [76]. This unsupervised method needs only a few samples to be trained and could be suitable for the task of citrus inspection.…”

Automated Systems Based on Machine Vision for Inspecting Citrus Fruits from the Field to Postharvest—a Review

“…To solve this, Blasco et al (2007a) exploited the contrast between sound peel and defects by means of an unsupervised segmentation method based on region growing, which, on the other hand, consumed a large amount of processing time. Another technique was used by López-García et al (2010), who employed a multivariate image analysis approach consisting in introducing textural information about each pixel and its 3x3 neighbourhood and later using all the variables in a model based on principal component analysis (PCA) (Jackson, 1991). Several segmentation techniques applied on citrus fruit images were compared by Vijayarekha (2012a) to detect defects such as iterative intensity enhancement, contrast stretching, comparison against a reference colour of sound peel and Euclidean distance between the colour of different regions of interest.…”

“…Table 2 summarises the different works carried out for the application of computer vision in the citrus inspection in postharvest ordered by different topics chronologically. Table 2 Reference Achievement Estimation of properties of the fruit Blasco et al (2007a) Used an unsupervised segmentation method based on region growing to separate defects from sound skin Blasco et al (2007b) Tested different colour spaces to discriminate among eleven types of defects in the citrus peel and the stem Kim et al (2009) Introduced textural features in colour images to distinguish between some serious damages and other cosmetic defects (Blasco et al 2009) Introduced spectral and morphological information to distinguish between some serious damages and other cosmetic defects Omid et al (2010) Estimated the volume of the citrus using two cameras and computing the volume by dividing the fruit in a series of discs López-García et al (2010) Used multivariate image analysis introducing textural information and PCA to separate defects from sound skin López et al (2011) Used colour and texture features extracted in the RGB and HSI colour spaces to discriminate among seven common defects of citrus fruits Vijayarekha (2012a) Used several segmentation techniques to detect defects in citrus fruits Vijayarekha (2012b) Used several segmentation techniques to identify defects in citrus fruits Li et al (2013) Used RGB image ratios to discriminate the stem from different defects in oranges Cubero et al (2014b) Developed a robust method to detect stalks in different fruits, including oranges and mandarins Iqbal et al (2016) Investigated several supervised segmentation methods based on colour information Detection of decay lesions Gomez et al (2007) Used a Mahalanobis kernel to classify pixels as decay or sound skin in hyperspectral images Gómez-Sanchis et al (2008) Used correlation analysis, mutual information, stepwise, and genetic algorithms based on linear discriminant analysis (LDA) to select the most relevant bands of hyperspectral images, and classification and regression trees and LDA for pixels classification in decay or sound skin Kondo et al (2009) Studied the compounds involved in the fluorescence process to detect decay in oranges Kurita et al (2009) Innovative technique which alternatively switched on UV and white pulsed LED, thus allowing the inspection with both types of illumination, and hence allowed both a fluorescent and a colour image to be captured Slaughter et al (2008) Detect freeze-damages in the skin of trough fluorescence imaging Blanc et al (2010) Patented a commercial sorter for decay detection in citr...…”

Selection of Optimal Wavelength Features for Decay Detection in Citrus Fruit Using the ROC Curve and Neural Networks

“…Kim et al (2009) used colour texture features based on HSI and colour co-occurrence method to detect peel diseases in grapefruit. López-García et al (2010) used multivariate image analysis with the same objective in citrus fruits. However, some defects, like decay or freeze damage, are very difficult to detect using standard artificial vision systems since they are hardly visible to the human eye and, consequently, by standard red-green-blue (RGB) cameras Blasco et al (2007).…”