Vis/NIR hyperspectral imaging for detection of hidden bruises on kiwifruits

Abstract

Lü Q., Tang M.-j., Cai j.-r., Zhao j.-w., Vittayapadung S. (2011): Vis/NIR hyperspectral imaging for detection of hidden bruises on kiwifruits. Czech j. Food Sci., 29: 595-602.It is necessary to develop a non-destructive technique for kiwifruit quality analysis because the machine injury could lower the quality of fruit and incur economic losses. Bruises are not visible externally owing to the special physical properties of kiwifruit peel.we proposed the hyperspectral imaging technique to inspect the hidden bruises on kiwifruit. The Vis/NIr (408-1117 nm) hyperspectral image data was collected. Multiple optimal wavelength (682, 723, 744, 810, and 852 nm) images were obtained using principal component analysis on the high dimension spectral image data (wavelength range from 600 nm to 900 nm). The bruise regions were extracted from the component images of the five waveband images using rBF-SVM classification. The experimental results showed that the error of hidden bruises detection on fruits by means of hyperspectral imaging was 12.5%. It was concluded that the multiple optimal waveband images could be used to constructs a multispectral detection system for hidden bruises on kiwifruits.

Unsupervised Spectral‐Spatial Feature Selection‐Based Camouflaged Object Detection Using VNIR Hyperspectral Camera

“…The baseline band selection method was principle components analysis (PCA), which is an effective data reduction technique that is used frequently in hyperspectral data analysis [6]. In PCA, a human manually selects a principle component (i.e., PC2 as shown in Figure 6(a)) that visualizes the abnormal region clearly.…”

“…Baseline method of band selection by PCA [6]: (a) principal component: PC2 and (b) load function curve of PC2.…”

Semi-supervised learning of hyperspectral image segmentation applied to vine tomatoes and table grapes

“…These methods can be divided into four groups, namely a) agglomerativehierarchicalclusteringinaone-step(1) and a two-step procedure (2), b) k-means clustering in a one-step (3) and a two-step (4) procedure, c) multivariateGaussianmixturemodelsinaone-step (5) and a two-step (6) procedure and d) a combination of two of the preceding methods, namelycombinationof(b)and(a) (7),and(b)and(c) (8).…”

“…Baranowski et al 7 useddifferentsupervisedclassificationmethods, suchassupportvectormachines(SVM),functionaltrees, nearest-neighbourclassifiersandregressionmethods,to distinguishbruisedappletissuefromsoundtissueand found that logistic regression gave the best classification resultswith a correct classification rate of 98.8%. Lü et al 8 applied SVM on hyperspectral data from kiwifruit to detect bruiseswith a misclassification rate of 12.5%. Partial least squares discriminant analysis (PLS-DA),asupervisedclassificationmethodbasedonpartial least squares regression (PLSR), 9 was used to classify undamaged, mechanically damaged and microbiological diseased mushroomswith a correct classification rate ofmorethan95% 10 and to classify hazelnuts into four qualityclasseswithmorethan90%accuracy.…”

Hyperspectral image applied to determine quality parameters in leafy vegetables

“…The classification results were about 93% of the non-bruised apples recognised, and an accuracy of about 86% for detecting bruises. Besides apple, hyperspectral imaging has also been used for the bruise detection of pear (Zhao et al, 2010), cucumber , mushroom (Esquerre et al, 2012;Gowen et al, 2008), strawberry (Nagata et al, 2006) and kiwifruit (Lü et al, 2011). Zhao et al (2010), using PCA and spectral angle mapper classification algorithm, obtained a detection accuracy of 95% for bruise detection on pears.…”