Search

In this study a hyperspectral imaging system (short wave infrared range from 1000 to 2500 nm) was used to model fish texture by experimental compression test. Partial least square–discriminate analysis modeling technique was used for classifying the samples by linking the hyperspectral information and their measured texture. TheR2of cross validation and prediction were 0.97 and 0.96, respectively. The root mean squared errors for cross validation and prediction were 0.07 and 0.09, respectively. Sensitivity and specificity for both class I and II were 1.00. Results indicated that hyperspectral imaging in short wave infrared range has ability to detect texture stiffness of rainbow trouts which is affected by freshness. [ABSTRACT FROM PUBLISHER]

A rapid, robust, unbiased, and inexpensive discriminant method capable of classifying hazelnut ( Corylus avellana, L.) in compliance with the statements set out by the Commission Regulation (EC) No. 1284/2002 is economically important to the fresh and processed industries. Thus, in this study, the feasibility of high dynamic range (HDR) hyperspectral imaging for hazelnut kernel sorting (cv. Tonda Gentile Romana) of four quality classes (Class Extra, Class I, Class II, and Waste) has been investigated. Two different exposure times (5 and 8 ms) were selected for experiments, and the respective spectra were combined to obtain a HDR over the full spectral range. The illumination setup was optimized to improve the intensity and uniformity of the light along the field of view of the camera. PLS-DA was used to classify the kernels based on their spectra and the spectral pretreatment was optimized through an iterative routine. The performance of each PLS-DA model was defined based on its accuracy, sensitivity, and selectivity rates. All of the selected models provided a very good (>90 %) or good (>80 %) sensitivity and selectivity for the predefined classes. Misclassified kernels were primarily assigned to the low-quality classes (i.e., Class II and Waste). Moreover, the spatial domain was used to evaluate the feasibility of distinguishing hazelnut classes on the basis of their size and shape. It was found that hazelnut dimensions can be used to improve the accuracy of the classification of the kernels. Thanks to this combination of both spectral and spatial information spectral imaging could be used for quality sorting of hazelnuts. [ABSTRACT FROM AUTHOR]

Thinning of fruit-tree blossoms is used to regulate the yearly tree bearing and to increase the fruit yield and quality. While this is still mostly done by hand, the increasing costs of manual labor have created a demand for mechanization. This has recently led to the development of several prototype thinning machines. The main disadvantage of these machines is that they are not selective, while the fruit bearing capacity of different floral buds is not equal. On-line information about the position and distribution of the floral buds on the tree can improve the efficiency of mechanized thinning. Therefore, the aim of this study was to identify the most informative wavebands to develop a multispectral vision sensor for detection of the floral buds of the pear cultivar Conference. Hyperspectral scans were taken from tree samples in five early phenological stages to create a database of reflectance spectra for the different tree features. A stepwise algorithm was then applied to this training set to select the best combination of wavebands having the highest discriminating power between the components of interest. Subsequently, canonical correlation analysis was used to create discriminant functions out of the selected wavebands. It was possible to correctly classify 95 % of the (pixel) observations with six selected wavebands. The discrimination performance was also tested as a function of the number of used wavebands. Analysis showed that when only the two most important wavebands were used, still over 90 % of the (pixel) observations could be correctly classified. [ABSTRACT FROM AUTHOR]

Banana and its dried products are one of the most common foods consumed all over the world. Quality attributes of dried banana slices are typically evaluated using time-consuming and labor intensive methods which do not allow to evaluate each individual slice. Replacement of the current destructive methods for quality evaluation of dried banana slices by fast, non-destructive methods would provide large added value to the food industry. Therefore, the aim of this study was to monitor the moisture content, texture and color of banana slices during the drying process in a fast and non-destructive way. VNIR hyperspectral reflectance imaging in the 400–1000 nm range was selected for this purpose. Thanks to a cross-polarized configuration the effects of glare or specular reflection on the banana slice surfaces in the hyperspectral diffuse reflectance images was largely reduced. Reference quality attributes for all the slices (moisture content, texture and color ( L*, a*, b* values) obtained using conventional destructive methods and colorimeter were predicted from their corresponding average reflectance spectra extracted from the hyperspectral images by means of partial least squares regression (PLSR). The PLSR calibration models were validated on samples which had not been used for model calibration. The results were very good for water content ( R 2 P  = 0.97, RMSEP  = 0.05 kg water/kg DM), quite good for and b* value ( R 2 P  = 0.83, RMSEP  = 1.95), and reasonable for texture ( R 2 P  = 0.66, RMSEP  = 11.8 N), a* value ( R 2 P  = 0.53, RMSEP  = 1.32) and L* value ( R 2 P  = 0.61, RMSEP  = 5.92). Subsequently, these calibration models were used to predict those quality attributes at pixel level for the validation slices to visualize the spatial distribution of these quality parameters at different stages during the drying process. The obtained results clearly indicate the potential of cross-polarized hyperspectral reflectance imaging for non-destructive monitoring of the quality attributes of banana slices during drying. [ABSTRACT FROM AUTHOR]

High speed data processing for online food quality inspection using hyperspectral imaging (HSI) is challenging as over hundred spectral images have to be analyzed simultaneously. In this study, a real-time pixel based early apple bruise detection system based on HSI in the shortwave infrared (SWIR) range has been developed. This systems consists of a novel, homogeneous SWIR illumination unit and a line scan camera. The system performance was tested on Jonagold apples bruised less than two hours before scanning. Partial least squares-discriminant analysis was used to discriminate bruised pixel spectra from sound pixel spectra. As the glossiness of many fruit and vegetables limits the accuracy in the detection of defects, several reflectance calibrations and pre-processing techniques were compared for glare correction and maximizing the signal to noise ratio. With the best combination of first derivative and mean centering, followed by image post-processing, this system was able to detect fresh bruises in thirty apples with 98% accuracy at the pixel level with a processing time per apple below 200 ms. [ABSTRACT FROM AUTHOR]

Spectroscopy in the visible near-infrared spectral (Vis-NIRS) range combined with imaging techniques (hyperspectral imaging, HSI) allows assessment of chemical composition, texture, and meat structure. The use of HSI in the meat and food industry has observed a significant growth in the last decade, yet its use for assessment of meat it is not optimal yet. The application of HSI for assessment of meat is reviewed with focus on its ability to capture meat unique chemical and structural characteristics. While HSI is widely used for assessment of chemical composition, a limited number of evidences on its ability to handle the effect of different sources of variation on the assessment is found. The use of spatially resolved spectroscopy has been able to detect structural information related to animal background, muscle type, rigor process and ageing. Similarly the use of texture features seem to capture unique characteristics of meat. [ABSTRACT FROM AUTHOR]

Bruises in apples is one of the most important quality factors during postharvest, which needs to be detected early and efficiently during sorting processes. In this study, a step-wise pixel based apple bruise detection system based on line scan hyperspectral imaging (HSI) in the shortwave infrared (SWIR) is demonstrated for three apple cultivars: ‘Jonagold’, ‘Kanzi’ and ‘Joly Red’. The SWIR HSI system performance was tested on apples from the different cultivars bruised at five different impact levels, and monitored from 1 to 36 h after bruising. While glare regions are commonly considered as anomalies and discarded from further analysis, their spectral signatures enabled in this work to distinguish between cultivars with a prediction accuracy up to 96%. Different partial least squares-discriminant analysis (PLS-DA) models were trained to discriminate cultivars and then to discriminate between sound, bruised, glossy and stem regions. Spectral area normalization pre-processing was found to be the most effective for pixel based bruise prediction, resulting in a prediction accuracy up to 90.1%. Post-processing of the binary images by exploiting spatial information further improved the bruise detection accuracy to 94.4%. [ABSTRACT FROM AUTHOR]

Blackspot is a subsurface potato damage resulting from impacts during harvesting. This type of bruising represents substantial economic losses every year. As the tubers do not show external symptoms, bruise detection in potatoes is not straightforward. Therefore, a nondestructive and accurate method capable of identifying bruised tubers is needed. Hyperspectral imaging (HSI) has been shown to be able to detect other subsurface defects such as bruises in apples. This method is nondestructive, fast and can be fully automated. Therefore, its potential for non-destructive detection of blackspot in potatoes has been investigated in this study. Two HSI setups were used, one ranging from 400 to 1000 nm, named Visible-Near Infrared (Vis-NIR) and another covering the 1000–2500 nm range, called Short Wave Infrared (SWIR). 188 samples belonging to 3 different varieties were divided in two groups. Bruises were manually induced and samples were analyzed 1, 5, 9 and 24 h after bruising. PCA, SIMCA and PLS-DA were used to build classifiers. The PLS-DA model performed better than SIMCA, achieving an overall correct classification rate above 94% for both hyperspectral setups. Furthermore, more accurate results were obtained with the SWIR setup at the tuber level (98.56 vs. 95.46% CC), allowing the identification of early bruises within 5 h after bruising. Moreover, the pixel based PLS- DA model achieved better results in the SWIR setup in terms of correctly classified samples (93.71 vs. 90.82% CC) suggesting that it is possible to detect blackspot areas in each potato tuber with high accuracy. [ABSTRACT FROM AUTHOR]

Hyperspectral reflectance imaging using unpolarised light has high potential for various applications in non-destructive quality evaluation of food and agricultural products. However, one of the most important challenges is the existence of glare in the obtained hyperspectral reflectance images due to specular reflection of the illuminating light on the sample surface. Whilst this can be avoided for flat surfaces by optimising the viewing angle, this is not possible for curved food products. Therefore, cross-polarisation has been implemented in this study to block the specular reflection and exclude it from the acquired hyperspectral reflectance images in the 450–1000 nm range. The added value of this approach has been successfully demonstrated for three glossy agricultural products: aubergine and two apple cultivars, for which the quality of the obtained hyperspectral images, expressed by signal-to-noise ratios, could be improved from 1.1 to 3 times depending on the wavelength regions. [ABSTRACT FROM AUTHOR]

Thinning of pome and stone fruit involves the reduction of tree crop load in order to regulate fruit set and quality. As it is typically carried out through manual labor, thinning comprises a large part of a grower’s production costs. Mechanized thinning has been shown to be a cost-effective alternative but the performance of existing thinning devices needs to be further improved by taking the variation in bearing capacity of the individual trees into account. In this work, a multispectral camera system is developed to detect the floral buds of pear (cv. Conference ) during the growth stages prior to bloom. During a two-year field trial, the multispectral system was used to measure orchard scenes in six distinct optical wavebands under controlled illumination. These wavebands are situated in the visible and near infrared region of the spectrum and were selected based on hyperspectral laboratory measurements described in previous work. The recorded multispectral images were converted to a database containing the spatial–spectral signatures of the objects present in the orchard. Subsequently, canonical correlation analysis was applied to create a spectral discriminant model that detects pixels originating from floral buds. This model was then applied to the recorded data after which an image analysis algorithm was designed and optimized to predict the number of floral buds. In total, approximately 87% of the visible floral buds were detected correctly with a low false discovery rate (<16%). Therefore, it is expected that the multispectral sensor can be used to improve the efficiency of existing thinning devices. Additionally, it could as well be used as a stand-alone sensor for early-season yield estimation. [ABSTRACT FROM AUTHOR]

A hyperspectral scatter imaging system was developed for the contactless acquisition of spatially resolved diffuse reflectance profiles for the optical characterization of turbid food products in the wavelength range from 500 to 960 nm. To investigate the potential of this concept sugar foams with different microstructures, but with similar chemical compositions, have been prepared by applying different mixing times to the same mixtures of sugar and albumin. Hyperspectral scatter images have been acquired from these samples and the absorption and reduced scattering coefficients have been derived from spatially resolved reflectance profiles based on the diffusion approximation of the radiative transfer equation describing the light propagation in turbid media. The estimated reduced scattering coefficients μ=sub=s=/sub=′ spectra clearly reflected the effect of the different mixing times on the foam microstructure. On the other hand, similar absorption coefficient spectra were observed, confirming the identical chemical composition of the sugar-albumin matrix. These results indicate that the hyperspectral scatter imaging technique has potential as a non-contact and rapid method for online quality control and process monitoring of foamed food products. [ABSTRACT FROM AUTHOR]

Abstract: The freshness of rainbow trout is one of the most important quality parameters to attract customers. Common methods to detect fish freshness are usually subjective to the skill of a quality evaluator and are time consuming and destructive. Therefore, an automatic, nondestructive, accurate and quick method is needed. Hyperspectral imaging has demonstrated its efficiency in the meat and fish industries for quality control purposes. This method is nondestructive, fast and automatic. In this study, two setups for hyperspectral imaging named “Visible–Near Infrared” (Vis–NIR) and “Short Wave Infrared” (SWIR) are used to determine fish freshness. Eighty fresh rainbow trouts were divided into four batches which were separately preserved in ice for 1, 3, 5 and 7days, respectively. Principle Component Analysis (PCA) and Partial Least Squares-Discriminate Analysis (PLS-DA) were used as unsupervised and supervised techniques for the evaluation of rainbow trout freshness. Results obtained by PCA technique indicated that four classes of samples can be detected using the Vis–NIR mean spectrum by applying a second derivative (D2) preprocessing method. The RCV 2 and RPre with D2 preprocessing were 0.97 and 0.98 for Vis–NIR and 0.84 and 0.67 for SWIR, respectively. The corresponding values of RMSECV and RMSEPre were 0.16 and 0.14 in Vis–NIR and 0.44 and 0.76 in SWIR, respectively. Classification model achieved an overall correct classification of 100% and 75% for Vis–NIR and SWIR, respectively. The obtained results using both PCA and PLS-DA methods indicated that the Vis–NIR imaging system performs better than SWIR. Among all applied preprocessing techniques, the second derivative preprocessing achieved the best performance. [Copyright &y& Elsevier]