Hyperspectral Transmittance Imaging Research Articles

Hyperspectral transmittance imaging detection of early decayed oranges caused by Penicillium digitatum using NFINDR-JMSAM algorithm with spectral feature separating

Decay caused by Penicillium spp. is the main cause of postharvest citrus quality loss, moreover, the fungus can quickly infect entire batches of citrus fruit resulting in significant economic losses. However, effective detection of early decay remains a challenge due to the lack of distinct visual features. In this study, a Vis-NIR hyperspectral imaging system was developed to acquire full-transmittance images and an NFINDR-JMSAM algorithm was proposed to segment different image pixels. By extracting pure pixels and separating spectral features, the overall classification accuracy of 99.3 % was obtained for all tested samples. The proposed method can also effectively identify scars on the flavedo, citrus stem-end and navel, thereby eliminating their interference with the detection of decayed orange fruit. This study provided a new idea for accurately detecting the early decayed citrus fruit and visualizing the detection results for different tissues by combining hyperspectral transmittance imaging and NFINDR-JMSAM algorithm.

Fertility detection of unincubated chicken eggs by hyperspectral transmission imaging in the Vis-SWNIR region

Detection of infertile eggs prior to incubation can lead to an increase in the hatchability rate and prevent the wastage of billions of non-fertile eggs ended up by failed incubation. In this study, the feasibility of a line-scan hyperspectral imaging system in the visible and short-wavelength near-infrared region was assessed for early detection of non-fertile eggs on day 0 before incubation. A total of 227 white-shell eggs including 131 fertile and 96 infertile eggs were collected from a flock with similar conditions in terms of hen age, feeding, and management. Hyperspectral images of eggs were captured on day 0 before incubation in a transmittance mode of illumination and then the eggs were incubated in a commercial incubator. The edge detection method was used to segment the egg, including both the white and yolk, from the background, and the image spectral information was extracted from the egg region. After applying various pretreatment methods, different classifiers including soft independent modeling of class analogy (SIMCA), linear discriminant analysis (LDA), quadratic discriminant analysis (QDA), and artificial neural networks (ANN) classifiers were utilized to extract the predictive models. Following the acceptable results of SIMCA analysis accomplished by 1st derivative pretreatment (accuracy of 86.67%), the discrimination power plot was used to select the most informative wavebands. The results showed that by using fewer variables in effective wavebands better performance (precision and accuracy of 92.59% and 93.33%, respectively) could be obtained in comparison with the ANN classifier based on the whole spectral data (precision and accuracy of 89.29% and 91.11%, respectively). This study revealed the potential application of hyperspectral transmittance imaging in the Vis-SWNIR region to discern the fertile and infertile eggs before starting the incubation process.

Determination of plumpness for kernel of semen ziziphi spinosae use of hyperspectral transmittance imaging technology coupled with improved Otsu algorithm

Semen ziziphi spinosae are the seeds of sour dates. They can be used as both food and medicine, with large and full kernels producing particularly good medicinal herbs. However, the kernels of semen ziziphi spinosae are wrapped in a shell, so it is difficult to analyze the fullness of the kernel without damaging the seeds themselves. In this study, shortwave near-infrared (SW-NIR) and longwave near-infrared (LW-NIR) hyperspectral imaging techniques are used to determine the recognition ability of kernels from shell regions in different wavelength ranges using a two-band ratio algorithm. LW-NIR is found to be more suitable for the non-destructive detection of the inner kernel of semen ziziphi spinosae. Three practical wavelength selection algorithms and two classification algorithms are used to build pixel-level kernel detection models that achieve classification accuracies of greater than 97%. A novel improved Otsu algorithm is developed and applied to multispectral images. The watershed algorithm, threshold segmentation, and improved Otsu algorithm are used to build image-level kernel detection models. The detection results indicate that the improved Otsu algorithm is superior to other algorithms, achieving accuracy levels of 100%, 97.5%, 95%, and 100% for four different types of plumpness, respectively. These results indicate that the proposed method has the potential for the online detection of suitable semen ziziphi spinosae kernels. Moreover, LW-NIR hyperspectral transmission imaging combined with the improved Otsu algorithm is shown to provide a useful non-destructive method for detecting kernel fullness for medicinal materials.

Fully convolutional networks for blueberry bruising and calyx segmentation using hyperspectral transmittance imaging

Early detection of internal bruises in blueberries is a significant challenge for the blueberry industry. The main goal of this study was to detect blueberries’ internal bruising accurately, after mechanical damage from hyperspectral transmittance images (HSTIs), using the deep learning-based method of fully convolutional networks (FCNs) for segmentation tasks. To improve detection accuracy, a total of three classes (bruised tissue, unbruised tissue, and calyx end of blueberries), were treated as segmentation targets. A near-infrared hyperspectral imaging system was used to acquire transmittance images of 1200 blueberries, and the images were divided randomly to form training, validation and testing sets. Three categories of input HSTIs were used to evaluate the FCN models using pre-trained weights (transfer learning) and random initialisation. Random forests and linear discriminant analysis were applied to generate 9-channel and 3-channel input images along with full-wavelength multi-channel images. The results indicate that when using the deep learning approach, blueberry bruises and calyx ends can be segmented from the blueberry fruit as early as 30 min after mechanical damage has been inflicted on the blueberries. The new full-wavelength model with random initialisation had the highest accuracy (81.2% over the entire test set), and can be used to research the resistance of blueberry fruit to mechanical damage. The new 3-channel and 9-channel models show potential for application to packing-line detection and online inspection.

Detection of blood spots in eggs by hyperspectral transmittance imaging

Blood spots are one of undesired inclusions in eggs, whose detection success is highly dependent on shell color. This research reports a method for detecting blood spots in light brown-shelled eggs on the basis of hyperspectral transmittance images. The normalized spectra of intact eggs and their shells were acquired. Five feature wavelengths of intact eggs selected by the successive projections algorithm and 3 absorption peak locations of eggshells were regarded as characteristic bands. The k-nearest neighbor (kNN) and support vector machine (SVM) algorithms were adopted to develop detection models. The latter achieved better performance. The overall classification accuracy increased to 100% by merging normalized spectra of intact eggs at 5 feature wavelengths with 3 absorption peaks of eggshells as input variables of SVM-based model. Moreover, a practical SVM-based model with 96.43% overall classification accuracy was established by replacing inputs with normalized spectra of intact eggs at characteristic bands. Keywords: hyperspectral transmittance imaging, non-destructive detection, blood-spot, egg DOI: 10.25165/j.ijabe.20191206.5376 Citation: Feng Z, Ding C Q, Li W H, Cui D. Detection of blood spots in eggs by hyperspectral transmittance imaging. Int J Agric & Biol Eng, 2019; 12(6): 209–214.

Detection of early decay on citrus using hyperspectral transmittance imaging technology coupled with principal component analysis and improved watershed segmentation algorithms

Abstract Decay caused by Penicillium spp. fungi is one of the main problems affecting marketing of citrus fruit after harvest because the fungal infection can spread fast from a small number of decayed fruit to the whole consignment. However, the automatic detection of decayed citrus is still a challenge. Early decay of citrus happen on surface peel and present a obvious symptom of water-soaked with cell tissue collapse, which may offer the feasibility of transmittance imaging mode to detect decayed region of citrus. In this study, image processing methods including principal component analysis (PCA), pseudo-color image transformation technology and improved watershed segmentation algorithms (IWSA) were employed to analyze the feasibility of decay detection based on the scanned hyperspectral transmittance images (325−1098 nm) of sound and decayed oranges. The results show that PC3 image is promising for decay segmentation. G components extracted from pseudo-color images of PC3 were selected to enhance image contrasts between decayed and sound tissues, and then decayed regions were segmented perfectly by IWSA whether the defects located on the edge or center position of oranges. However, stem-end tissue had similar features with decayed tissue and therefore were easily misidentified as decayed tissues for those decayed samples with stem-end tissue, and so stem-end identification was carried out. PC2 image and R components extracted from pseudo-color images of PC2 were promising for stem-end identification, then IWSA and morphological parameters were used to extract stem-end region. The stem-end was marked in both operations of decay segmentation and stem-end identification, hence decayed region were further determined for eliminating the misclassification interference of stem-end tissue on decay detection by removing the stem-end region from the operation of decay segmentation. For a validation set including of 84 decayed and 66 sound fruit, the success rates were 93 % and 96 %, respectively, and 94 % for decayed, sound and all fruit. Hyperspectral transmittance imaging offers a novel method for automatic detection of early decayed orange caused by fungus.

Measurement of SSC in processing tomatoes (Lycopersicon esculentum Mill.) by applying Vis‐NIR hyperspectral transmittance imaging and multi‐parameter compensation models

AbstractProcessing tomato (Lycopersicon esculentum Mill.) is a very important horticultural product all over the world. Soluble solid content (SSC) as a key quality assessment parameter of processing tomato directly affects quality and cost of processing tomato products such as tomato paste. This study proposes an analytical method for quantitatively assessment of SSC in processing tomatoes by using the visible and near‐infrared (Vis‐NIR) hyperspectral transmittance imaging technique and multi‐parameter compensation models. Monte‐Carlo outlier detection method was applied to optimize sample set. The competitive adaptive reweighted sampling (CARS) algorithm was used to identify the optimal wavelengths from transmittance spectra. Two characteristic parameters including area (size) and weight of samples were measured and different models including conventional PLS/LS‐SVM models and multi‐parameter compensation PLS/LS‐SVM models were established, respectively. The results obtained by comparing all the established models showed that the multi‐parameter (spectrum, area, and weight) compensation CARS‐LS‐SVM model with 47 important wavelengths had the best assessment ability of SSC in processing tomatoes. The prediction accuracies of the optimal model are rcal = 0.95 and RMSEC = 0.16 for calibration set, and rpre = 0.94, RMSEP = 0.17 and RPD = 2.94 for prediction set, respectively. Research results indicated that the newly proposed Vis‐NIR hyperspectral transmittance imaging combining with multi‐parameter compensation LS‐SVM model would be potential as a noninvasive technique to quantitatively evaluate the SSC in processing tomatoes.Practical applicationsFast assessment of soluble solid content (SSC) in processing tomatoes has significant value for tomato processing in practical application. High quality processing tomatoes can reduce the processing cost and raise product prices. The newly proposed SSC measurement technique integrated both Vis‐NIR hyperspectral transmittance imaging and multi‐parameter compensation models. This technique can effectively overcome some disadvantageous influences from size and weight of processing tomatoes in the assessment process of SSC. Use of transmittance technology further improved the acquisition capability of effective spectral information from the tested object. This work was valuable for setting up the fast multispectral inspection system that can be used for SSC measurement of processing tomatoes in practical on‐line grading applications.

Use of hyperspectral transmittance imaging to evaluate the internal quality of nectarines

The internal quality of nectarines (Prunus persica L. Batsch var. nucipersica) cv. ‘Big Top’ (yellow flesh) and ‘Magique’ (white flesh) has been inspected using hyperspectral transmittance imaging. Hyperspectral images of intact fruits were acquired in the spectral range of 630–900 nm using transmittance mode during their ripening under controlled conditions. The detection of split pit disorder and classification according to an established firmness threshold were performed using PLS-DA. The prediction of the Internal Quality Index (IQI) related to ripeness was performed using PLS-R. The most important variables were selected using interval-PLS. As a result, an accuracy of 94.7% was obtained in the detection of fruits with split pit of the ‘Big Top’ cultivar. Accuracies of 95.7% and 94.6% were achieved in the classification of the ‘Big Top’ and ‘Magique’ cultivars, respectively, according to the firmness threshold. The internal quality was predicted through the IQI with R2 values of 0.88 and 0.86 for the two cultivars. The results obtained indicate the great potential of hyperspectral transmittance imaging for the assessment of the internal quality of intact nectarines.

Detection of blood spots in eggs by hyperspectral transmittance imaging

Blood spots are one of undesired inclusions in eggs, whose detection success is highly dependent on shell color. This research reports a method for detecting blood spots in light brown-shelled eggs on the basis of hyperspectral transmittance images. The normalized spectra of intact eggs and their shells were acquired. Five feature wavelengths of intact eggs selected by the successive projections algorithm and 3 absorption peak locations of eggshells were regarded as characteristic bands. The k-nearest neighbor (kNN) and support vector machine (SVM) algorithms were adopted to develop detection models. The latter achieved better performance. The overall classification accuracy increased to 100% by merging normalized spectra of intact eggs at 5 feature wavelengths with 3 absorption peaks of eggshells as input variables of SVM-based model. Moreover, a practical SVM-based model with 96.43% overall classification accuracy was established by replacing inputs with normalized spectra of intact eggs at characteristic bands. Keywords: hyperspectral transmittance imaging, non-destructive detection, blood-spot, egg DOI: 10.25165/j.ijabe.20191206.5376 Citation: Feng Z, Ding C Q, Li W H, Cui D. Detection of blood spots in eggs by hyperspectral transmittance imaging. Int J Agric & Biol Eng, 2019; 12(6): 209–214.

Detection of blood spots in eggs by hyperspectral transmittance imaging

Detection of blood spots in eggs by hyperspectral transmittance imaging

Active learning algorithm can establish classifier of blueberry damage with very small training dataset using hyperspectral transmittance data

The aim of this study was to estimate the performance of active learning algorithm for detecting blueberry damage using hyperspectral transmittance data with the very low labeling cost. A hyperspectral transmittance imaging system was first applied to collect the hyperspectral transmittance data of blueberries. Subsequently, the mean hyperspectral transmittance data was extracted. With only 9 labeled berries, the estimated error reduction could achieve the accuracy, precision and recall of 0.87, 0.93 and 0.78 respectively, and it consistently improved or maintained the performance of classifier for the remainder of the queries. In contrast to the SOM and SVM models, the classifier based on estimated error reduction also provided higher accuracy, precision and recall with the much fewer labeled samples. The active learning algorithms can be extended to the large scale applications in which the labeled samples are very limited or expensive and the models are required to be frequently transferred. In our case, due to the significant biological variations existing among blueberry samples, the classifier required frequent updates in practical applications, and the active learning algorithms could remarkably reduce label effort during the model updating processes.

Algorithms for detecting cherry pits on the basis of transmittance mode hyperspectral data

Abstract The suitability of the hyperspectral transmittance imaging technique was assessed in terms of detecting the internal intrusions (pits and their fragments) in cherries. Herein, hyperspectral transmission images were acquired in the visible and near-infrared range (450-1000 nm) from pitted and intact cherries of three popular cultivars: ‘Łutówka’, ‘Pandy 103’, and ‘Groniasta’, differing by soluble solid content. The hyperspectral transmittance data of fresh cherries were used to determine the influence of differing soluble solid content in fruit tissues on pit detection effectiveness. Models for predicting the soluble solid content of cherries were also developed. The principal component analysis and the second derivative pre-treatment of the hyperspectral data were used to construct the supervised classification models. In this study, five classifiers were tested for pit detection. From all the classifiers studied, the best prediction accuracies for the whole pit or pit fragment detection were obtained via the backpropagation neural networks model (87.6% of correctly classified instances for the training/test set and 81.4% for the validation set). The accuracy of distinguishing between drilled and intact cherries was close to 96%. These results showed that the hyperspectral transmittance imaging technique is feasible and useful for the non-destructive detection of pits in cherries.

Classification of foreign matter embedded inside cotton lint using short wave infrared (SWIR) hyperspectral transmittance imaging

Cotton is an important source of natural fiber around the world. Cotton lint, however, could be contaminated by various types of foreign matter (FM) during harvesting and processing, leading to reduced quality and potentially even defective textile products. Current sensing methods can detect the presence of foreign matter on the surface of cotton lint, but they are not able to efficiently detect and classify foreign matter that is mixed with or embedded inside cotton lint. This study focused on the detection and classification of common types of foreign matter hidden within the cotton lint by a short wave near infrared hyperspectral imaging (HSI) system using the transmittance mode. Fourteen common categories of foreign matter and cotton lint were collected from the field and the foreign matter particles were sandwiched between two thin cotton lint webs. Operation parameters were optimized through a series of experiments for the best performance of the transmittance mode. After acquiring transmittance images of the cotton lint and foreign matter mixture, minimum noise fraction (MNF) rotation was utilized to obtain component images to assist visual detection and mean spectraextractionfrom a total of 141 wavelength bands. The optimal spectral bands were identified by using the minimal-redundancy-maximal-relevance (mRMR)-based feature selection method. Linear discriminant analysis (LDA) and a support vector machine (SVM) were employed to classify foreign matter at the spectral and pixel level, respectively. Over 95% classification accuracies for the spectra and the images were achieved using the selected optimal wavelengths. This study indicated that it was feasible to detect botanical (e.g. seed coat, seed meat, stem, and leaf) and non-botanical (e.g. paper, and plastic package) types of foreign matter that were embedded inside cotton lint using short wave infrared hyperspectral transmittance imaging.

Detection of Internally Bruised Blueberries Using Hyperspectral Transmittance Imaging

Abstract. Internal bruise damage that occurs in blueberry fruit during harvest operations and postharvest handling lowers the overall quality and causes significant economic losses. The main goal of this study was to nondestructively detect internal bruises in blueberries after mechanical damage using hyperspectral transmittance imaging. A total of 600 hand-harvested blueberries were divided into 20 groups of four storage times (30 min, 3 h, 12 h, and 24 h), two storage temperatures (22°C and 4°C), and three treatments (stem bruise, equator bruise, and control). A near-infrared hyperspectral imaging system was used to acquire transmittance images from 970 to 1400 nm with 5 nm bandwidth. Images were acquired from three orientations (calyx-up, stem-up, and equator-up) for fruit in the control and stem bruise groups and from four orientations (calyx-up, stem-up, equator-up, and equator-down) in the equator bruise groups. Immediately after imaging, the fruit samples were sliced, and the sliced surfaces were photographed. The color images of sliced fruit were used as references. By comparing with the reference color images, the profiles of spatial and spectral intensities were evaluated to observe the effect of orientation and help extract regions of interest (ROIs) of bruised and healthy tissues. A support vector machine (SVM) classifier was trained and tested to classify pixels of bruised and healthy tissues. Classification maps were produced, and the bruise ratio was calculated to identify bruised blueberries (bruise ratio >25%). The average accuracy of blueberry identification was 94.5% with the stem-up orientation. The results indicate that detecting bruised blueberries as soon as 30 min after mechanical damage is feasible using hyperspectral transmittance imaging. Keywords: Blueberry, Bruise detection, Classification, Hyperspectral imagery, Transmittance mode.

Black Heart Detection in White Radish by Hyperspectral Transmittance Imaging Combined with Chemometric Analysis and a Successive Projections Algorithm

Radishes with black hearts will lose edible value and cause food safety problems, so it is important to detect and remove the defective ones before processing and consumption. A hyperspectral transmittance imaging system with 420 wavelengths was developed to capture images from white radishes. A successive-projections algorithm (SPA) was applied with 10 wavelengths selected to distinguish defective radishes with black hearts from normal samples. Pearson linear correlation coefficients were calculated to further refine the set of wavelengths with 4 wavelengths determined. Four chemometric classifiers were developed for classification of normal and defective radishes, using 420, 10 and 4 wavelengths as input variables. The overall classifying accuracy based on the four classifiers were 95.6%–100%. The highest classification with 100% was obtained with a back propagation artificial neural network (BPANN) for both calibration and prediction using 420 and 10 wavelengths. Overall accuracies of 98.4% and 97.8% were obtained for calibration and prediction, respectively, with Fisher's linear discriminant analysis (FLDA) based on 4 wavelengths, and was better than the other three classifiers. This indicated that the developed hyperspectral transmittance imaging was suitable for black heart detection in white radishes with the optimal wavelengths, which has potential for fast on-line discrimination before food processing or reaching storage shelves.

Assessment of internal quality of blueberries using hyperspectral transmittance and reflectance images with whole spectra or selected wavelengths

Hyperspectral imaging has been used in previous studies for assessing firmness and soluble solids content of fresh fruit. To assess the applicability of this technique for automatic sorting and grading of blueberries, we investigated different sensing modes (i.e., reflectance and transmittance), evaluated the effect of fruit orientation on fruit quality prediction, and developed robust prediction models with fewer wavelengths. In this study, a hyperspectral imaging system was used to acquire reflectance and transmittance images from 420 blueberries in three fruit orientations (i.e., stem end, calyx end and equator) for the spectral region of 400–1000 nm. Mean spectra were extracted from the hyperspectral images of each blueberry. Calibration models for soluble solids content (SSC) and firmness index (FI) were developed using partial least squares regression for the reflectance and transmittance spectra as well as their combination. Further, interval partial least squares (iPLS) regression with 10 different intervals of nine wavelengths was used to reduce the spectral dimensionality. Overall, reflectance gave better results (the best correlation for prediction (Rp) of 0.90 for SSC and 0.78 for FI) than transmittance (Rp of 0.76 for SSC and 0.64 for FI). For reflectance, SSC and FI predictions for the stem end orientation were better than for the other two orientations, while fruit orientation had little or insignificant effect on transmittance predictions. Combination of reflectance and transmittance spectra did not yield improved prediction results for both SSC and FI. On average, the prediction errors for iPLS increased by only 5%, compared to PLS for the whole spectra. The research demonstrated that it is feasible to use hyperspectral imaging technique for prediction of internal quality of blueberries with a few selected wavelengths with results similar to that with whole spectral information. Because of the distance traveled from the South to the North hemisphere, it is especially important to perform internal and external quality determination for individual fresh blueberries to ensure their quality upon arrival at the destination. Soluble solids content and firmness are important fruit quality parameters. Hyperspectral imaging has emerged as a new technique for quality and safety inspection of food and agricultural products and could be useful for blueberry quality assessment. However, there are several limitations to be afforded before: technique implementation velocity since this method uses multiple images from contiguous wavelengths (increasing computational costs), fruit light interaction, and fruit orientation effect between others. Specifically, the submitted manuscript presents results in order to demonstrate the hyperspectral imaging technique feasibility with a few selected wavelengths to achieve acceptable results for the prediction of internal quality of blueberries, thus, this would make it possible to implement the technique in the near future for online commercial sorting and grading of blueberries.

Hyperspectral transmittance imaging of the shell-free cooked clam Mulinia edulis for parasite detection

The clam Mulinia edulis is occasionally infected by the parasite Edotea magellanica. In this paper the normalized optical transmittance of the clam and its parasite have been disclosed for the first time using a hyperspectral imaging system. It has been identified that, in the spectral band of 600–950nm, the normalized optical transmittance of clam’s mantle cavity changes in the presence or absence of the parasite. This relative change in the normalized optical transmittance has been used as an effective spectral feature for designing parasite detectors. As a proof of concept two detectors have been designed. The first detector, which relies on all the hyperspectral information, achieved a perfect detection accuracy in identifying parasite-infected clams. The second detector, which is based on a reduced number of hyperspectral bands, achieved an 85% detection accuracy. These results provide insights on the fundamental trade-off between detection accuracy and the amount of hyperspectral information for parasite detection in clams.

Detection of omega-3 fatty acid in designer eggs using hyperspectral imaging

Quantification of n-3 polyunsaturated fatty acids (PUFA) is an important aspect in quality control of designer eggs. Prediction of these fatty acids would be of value for egg grading stations. The aim of the present study was to evaluate the ability of hyperspectral imaging (HSI) to classify the eggs according their content in n-3 PUFA as a rapid and non-destructive technique. Furthermore, HSI was used to predict the content of n-3 PUFA in designer eggs. Normal eggs, eggs containing 75 mg and eggs containing 125 mg docosahexaenoic acid were purchased from a grocery store. After hyperspectral transmittance images were collected in the spectral region of 900–1,700 nm, the eggs were broken and analyzed chemically in order to measure fatty acid content. Regions of interest (ROIs) of about 3,000 pixels from each egg were selected. K-means analysis was performed to classify eggs into the three aforementioned types. The classification accuracy was 100%. A partial least-squares regression model was built and used to link the ROIs with results obtained from the chemical analysis. The correlation coefficients between the measured and predicted values of alpha-linolenic acid, eicosapentaenoic acid, and docosahexaenoic acid were 0.94, 0.73 and 0.87, respectively. The root mean square error values for the validation sets were 38.65, 3.31 and 7.03, respectively. These results showed that HSI could be used both for discriminating the eggs between normal and designer eggs and for prediction of n-3 PUFA content in eggs.

Determining optimal wavebands using genetic algorithm for detection of internal insect infestation in tart cherry

This paper reports the results of waveband selection for detecting internal insect infestation in tart cherries as a precursor to development of a dedicated multispectral vision system. A genetic algorithm (GA) approach was applied on hyperspectral transmittance images (580–980 nm) and reflectance spectral data (590–1,550 nm) acquired from both intact and infested tart cherries. The GA analysis indicates that the ability of using transmittance imaging approach for detecting internal insect infestation in tart cherries would be limited. According to the GA analysis on the reflectance spectra, visible wavelengths were of less importance than NIR wavelengths for the purpose of distinguishing intact cherries from infested ones. The PLSDA results indicate that models built with three or four GA selected wavelength regions gave similar classification accuracy to the model built with full wavelength region, which demonstrates the efficiency of the GA variable selection procedure. However, due to the stochastic nature of the GA, the efficiency of using these wavebands in a multispectral vision system needs to be verified in future work.

Embedded bone fragment detection in chicken fillets using transmittance image enhancement and hyperspectral reflectance imaging

This paper is concerned with the detection of bone fragments embedded in compressed de-boned skinless chicken breast fillets by enhancing single-band transmittance images generated by back-lighting and exploiting spectral information from hyperspectral reflectance images. Optical imaging of chicken fillets is often dominated by multiple scattering properties of the fillets. Thus, resulting images from multiple scattering are diffused, scattered and low contrast. In this study, a fusion of hyperspectral transmittance and reflectance imaging, which is a non-ionized and non-destructive imaging modality, was investigated as an alternative method to the conventional transmittance X-ray imaging technique which is an ionizing imaging modality. An image formation model, called an illumination–transmittance model, was applied for correcting non-uniform illumination effects so that embedded bones are more easily detectable by a simple segmentation method using a single threshold value. Predicted bones from the segmentation were classified by the nearest neighbor classifier that was trained by the spectral library of mean reflectance of chicken tissues like fat, meat and embedded bones. Experimental results with chicken breast fillets and bone fragments are provided.