Yuzhen Lu

Fast and nondestructive determination of protein content in rapeseeds (Brassica napus L.) using Fourier transform infrared photoacoustic spectroscopy (FTIR‐PAS)

BACKGROUND Fast and non-destructive determination of rapeseed protein content carries significant implications in rapeseed production. This study presented the first attempt of using Fourier transform mid-infrared photoacoustic spectroscopy (FTIR-PAS) to quantify protein content of rapeseed. The full-spectrum model was first built using partial least squares (PLS). Interval selection methods including interval partial least squares (iPLS), synergy interval partial least squares (siPLS), backward elimination interval partial least squares (biPLS) and dynamic backward elimination interval partial least squares (dyn-biPLS) were then employed to select the relevant band or band combination for PLS modeling. RESULTS The full-spectrum PLS model achieved an ratio of prediction to deviation (RPD) of 2.047. In comparison, all interval selection methods produced better results than full-spectrum modeling. siPLS achieved the best predictive accuracy with an RPD of 3.215 when the spectrum was sectioned into 25 intervals, and two intervals (1198-1335 and 1614-1753 cm(-1) ) were selected. iPLS excelled biPLS and dyn-biPLS, and dyn-biPLS performed slightly better than biPLS. CONCLUSION FTIR-PAS was verified as a promising analytical tool to quantify rapeseed protein content. Interval selection could extract the relevant individual band or synergy band associated with the sample constituent of interest, and then improve the prediction accuracy of the full-spectrum model.

Determination of the contents of magnesium and potassium in rapeseeds using FTIR-PAS combined with least squares support vector machines and uninformative variable elimination

Fourier transform mid-infrared photoacoustic spectroscopy (FTIR-PAS) was employed to determine the contents of magnesium and potassium in rapeseeds. A total of 180 samples were collected for this purpose. A Savitzky–Golay filter was used for the spectral pretreatment. The whole sample set was divided into calibration and prediction sets composed of 135 and 45 samples, respectively. To build calibration models, partial least squares (PLS), least squares support vector machines (LS-SVM) and least squares support vector machines combined with uninformative variable elimination (UVE-LS-SVM) were used. The best results for quantification of both magnesium and potassium were achieved by UVE-LS-SVM models compared to the PLS models. The highest values of RPD (ratio of percentage deviation) were 2.5 and 2.25 for the prediction of magnesium and potassium, respectively. This work verified the good promise of FTIR-PAS combined with LS-SVM to quantify mineral nutrients of rapeseeds.

Fast demodulation of pattern images by spiral phase transform in structured-illumination reflectance imaging for detection of bruises in apples

Structured-illumination reflectance imaging (SIRI) was used to detect apple bruises.Spiral phase transform (SPT) was proposed for amplitude demodulation in SIRI.SPT showed improved performance compared to conventional demodulation methods.The use of SPT can increase the imaging speed of SIRI by fewer pattern images. A structured-illumination reflectance imaging (SIRI) system was recently developed in our laboratory for enhanced quality evaluation of horticultural products. It was implemented using a digital camera to acquire reflectance images from food products subjected to sinusoidal patterns of illumination, instead of conventional diffuse or uniform illumination. The reconstruction of amplitude images is a key step in implementing SIRI technique. Conventional methods require acquisition and subsequent demodulation of three phase-shifted pattern images, which limits the speed of image acquisition. This study proposed the use of spiral phase transform (SPT) for demodulation of pattern images acquired by SIRI. Three SPT-based methods involving only one or two images were investigated through numerical simulation, followed with experiments on the detection of fresh bruises in apples. Compared to conventional three-phase based and two-phase based Gram-Schmidt (GS) orthonormalization demodulation methods, the two-phase SPT method achieved the same or even higher demodulation accuracy, which applies to two pattern images that can be arbitrarily phase-shifted. SPT also allowed single-image based demodulation, although its performance still needs to be improved. This study demonstrated that SPT for amplitude demodulation can increase the imaging speed of SIRI by using one or two pattern images, which is a significant step towards real-time implementation of the technique.

Gram-Schmidt orthonormalization for retrieval of amplitude images under sinusoidal patterns of illumination.

Structured illumination using sinusoidal patterns has been used for optical imaging of biological tissues in biomedical research, and of horticultural products in food quality evaluation. Implementation of structured-illumination imaging relies on retrieval of amplitude images, which is conventionally achieved by a phase-shifting technique that requires collecting a minimum of three phase-shifted images. In this study, we have proposed Gram-Schmidt orthonormalization (GSO) to retrieve amplitude component (AC) images using only two phase-shifted images. We have proposed two forms of GSO implementation, and prior to GSO processing, we eliminated the direct component (DC) background by subtracting a DC image we recovered using a spiral phase function (SPF) in the Fourier space. We demonstrated the GSO methods through numerical simulations and application examples of detection of bruise defects in apples by structured-illumination reflectance imaging (SIRI). GSO performed comparably to conventional three-phase-based demodulation. It is simple, fast and effective for amplitude retrieval and requires no prior phase information, which could facilitate fast implementation of structured-illumination imaging.

Classification of rapeseed colors using Fourier transform mid-infrared photoacoustic spectroscopy

Fourier transform mid-infrared photoacoustic spectroscopy (FTIR-PAS) combined with multivariate discriminant analysis was employed to classify colors of rapeseeds. A total of 129 rapeseed varieties representing three colors (black, reddish and mottled-yellow) were scanned in the range of 500–4000 cm−1. A Savitzky–Golay algorithm was used for the spectral pretreatment. Principal components analysis (PCA) gave an overview of sample distribution in the score space of principal components. The whole sample set was divided into calibration and prediction sets, according to the Kennard–Stone algorithm. Classification models were developed using linear discriminant analysis combined with principal components analysis (PCA-LDA), partial least square discriminant analysis (PLS-DA), and support vector machine (SVM). Results showed that the best accuracy was achieved by the SVM model, with the overall error rates (ERs) of 1.1% and 2.5%, in calibration and prediction sets, respectively. Besides, the PLS-DA model performed slightly better than the PCA-LDA model. This work had demonstrated the good potential of FTIR-PAS to classify rapeseed colors.

Detection of fresh bruises in apples by structured-illumination reflectance imaging

Detection of fresh bruises in apples remains a challenging task due to the absence of visual symptoms and significant chemical alterations of fruit tissues during the initial stage after the fruit have been bruised. This paper reports on a new structured-illumination reflectance imaging (SIRI) technique for enhanced detection of fresh bruises in apples. Using a digital light projector engine, sinusoidally-modulated illumination at the spatial frequencies of 50, 100, 150 and 200 cycles/m was generated. A digital camera was then used to capture the reflectance images from ‘Gala’ and ‘Jonagold’ apples, immediately after they had been subjected to two levels of bruising by impact tests. A conventional three-phase demodulation (TPD) scheme was applied to the acquired images for obtaining the planar (direct component or DC) and amplitude (alternating component or AC) images. Bruises were identified in the amplitude images with varying image contrasts, depending on spatial frequency. The bruise visibility was further enhanced through post-processing of the amplitude images. Furthermore, three spiral phase transform (SPT)-based demodulation methods, using single and two images and two phase-shifted images, were proposed for obtaining AC images. Results showed that the demodulation methods greatly enhanced the contrast and spatial resolution of the AC images, making it feasible to detect the fresh bruises that, otherwise, could not be achieved by conventional imaging technique with planar or uniform illumination. The effectiveness of image enhancement, however, varied with spatial frequency. Both 2-image and 2-phase SPT methods achieved the performance similar to that by conventional TPD. SIRI technique has demonstrated the capability of detecting fresh bruises in apples, and it has the potential as a new imaging modality for enhancing food quality and safety detection.

Determination of Nitrogen in Rapeseed by Fourier Transform Infrared Photoacoustic Spectroscopy and Independent Component Analysis

Fourier-transform mid-infrared photoacoustic spectroscopy was utilized for rapid and nondestructive determination of nitrogen in rapeseeds. Rapeseed spectra were characterized by independent component analysis for quantitative calibration. A calibration model was built by using independent components as the input for partial least squares. Compared to full-spectrum partial least squares, the combined model achieved higher prediction accuracy with a residual predictive deviation of 2.06. Moreover, a genetic algorithm coupled with partial least squares was adopted to optimize the independent components for partial least square modeling and provide a further refined model with the highest residual predictive deviation of 2.12. A t-test verified a high congruence between results obtained by calibration models and the reference Kjeldahl method. This study demonstrated the promise of Fourier-transform mid-infrared photoacoustic spectroscopy for the determination of nitrogen in rapeseeds and the applicability of independent components for multivariate calibration.

Fast Bi-dimensional empirical mode decomposition as an image enhancement technique for fruit defect detection

Abstract Image enhancement is critical to detection of fruit defects by imaging techniques. Vignetting and noise are major image artifacts, which can seriously affect image segmentation results, especially in inspecting the curved-surface objects like fruit. It is common to use a calibration object or a mathematic model to reduce the vignetting effect in defect detection, but the approach is often cumbersome, inflexible, and difficult to achieve desired results. In this study, a new image enhancement method based on bi-dimensional empirical mode decomposition (BEMD) of images was proposed to isolate and subsequently remove the effects of vignetting and noise by means of selective image reconstruction. The new BEMD method, along with three other BEMD methods, was first tested on decomposing a synthetic image with artificially added vignetting and noise. The BEMD was found to be the most efficient for image decomposition in terms of computation time, and also give high-quality reconstructed images. Experiments were further conducted by applying the BEMD to the direct and amplitude component images of apple samples with subsurface bruising and surface defects, which were acquired by using a structured-illumination reflectance imaging (SIRI) system. BEMD effectively reduced the image vignetting and greatly enhanced the defect features of the apples, based on both visual inspection and quantitative evaluation. BEMD offers an effective tool for enhancing SIRI images, and it is also promising for image enhancement with other imaging modalities for fruit defect detection.