Qibing Zhu

Prediction of moisture content uniformity using hyperspectral imaging technology during the drying of maize kernel

Abstract Moisture content uniformity is one of critical parameters to evaluate the quality of dried products and the drying technique. The potential of the hyperspectral imaging technique for evaluating the moisture content uniformity of maize kernels during the drying process was investigated. Predicting models were established using the partial least squares regression method. Two methods, using the prediction value of moisture content to calculate the uniformity (indirect) and predicting the moisture content uniformity directly, were investigated. Better prediction results were achieved using the direct method (with correlation coefficients RP = 0.848 and root-mean-square error of prediction RMSEP = 2.73) than the indirect method (RP = 0.521 and RMSEP = 10.96). The hyperspectral imaging technique showed significant potential in evaluating moisture content uniformity of maize kernels during the drying process.

Penetration Depth Measurement of Near-Infrared Hyperspectral Imaging Light for Milk Powder

The increasingly common application of the near-infrared (NIR) hyperspectral imaging technique to the analysis of food powders has led to the need for optical characterization of samples. This study was aimed at exploring the feasibility of quantifying penetration depth of NIR hyperspectral imaging light for milk powder. Hyperspectral NIR reflectance images were collected for eight different milk powder products that included five brands of non-fat milk powder and three brands of whole milk powder. For each milk powder, five different powder depths ranging from 1 mm–5 mm were prepared on the top of a base layer of melamine, to test spectral-based detection of the melamine through the milk. A relationship was established between the NIR reflectance spectra (937.5–1653.7 nm) and the penetration depth was investigated by means of the partial least squares-discriminant analysis (PLS-DA) technique to classify pixels as being milk-only or a mixture of milk and melamine. With increasing milk depth, classification model accuracy was gradually decreased. The results from the 1-mm, 2-mm and 3-mm models showed that the average classification accuracy of the validation set for milk-melamine samples was reduced from 99.86% down to 94.93% as the milk depth increased from 1 mm–3 mm. As the milk depth increased to 4 mm and 5 mm, model performance deteriorated further to accuracies as low as 81.83% and 58.26%, respectively. The results suggest that a 2-mm sample depth is recommended for the screening/evaluation of milk powders using an online NIR hyperspectral imaging system similar to that used in this study.

Maize and weed classification using color indices with support vector data description in outdoor fields

Abstract An automated method for maize and weed detection is very important to efficiently remove weeds and precisely calculate the quantity of maize. Color features were used in this study to investigate a simple maize-detection method using a color machine-vision system. Conventional image segmentation methods based on RGB values cannot separate maize from weeds because of the highly similar image RGB values of these plants. Thus, a post-processing algorithm was developed to distinguish maize from weeds after image preprocessing. Color indices were used to develop a classification model. The nine optimal features were selected by principal component analysis to reduce the effect of illumination. Finally, support vector data description was used as a classifier to differentiate maize from the mixes of different species of weeds. Pictures were taken by a commercial camera and used to verify the stability of the algorithm. Results show that the overall accuracy for three years is 90.19%, 92.36% and 93.87%, respectively. And the color indices used in this work were stable under various weather conditions and over time.

Model updating for the classification of different varieties of maize seeds from different years by hyperspectral imaging coupled with a pre-labeling method

Abstract The use of hyperspectral imaging technology combined with chemometrics is an effective nondestructive method for sorting seed varieties. However, the performance of the method is susceptible to the influence of time and depends on the training set used in the modeling process. The accuracy of classification models maybe deteriorate when they are used to differentiate the same variety of seeds harvested in different years, due to new variances in the test set are introduced by changes in the cultivation conditions, soil environmental conditions and climatic changes from one year to another. To maintain the accuracy and robustness of model, a model-updating algorithm for differentiating maize seed varieties from different years based on hyperspectral imaging coupled with a pre-labeling method was proposed in this work. The pre-label of each unlabeled sample was obtained using the original classification models developed by the least squares support vector machine classifier. The representative unlabeled samples, which had reliable pre-labels, were selected for updating classification models based on Pearson correlation coefficients. After model updating, the average classification accuracies were improved by 8.9%, 35.8% and 9.6%, compared with those of non-updated models for three test sets, respectively. This shows the effectiveness of the proposed method for classifying maize seeds of different years.

Pose Estimation of Sweet Pepper through Symmetry Axis Detection

The space pose of fruits is necessary for accurate detachment in automatic harvesting. This study presents a novel pose estimation method for sweet pepper detachment. In this method, the normal to the local plane at each point in the sweet-pepper point cloud was first calculated. The point cloud was separated by a number of candidate planes, and the scores of each plane were then separately calculated using the scoring strategy. The plane with the lowest score was selected as the symmetry plane of the point cloud. The symmetry axis could be finally calculated from the selected symmetry plane, and the pose of sweet pepper in the space was obtained using the symmetry axis. The performance of the proposed method was evaluated by simulated and sweet-pepper cloud dataset tests. In the simulated test, the average angle error between the calculated symmetry and real axes was approximately 6.5°. In the sweet-pepper cloud dataset test, the average error was approximately 7.4° when the peduncle was removed. When the peduncle of sweet pepper was complete, the average error was approximately 6.9°. These results suggested that the proposed method was suitable for pose estimation of sweet peppers and could be adjusted for use with other fruits and vegetables.

Optical property inversion of biological materials using Fourier series expansion and LS-SVM for hyperspectral imaging

Abstract Determination of the optical properties of biological materials based on steady-state spatially resolved diffuse reflectance imaging is a complicated inverse problem-solving process. An effective inverse algorithm is first necessary and validated. This article proposed a Fourier series expansion (FSE) coupled with least squares support vector machine (LS-SVM) as an inverse algorithm for determining the absorption coefficient (μa) and the reduced scattering coefficient () of biological materials. A hyperspectral imaging system was used to acquire scattering images and steady-state spatially resolved diffuse reflectance profiles of liquid phantoms. Experiment results demonstrated that the hyperspectral imaging system coupled with this inverse algorithm effectively improved prediction accuracy of both μa and of liquid phantoms. Tests on liquid phantoms showed that the mean relative errors of this inverse algorithm to be 11.03% for the absorption coefficient and 7.16% for the reduced scattering coefficient when corresponding Fourier coefficients of liquid phantoms were used to develop the prediction model. To further study the method, the Fourier coefficients calculated from normalized Monte Carlo simulation data were used to develop the prediction model for determining the optical properties of 36 liquid phantoms. The prediction errors were 15.96 and 10.91% for μa and , respectively. For all liquid phantoms, it was found that the prediction values of both μa and were generally in good agreement with their actual values. Therefore, the FSE–LS-SVM method provides an effectively means for improving the prediction accuracy of optical properties of biological materials.

Classification of Chinese green tea grade using laser-induced breakdown spectroscopy

Abstract. Tea is one of the most common and popular beverages all over the world. The accurate identification of tea grade is of great significance to ensure the interests of tea producers and consumers. In this paper, three grades of Wuxi baikhovi tea leaves were analyzed and identified using laser-induced breakdown spectroscopy (LIBS). A total of 40 optimal spectral peaks were automatically selected from full LIBS spectra by using successive projection algorithm (SPA), and the selected spectral peaks mainly represent the elemental difference in C, Fe, Mg, Mn, Al and Ca. Finally, partial least squares discriminant analysis (PLSDA) was used for developing classification model using selected optimal spectral peaks, the result shows the accuracy of 98.8% for 150 test samples. This study demonstrates LIBS is a useful technique for the identification and discrimination of tea grade in various tea products and is promising for real-time, fast, and reliable measurement.