Non-Destructive Detection of Abnormal Chicken Eggs by Using an Optimized Spectral Analysis System

Environmental pressures, such as temperature and light intensity, food, and genetic factors, can cause chicken eggs to develop abnormalities. The common types of internal egg abnormalities include bloody and damaged egg yolk. Spectrometers have been recently used in real-time abnormal egg detection research. However, there are very few studies on the optimization of measurement systems. This study aimed to establish optimum parameters for detecting of internal egg abnormalities (bloody and damaged-yolk eggs) using visible and near-infrared (Vis/NIR) spectrometry (192-1110 nm range) and multivariate data analysis. The detection performance using various system parameters, such as the types of light sources, the configuration of the light, and sensor positions, was investigated. With the help of collected data, a partial least-squares discriminant analysis (PLS-DA) model was developed to classify normal and abnormal eggs. The highest classification accuracy for the various system parameters was 98.7%. Three band selection methods, such as weighted regression coefficient (WRC), sequential feature selection (SFS), and successive projection algorithm (SPA) were used for further model optimization, to reduce the spectral bands from 1028 to less than 7. In conclusion the results indicate that the types of light sources and the configuration design of the sensor and illumination affect the detection accuracy for abnormal eggs.

Mathematical progression of avian egg shape with associated area and volume determinations

Development of nondestructive techniques for estimating egg parameters requires a comprehensive approach based on mathematical theory. Basic properties used to solve theoretical and applied problems in this respect are volume (V) and surface area (S). There are respective formulae for calculating V and S of spherical, ellipsoidal, and ovoid eggs in classical egg geometry; however, the mathematical description and calculation of these parameters for pyriform eggs have remained elusive. In the present study, we derived the appropriate formulae and established that this would be not only applicable and valid for the category of pyriform eggs, but also universal and explicit for all other naturally occurring avian egg shapes. Thus, we have demonstrated "mathematical progression" of this natural object, considering the egg as a sequence of geometric figures that transform from one to another in the following sequence of shapes: sphere → ellipsoid → ovoid (whose profile corresponds to Hügelschäffer's model) → pyriform ovoid.

A Nondestructive Measurement Method of Optical Fiber Young's Modulus Based on OFDR

A nondestructive measurement method based on an Optical frequency domain reflectometry (OFDR) was demonstrated to achieve Young’s modulus of an optical fiber. Such a method can be used to measure, not only the averaged Young’s modulus within the measured fiber length, but also Young’s modulus distribution along the optical fiber axis. Moreover, the standard deviation of the measured Young’s modulus is calculated to analyze the measurement error. Young’s modulus distribution of the coated and uncoated single mode fiber (SMF) samples was successfully measured along the optical fiber axis. The average Young’s modulus of the coated and uncoated SMF samples was 13.75 ± 0.14, and 71.63 ± 0.43 Gpa, respectively, within the measured fiber length of 500 mm. The measured Young’s modulus distribution along the optical fiber axis could be used to analyze the damage degree of the fiber, which is very useful to nondestructively estimate the service life of optical fiber sensors immersed into smart engineer structures.

Classification of Watermelon Seeds Using Morphological Patterns of X-ray Imaging: A Comparison of Conventional Machine Learning and Deep Learning

In this study, conventional machine learning and deep leaning approaches were evaluated using X-ray imaging techniques for investigating the internal parameters (endosperm and air space) of three cultivars of watermelon seed. In the conventional machine learning, six types of image features were extracted after applying different types of image preprocessing, such as image intensity and contrast enhancement, and noise reduction. The sequential forward selection (SFS) method and Fisher objective function were used as the search strategy and feature optimization. Three classifiers were tested (linear discriminant analysis (LDA), quadratic discriminant analysis (QDA), and k-nearest neighbors algorithm (KNN)) to find the best performer. On the other hand, in the transfer learning (deep learning) approaches, simple ConvNet, AlexNet, VGG-19, ResNet-50, and ResNet-101 were used to train the dataset and class prediction of the seed. For the supervised model development (both conventional machine learning and deep learning), the germination test results of the samples were used where the seeds were divided into two classes: (1) normal viable seeds and (2) nonviable and abnormal viable seeds. In the conventional classification, 83.6% accuracy was obtained by LDA using 48 features. ResNet-50 performed better than other transfer learning architectures, with an 87.3% accuracy which was the highest accuracy in all classification models. The findings of this study manifested that transfer learning is a constructive strategy for classifying seeds by analyzing their morphology, where X-ray imaging can be adopted as a potential imaging technique.

Nondestructive measurement of kiwifruit firmness, soluble solid content (SSC), titratable acidity (TA), and sensory quality by vibration spectrum

Maturity is a key attribute to evaluate the quality and acceptability of fruit products. In this study, the impact method was used for nondestructive measurement of kiwifruit maturity. The fruit was vertically dropped onto an impact plate, and an accelerometer was used to measure the response signal. Then, fruit firmness, soluble solid content (SSC), titratable acidity (TA), and sensory scores were measured to determine the kiwifruit maturity. In addition, different modeling methods were proposed for data analysis. The results showed that the optimized prediction results were obtained by the principal component analysis-back-propagation neural network (PCA-BPNN) method for both quantitative and qualitative analysis. The optimized correlation coefficient between prediction and actual values (r p) and root mean square error of prediction (RESEP) for firmness, SSC, TA, and sensory score were 0.881 (2.359N), 0.641 (1.511 Brix), 0.568 (0.023%), and 0.935 (0.693), respectively. The optimized discriminant accuracy for immature, mature, and overmature kiwifruits was 94.2% and 92.1% for calibration and validation, respectively. Such results indicated the feasibility of the proposed impact method for kiwifruit maturity evaluation.

Near-Infrared Transmittance Spectral Imaging for Nondestructive Measurement of Internal Disorder in Korean Ginseng

The grading of ginseng (Panax ginseng) including the evaluation of internal quality attributes is essential in the ginseng industry for quality control. Assessment for inner whitening, a major internal disorder, must be conducted when identifying high quality ginseng. Conventional methods for detecting inner whitening in ginseng root samples use manual inspection, which is time-consuming and inaccurate. This study develops an internal quality measurement technique using near-infrared transmittance spectral imaging to evaluate inner whitening in ginseng samples. Principle component analysis (PCA) was used on ginseng hypercube data to evaluate the developed technique. The transmittance spectra and spectral images of ginseng samples exhibiting inner whitening showed weak intensity characteristics compared to normal ginseng in the region of 900-1050 nm and 1150-1400 nm respectively, owing to the presence of whitish internal tissues that have higher optical density. On the basis of the multivariate analysis method, even a simple waveband ratio image has the great potential to quickly detect inner whitening in ginseng samples, since these ratio images show a significant difference between whitened and non-whitened regions. Therefore, it is possible to develop an efficient and rapid spectral imaging system for the real-time detection of inner whitening in ginseng using minimal spectral wavebands. This novel strategy for the rapid, cost-effective, non-destructive detection of ginseng's inner quality can be a key component for the automation of ginseng grading.

Classification Method for Viability Screening of Naturally Aged Watermelon Seeds Using FT-NIR Spectroscopy

Viability analysis of stored seeds before sowing has a great importance as plant seeds lose their viability when they exposed to long term storage. In this study, the potential of Fourier transform near infrared spectroscopy (FT-NIR) was investigated to discriminate between viable and non-viable triploid watermelon seeds of three different varieties stored for four years (natural aging) in controlled conditions. Because of the thick seed-coat of triploid watermelon seeds, penetration depth of FT-NIR light source was first confirmed to ensure seed embryo spectra can be collected effectively. The collected spectral data were divided into viable and nonviable groups after the viability being confirmed by conducting a standard germination test. The obtained results showed that the developed partial least discriminant analysis (PLS-DA) model had high classification accuracy where the dataset was made after mixing three different varieties of watermelon seeds. Finally, developed model was evaluated with an external data set (collected at different time) of hundred samples selected randomly from three varieties. The results yield a good classification accuracy for both viable (87.7%) and nonviable seeds (82%), thus the developed model can be considered as a “general model” since it can be applied to three different varieties of seeds and data collected at different time.

Nondestructive In Situ Measurement Method for Kernel Moisture Content in Corn Ear

Moisture content is an important factor in corn breeding and cultivation. A corn breed with low moisture at harvest is beneficial for mechanical operations, reduces drying and storage costs after harvesting and, thus, reduces energy consumption. Nondestructive measurement of kernel moisture in an intact corn ear allows us to select corn varieties with seeds that have high dehydration speeds in the mature period. We designed a sensor using a ring electrode pair for nondestructive measurement of the kernel moisture in a corn ear based on a high-frequency detection circuit. Through experiments using the effective scope of the electrodes’ electric field, we confirmed that the moisture in the corn cob has little effect on corn kernel moisture measurement. Before the sensor was applied in practice, we investigated temperature and conductivity effects on the output impedance. Results showed that the temperature was linearly related to the output impedance (both real and imaginary parts) of the measurement electrodes and the detection circuit’s output voltage. However, the conductivity has a non-monotonic dependence on the output impedance (both real and imaginary parts) of the measurement electrodes and the output voltage of the high-frequency detection circuit. Therefore, we reduced the effect of conductivity on the measurement results through measurement frequency selection. Corn moisture measurement results showed a quadric regression between corn ear moisture and the imaginary part of the output impedance, and there is also a quadric regression between corn kernel moisture and the high-frequency detection circuit output voltage at 100 MHz. In this study, two corn breeds were measured using our sensor and gave R² values for the quadric regression equation of 0.7853 and 0.8496.