Search

Chicken disease identification with manual monitoring cannot fully meet the large-scale production in livestock and poultry breeding industry, due to slow speed and chickens prone to cross-infection. The characteristics of chicken feces can be one of the important indicators to reflect the health status of chickens. The different colors and traits of feces are associated with different chicken diseases. In this study, a classification and discrimination model was proposed for the chicken feces using visible-near-infrared spectroscopy. A research foundation was laid for the final realization of early warning of chicken diseases. Four types of typical chicken feces sample were selected, including the normal, red blood streak, green and fodder feces. The spectral data of samples was scanned in the 400-900 nm band. Each type of chicken feces samples was divided into the correction sets (162 samples) and test sets (53 samples) at 3:1, according to the principle of randomness. Multivariate scattering correction, Savitzky-Golay convolution smoothing, and Z-scores normalization were used to preprocess the data. A PLS-DA (partial least squares discriminant analysis) model was then established to select the optimal pretreatment, according to the evaluation index. An improved shuffled frog leaping algorithm was proposed to further optimize the existing partial least squares discriminant analysis model. The calculation rate of the model was effectively improved to obtain a more accurate number of principal factors. The number of iterations was determined to minimize the root mean square error of crossvalidation. The PCA (principal component analysis), CARS (competitive adaptive reweighted sampling), and ISFLA (improved shuffled frog leaping algorithm) were used to reduce the data dimensionality of the processed samples using the preferred pretreatment. Finally, a categorical discriminant model was established after optimization. The results showed that the PLS-DA model established by the SG convolution smoothing-MSC-Z-Scores shared the better performance of normal stool sample data, and for the three types of abnormal samples, the PLS-DA model established by MSC performed better. The accuracies of the test sets were achieved at 74.07%, 91.98%, 96.15%, and 100%, respectively, for four types of feces samples. The optimal data preprocessing was determined using model evaluation. The ISFLA-PLS-DA was used to distinguish between the normal, redblooded feces and green feces, where the accuracies of the test set were 92.27%, 92.59% and 100%, respectively. More importantly, the PCA-PLS-DA should be used for the discrimination in the fodder feces, where the accuracy of the test set was 100%. The dimensionality of the data was reduced by feature extraction to construct the lightweight PLS-DA model. The accuracy of the correction set was significantly improved in each typical stool sample. Specifically, the accuracies of the correction set were significantly improved before optimization, which increased by 7.70, 0.61, and 3.85 percentage point, respectively, of normal, red-blooded feces and green feces. Therefore, rapid, accurate and non-contact detection was achieved in the different types of typical chicken feces using visible-near-infrared spectroscopy detection combined with the characteristic wavelength selection and PLS-DA discrimination. The finding can also provide the strong reference for the intelligent detection technology of chicken disease. [ABSTRACT FROM AUTHOR]

Remote sensing technology has been widely used to monitor the changes in SPAD, which is an important parameter. In this study, the multispectral images were acquired by a six-rotor unmanned aerial vehicle, and the SPAD of winter wheat was measured to carry out the estimation research. The four growth stages with the most obvious changes in SPAD were selected, namely the jointing stage, booting stage, heading stage, and flowering stage. The camera with five bands (475, 560, 668, 717, and 840 nm) was used to collect multispectral canopy leaves at the four stages. A total of four data collections were performed to extract spectral reflectance data and the SPAD was measured from 1st April to 27th April 2018. A total of 65 samples were selected and recorded with GPS. The test area was divided into 65 sample zones with each one measuring 2.5 m×25 m, of which one sample area of 1 m × 1 m was selected. All the zones were in a rectangle, so they could be evenly distributed 8 m from the center of the cell in the horizontal direction. The overall samples were S-shaped distribution. The samples in the middle were located at the center of the rectangular cell. The SPAD of 65 samples were measured by SPAD-502 chlorophyll meter at the same time when the UAV data was collected. In the sample area, seven leaves of different canopy parts were selected to measure the tip, middle, and base. The average of the three parts was used as the SPAD values of the leaf. Finally, the average value of the leaf blades was taken as the final SPAD value of the sample. The canopy reflectance data was extracted from multispectral images. And then the correlation coefficients of SPAD values and spectral reflectance data in four growth stages were analyzed. Herein, the reflectivity of single-band and SPAD directly had serious collinearity problems so principal component regression, stepwise regression, and ridge regression these three methods were chosen to solve it. After that, the SPAD inversion models were established separately by using the reflectance data and the SPAD values as the data source. The best inversion model and stage were selected by comparison. The results showed that a high correlation was obtained between the SPAD and canopy spectral reflectance. In the visible light band, the negative correlation was observed between canopy spectral reflectance and SPAD at the jointing stage, booting stage, and flowering stage. On the contrary, it was a positive correlation at the heading stage and a positive correlation at the red-edge and near-infrared bands at all four stages. Compared with the main bands in the model expression, the frequency of passing the screening in different growth stages was different. The highest passing frequency was the near-infrared band in the jointing stage. The blue band was selected at the booting stage, the near-infrared band at the heading stage, and the green and red bands at the flowering stage. This study compared the prediction accuracy of the models established by three regression methods. The results showed that the models of stepwise regression established at the heading stage had the highest inversion accuracy with the adjusted coefficient of determination was 0.77, and the root mean square error was 0.61. The validation showed the coefficient of determination was 0.73, and the root mean square error was 0.56. It indicated that the model could be used to estimate the crop coefficient. Compared with the four periods, the heading stage was the best inversion stage of SPAD value. The study results proved the feasibility of inversion of the winter wheat SPAD value by unmanned aerial vehicle multispectral remote sensing, and at the same time, it could provide a reference for the rapid monitoring of the SPAD value of other crops. [ABSTRACT FROM AUTHOR]