Search

The behaviours of the pig are often closely related to their health. Pig recognition is very important for pig behaviour analysis and digital breeding. Currently, the early signs and abnormal behaviours of sick pigs in breeding farms are mainly completed by human observation. However, visual inspection is labour intensive and time-consuming, and it suffers from the problems of individual experiences and varying environments. An improved ResNet model was proposed and applied to detect individual pigs in this study based on deep learning knowledge. The developed model captured the features of pigs applying across layer connections, and the ability of feature expression was improved by adding a new residual module. The number of layers was reduced to minimise the net complexity. Generally, the ResNet frame was developed by reducing the number of convolution layers, constructing different types of the residual module and adding the number of convolution kernels. The training accuracy and testing accuracy reached 98.2% and 96.4%, respectively, when using the improved model. The experiment results showed that the method proposed in this paper for checking living situations and disease prevention of commercial pigs in pig farms is potential.

It is of great significance to identify soybean seeds with high oil content since the oil content of soybean seeds decides oil yield. At present, related researches mostly used machine learning algorithm to identify soybean varieties with small samples. In this study, 5800 spectral data samples of 58 varieties in the range of 400–1000 nm were obtained. An acceptable method that combines hyperspectral imaging with one-dimensional convolutional neural network was proposed to distinguish high oilcontent soybean seeds. Moreover, traditional machine learning models, including support vector machine, k-nearest neighbor algorithm, and partial squares discriminant analysis, were also established in the experimental study. The effects of four preprocessing methods, namely moving window smoothing, standard normal variate, multivariate scattering correction, and Savitzky–Golay, were compared when building support vector machine-based identification models. The results showed that the model using multivariate scattering correction gave better test accuracy (94.5%), indicating that for this study, multivariate scattering correction was a more suitable method than others. Meanwhile, the study compared the performance of the four models by expanding the number of samples. The results showed that the proposed one-dimensional convolutional neural network model was more stable. The average accuracy of the training set and test set was 96% and 93%, respectively. Therefore, hyperspectral data combined with one-dimensional convolutional neural network was effective in identifying soybean seeds with high oil content. [ABSTRACT FROM AUTHOR]

The use of computer vision for estimating quality in agriculture products has become wide spread in recent years and the composition, variety, or ripeness can be estimated. On the other hand, the appearance is one of the most worrying issues for producers due to its influence on quality. In this research, computer vision technology combined with BP artificial neural network (ANN) was developed to identify soybean frogeye, mildewed soybean, worm-eaten soybean and damaged soybean. Thirty-nine characteristic parameters from color, texture and shape characteristics were computed after preprocessing the acquired soybean images. The dimensionality of the characteristic parameters was reduced from 39 dimensionalities to 12 dimensionalities using the method of principal component analysis (PCA). MALAB software was used to build a prediction model according to 12 characteristic parameters. The identification accuracies of soybean frogeye, mildewed soybean, damaged soybean and worm-eaten soybean are 96%, 95%, 92%, and 92%, respectively. And the accuracy for heterogeneous soybean seeds with several diseases is 90%. The results show that the prediction model constructed by BP neural network can identify the diseases of soybean seeds. And it is useful to estimate appearance quality of soybean by computer vision applying BP neural network. [ABSTRACT FROM AUTHOR]

Different soybean seed varieties have different components (oil, protein, fat etc.) content. Identification of soybean seed varieties is a critical factor that improves the quality of produced soybean. In this study, hyperspectral image technique was applied in order to classify soybean seeds based on their varieties. The spectral reflectance data was collected using the optical sensor system with spectral region of 1000-2500nm. Principal component analysis (PCA) was performed to reduce the dimensionality of the data and remove the redundancy. Scores of four PCs were used as input features in the classification algorithm. Four texture feature parameters (angular second moment, energy, entropy and correlation) were extracted from each feature image selected by PCA. For the extraction of specific features, four significant feature parameters were computed from the 16 characteristic variables. Artificial neural network (ANN) classifier was employed for classification using top selected features. The obtained average training accuracy rate was 97.50% and the average testing accuracy rate was 93.88%. Thus, the results confirmed that hyperspectral image technique in-conjunction with BP neural network could be useful for soybean seed varieties classification. [ABSTRACT FROM AUTHOR]