In order to study the feed processing and development technology of the renewable plant Humulus scandens, Humulus scandens were collected, dried and comminuted at the stage of leaf unfolding, flower opening, flower blooming and yellow withering, and processed into direct raw material group, silage natural fermentation group and silage plus bacteria fermentation group. The contents of main nutrients and total flavonoids in Humulus scandens feed were determined and compared. The microbial quantity and fermentation quality of Humulus scandens in withered and yellow period in silage natural fermentation group and silage plus bacteria fermentation group were detected. The results showed that there was no significant change in calcium and phosphorus content in the three groups. The other nutrients in silage natural fermentation group and silage plus bacteria fermentation group were slightly lower than those in direct raw material group. The content of total flavonoids in silage plus bacteria fermentation group was significantly higher than that in other groups. In terms of fermentation quality and microbial quantity, the fermentation of silage natural fermentation group was not obvious, the pH value was on the high side, the fermentation of silage plus bacteria group was obvious, and the quality was also higher. In conclusion, if Humulus scandens is directly used as feedstuff, the best collection time is the blooming period of flowers. If silage and lactobacillus fermentation are used, the value of feedstuff is the highest, but the value of silage natural fermentation is the lowest, so it is not recommended to choose.

Taking the extraction voltage, liquid-solid ratio and extraction time as the factors, and the yield of Ligustrum lucidum seed oil as the evaluation index, response surface methodology was used to explore the best technology of Ligustrum lucidum seed oil flash extraction method, and its antioxidant activity was analyzed. The results showed that the optimum extraction conditions were as follows: voltage 140 V, ratio of liquid to material 12:1, extraction time 150s, and the oil yield was 13.53%. The results showed that Ligustrum lucidum seed oil had certain scavenging ability to DPPH · free radical and hydroxyl radical, and its scavenging ability were positively correlated with its mass concentration. When the mass concentration was 2.5mg/ml, the DPPH · free radical scavenging rate was 32.25% and the hydroxyl radical scavenging rate was 43.83%.

A primary problem faced during previous research was the gap in limited and unbalanced quantity of prior samples between computer classification tasks and targeted remote sensing applications. This paper presents the fusion method to overcome this limitation. It offers a novel method based on knowledge transfer and feature association, a strong combination of transfer learning and data fusion. The former reuses layers trained on complete data sets to compute a mid-level representation of the specific target. The latter brings additional information from heterogeneous sources to enrich the features in the target domain. Firstly, a basic convolutional neural network (B_CNN) is pretrained on to the CIFAR-10 dataset to produce a stable model responsible for general feature extraction from multiple inputs. Secondly, a transfer CNN (Trans_CNN) with fine-tuned and transferred parameters is constraint-trained to fit and switch between differing tasks. Meanwhile, the feature association (FA) frames a new feature space to achieve integration between training and testing samples from different sensors. Finally, on-line detection can be completed based on Trans_CNN to explore a state-of-the-art method to overcome the inadequate sample problems in real remote sensing applications rather than produce an unrolled version of training methods or structural improvement in CNN. Experimental results show that target detection rates without homogeneous prior samples can reach 85%. Under these conditions, the traditional CNN model is invalid.

No abstract

No abstract