Paeonia ostii is the representative of oil-utilized peony, its roots (PR) and leaves (PL) are discarded as by-products, resulting in a waste of resources. The exploration of extraction process of active ingredients from PR/PL is beneficial to the comprehensive utilization. In this work, the optimum process of Ultrasound-Assisted Extraction (UAEo) of total flavonoids content (TFC) from PR and PL was determined by single factor and response surface methodology. The results showed that UAEo was 80 min, 1:5 g/mL, 250 W, 33.83 mg Rutin/g dw for PR and 60 min, 1:10 g/mL, 250 W, 25.85 mg Rutin/g dw for PL. Then, Homogenization-Assisted Extraction (HAE), Homogenization-Ultrasound-Assisted Extraction (HUAEo) and Ultrasound-Homogenization-Assisted Extraction (UHAEo) were further analyzed. The highest PR(PL)-TFC by HUAEo at homogenization 5 min was 49.58 ± 0.25 mg/g with an increase of 46.6% (33.02 ± 0.04 mg/g with an increase of 27.7%). The highest PR/PL-total phenolic content by the HUAEo reached 77.84 ± 0.52 mg/g dw and 146.62 ± 2.77 mg/g dw for homogenization 3 min. However, there was no significant difference between HUAEo and UHAEo. In conclusion, the TFC increased with the extension of HAE time, and the combined extraction was higher yield than the single extraction.

Aiming at the low accuracy of traditional dance movement recognition methods, a movement recognition algorithm based on human posture estimation is proposed. Firstly, PAFs algorithm is adopted to recognize the spatial skeleton nodes of the human body model and the connection of human body joints, thus the human movement skeleton is obtained. According to the movement skeleton, the human body posture can be estimated. After the posture information is preprocessed and features are extracted, LSTM time series algorithm is used to classify and recognize the dance movements. The results show that the algorithm can clearly identify the dance movement skeleton nodes. For different movement categories, the recognition accuracy and recall rate of different movement categories are above 85%, and the recognition accuracy of curtsey movement is up to 95.2%. It can be seen that the recognition accuracy of this algorithm is significantly improved and different dance movement categories can be accurately recognized.

In order to adapt to the development and changes in the era of big data, many industries are undergoing reforms or are about to face reforms, including the education industry. In the era of big data, the teaching informatization level of many teachers needs to be improved, and the educational informatization methods of many schools also need to be improved, especially PE (sports) in schools. The purpose of this paper is to study the informatization of physical education courses based on data analysis technology. Using BDT to obtain data information in physical education courses and analyze the results of teaching information satisfaction, we selected two physical education colleges, C and D, with different teaching methods. Experimental research shows that the teaching performance of C college is significantly better than that of D college in one academic year. The highest score of College C is 96 and the lowest is 83. The highest score of D College is 77 points and the lowest score is 63 points. The teachers of College C are also satisfied with the new teaching method implemented by the whole school. The total proportion of satisfaction with the new teaching method is 61%. This method can help sports innovation and reform.