Liu, T., 2019. Injury prevention methods in island survival training. In: Guido-Aldana, P.A. and Mulahasan, S. (eds.), Advances in Water Resources and Exploration. Journal of Coastal Research, Special Issue No. 93, pp. 536–541. Coconut Creek (Florida), ISSN 0749-0208.

Aiming at the problem of low coincidence between simulated and actual values of injury prevention methods in island survival training for university students, a method of injury prevention in island survival training based on PCA-SIFT features and Bayesian decision-making is proposed. This paper analyzed the most common injuries in island survival training for university students: ankle and knee joint; it also used generalized cross validation criteria and genetic method to select the optimal threshold to denoise the curved wave coefficients obtained. The extension frequency of the denoised image is used to enhance the ankle and knee joint images of university students. The PCA-SIFT features of each kind of image are extracted randomly from the sample set, and the unstable features are eliminated by the matching ratio. Bayesian decision-making is used to classify and process the images to be classified, the relevant information of ankle and knee joint injury of university students is obtained, and the prevention and treatment plan is obtained according to the information of injury and disease. The experimental results show that, compared with other methods, the proposed method has 99% coincidence between simulated and actual values in predicting injury and illness in island survival training for university students, which verifies the effectiveness of the proposed method.