In order to accurately estimate the tire wear degree and improve the safety of vehicle operation, a tire wear degree estimation method was proposed based on the strain features inside the tire and IWOA-SVR(improved whale optimization algorithm-support vector regression). The tire models with different wear degrees, pressures, loads and speeds were established by the finite element method, and then the strain features inside the tire were extracted and the key features were selected by the correlation analysis. Based on the pressures, loads, speeds and the selected key features, the IWOA-SVR model was established to estimate the tire wear degree, and the parameters of SVR were optimized by IWOA. Finally, the test was carried out under different wear degrees and different operating conditions, and the results show that the estimation accuracy was improved with the IWOA-SVR model, and the estimation mean absolute error is not more than 0.21 mm, which can meet the needs of tire wear detection.