As the core field in the development of global renewable energy, wind power is accelerating the green transformation of energy structure. In the design and efficiency assessment of wind power systems, wind turbine height is a critical parameter that directly influences turbine performance and overall power generation efficiency. Therefore, accurately and efficiently estimating turbine height is significant for site selection, layout optimization and system operation of wind power projects. This paper proposes a wind turbine height estimation method that integrates deep learning and machine learning, based on high-resolution remote sensing imagery. The method first utilizes a structurally improved DeepLabv3+ network to achieve precise segmentation of wind turbine shadows, from which shadow contour features are extracted and shadow lengths are calculated. Subsequently, in combination with geometric variables such as solar elevation angle, solar azimuth angle, satellite elevation angle and satellite azimuth angle, both multiple linear regression and random forest regression models are constructed to estimate wind turbine height. To validate the accuracy and adaptability of the proposed method, a dataset covering various geographic environments and turbine types is constructed to reflect practical application scenarios under diverse conditions. Experimental results demonstrate that, compared to the multiple linear regression model, the random forest regression model achieves superior estimation accuracy and stronger generalization ability, with mean relative errors of 1.98% on the training set and 5.33% on the test set. The proposed method, which combines deep learning and machine learning, not only reduces reliance on manual parameter inputs and complex modeling, but also achieves a high level of automation. It offers an efficient and cost-effective technical solution for the planning, construction and operation of wind farms, and holds significant potential for improving the overall efficiency and economic performance of wind power generation.