The traditional multispectral algorithm used for temperature field measurement on the surface of aeroengine turbine blades has high computational complexity, requiring singular value decomposition of spectral data for temperature field data calculation. In addition, the accuracy of turbine blade positioning signals is affected by simulated speed signals, resulting in deviations in temperature data. If no speed offset correction is performed, the calculated temperature data cannot accurately reflect the uneven combustion temperature and design defects of the blade discs and blades. To address the above issues, this paper optimizes the data processing algorithm for temperature field measurement on the surface of turbine blades in complex thermal environments, analyzes and constructs a multispectral temperature measurement model, and performs monochromatic temperature auxiliary calculation. Compared with the existing traditional multispectral optimization algorithm, the "multispectral + monochromatic temperature auxiliary correction" method improves the calculation efficiency by more than 30% without loss of temperature measurement accuracy. In addition, this paper proposes an adaptive speed offset correction algorithm that dynamically and adaptively adjusts the filter parameters. Compared with the existing adaptive correction algorithm, this method reduces the offset correction error by 15%. The optimized data processing algorithm for temperature field measurement on the surface of turbine blades can adapt to complex working conditions with multiple speeds, making up for the shortcomings of existing temperature measurement algorithms in engineering applications.