This work presents a new light field multispectral radiation thermometry technique and apply it to accurately validate the design effectiveness of a composite cooling structure for aero-engine turbine blades. This temperature measurement method integrates multispectral radiation thermometry with light field imaging technology: by coupling an optical filter array with a light field camera, it enables synchronous acquisition of multi-band radiation information via a single snapshot. Combined with an intelligent optimized temperature inversion algorithm, it effectively minimizes measurement errors induced by emissivity assumptions and background radiation. Compared with the traditional discrete-point measurement method of embedding thermocouples, this method can accurately obtain the two-dimensional temperature field distribution of the blade, with a temperature measurement deviation of no more than 0.33% in the absence of cooling air. Experiments were performed under a mainstream gas temperature of 823 K to obtain the influence laws of flow ratio and temperature ratio on the blade surface temperature distribution and cooling efficiency. The results indicate that higher temperatures are concentrated at the blade leading edge and regions adjacent to the endwalls; the cooling efficiency reaches its peak at the position of 0.2 times the arc length on the pressure surface, while the regions near 0.4 times the arc length on the suction surface and the trailing edge of the pressure surface exhibit relatively low cooling efficiency. The flow ratio exerts a more significant influence on cooling efficiency than the temperature ratio: when the temperature ratio is maintained constant at 1.7, the average cooling efficiency increases from 0.30 to 0.71 as the flow ratio rises from 0.02 to 0.08, and the improvement in average cooling efficiency enters a bottleneck when the flow ratio exceeds 0.06. When the flow ratio is fixed at 0.08, the average cooling efficiency decreases from 0.72 to 0.68 with the temperature ratio increasing from 1.56 to 2.36.