In the process of cryo-electron microscopy (Cryo-EM) data collection, the Weighted Back-Projection (WBP) algorithm serves as a classical reconstruction method that reconstructs the three-dimensional (3D) structure of a sample by weighting and back-projecting tilt images. However, as data scales continue to grow, existing serial WBP algorithms face challenges such as excessive computation time and the inability to fully utilize underlying hardware resources, significantly limiting their application in large-scale data processing scenarios. To address this issue, this paper proposes a high-efficiency parallel WBP algorithm optimized for multi-core CPUs. By incorporating techniques such as multi-threading, SIMD vectorization, memory access optimization, instruction pipeline reordering, and load balancing, the proposed algorithm significantly accelerates the WBP computation process, enhancing parallelism and computational efficiency. Experimental results demonstrate that, compared to the original serial algorithm, the parallel WBP algorithm achieves a maximum speedup of 93.67x, substantially improving the computational efficiency of Cryo-EM reconstruction. This work provides strong performance support for further research and advancements in this field.