The load-bearing capacity analysis of prestressed concrete in bridge engineering is a core technology for structural safety evaluation. It has long faced challenges in insufficient detection accuracy under complex stress environments and low efficiency in multi-source data fusion. Traditional analysis methods rely on a single mechanical model or empirical experience, making it difficult to accurately capture the nonlinear relationship between crack development and load-bearing capacity degradation. Therefore, this study proposes a prestressed concrete load-bearing capacity analysis model based on a dual-threshold edge detection algorithm. Experimental results show that the accuracy of the improved edge detection algorithm reaches a maximum of 89.5% after iteration, with the misdetection rate of bridge cracks under various noise influences being as high as 9%. Evaluation of the fusion analysis model shows that the Mean Square Error of its load-bearing capacity is only 0.015 kN·m2, and the Coefficient of Determination is 0.98. These results indicate that the proposed prestressed concrete load-bearing capacity analysis model can effectively improve the prediction accuracy of load-bearing capacity under complex stress environments and accurately capture the nonlinear relationship between crack development and load-bearing capacity degradation. This study provides a new technical approach for bridge structural safety assessment and contributes to the development of intelligent monitoring and full-life-cycle maintenance technologies for prestressed concrete structures.