During May 2026, contributed to the opencv/opencv repository by enhancing validation for feature detector parameters within the features2d module. This work focused on strengthening error handling and input validation across multiple computer vision algorithms, ensuring that invalid configurations are caught early and do not propagate into runtime. The approach involved integrating robust parameter checks and improving error messaging, which streamlines debugging and reduces support overhead. Implemented in C++ with an emphasis on unit testing, these changes improved the reliability and stability of downstream pipelines, while maintaining compatibility with existing workflows and providing comprehensive regression tests for detector parameter configurations.