Abstract: A landmark recognition and localization system with an enhanced robustness against large changes of lighting conditions eas presented for outdoor mobile robot. A near infrared illuminator was used for reducing the interference of natural light and shadow. The omnidirectional camera could capture images with more landmarks in a wide range of environment. For improving the landmark's recognition rate, a dynamic OTSU binarization method was adopted for enhancing the feature of landmarks in uneven lighting image. The outdoor realrobot localization experiments demonstrate the system’s robustness.