The invention belongs to the field of an intelligent recording and broadcasting system, and relates to a method of detecting and positioning the face of a standing person in an intelligent recording and broadcasting system. The specific operating steps include the steps of initializing a video file recorded by a camera and performing frame processing; selecting an appropriate threshold to binarize a differential image; then morphologically operating the image; marking and retaining active areas with the minimum bounding rectangle, and performing illumination compensation; afterwards, screening out the interfering active areas by skin color detection; performing classified screening according to the human face area proportion, and excluding the areas with the skin color but no human face; and finally conducting a data analysis according to the processed data to mark the human face. The design concept of the method is clever, the detection principle is simple, the operation and usage are easy, the face of a standing person can be detected and positioned only through one camera, equipment requirements are low, meanwhile, the detect results are highly accurate, the application environment is friendly, and the market prospects are broad.