The performance of an OCR system depends upon printing quality of the input document. Many OCRs have been designed which correctly identify fine printed documents in Indian and other scripts. But, little reported work has been found on the recognition of the degraded documents. The performance of any standard OCR system working for fine printed documents decreases, if it is tested on degraded documents. Feature extraction is an important task for designing an OCR for recognizing degraded documents. In this paper, we have discussed efficient structural features selected for recognizing degraded printed Gurmukhi script characters.