A novel method, the Qth order Mixed Moment Structure Partition Function (Q-MMSPF) method, is proposed for the detection of long range cross- correlations and multifractality. With this method, the multifractal cross-correlation characteristic of sea clutters was investigated. The analysis, based on the real sea clutter data, shows that the cross-correlation multifractality is quite weaker between the two sea clutter series, and the multifractality is significant between two time series of target plus sea clutter, whereas the multifractality is in the middle for the case, which is between target time series and sea clutter time ones. Consequently, an approach was suggested to detect the targets in sea clutter. The test results show that the target can be clearly distinguished from the sea clutter background with the proposed feature-based method. 