As a common passive interference method in modern battlefield,smoke screen jamming has excellent photoelectric countermeasure effect,strong tactical application mobility and high cost effectiveness ratio.In view of the lack of research on quantitative assessment of smoke screen interference based on images,five methods are used to extract features from infrared images of targets under smoke screen interference,including grayscale features,LBP based texture features,Haar like rectangular features,directional gradient histogram features and depth features from the perspective of tracking.By calculating the similarity between the feature extraction result and the subjective score,the ability of different image features to characterize the target under smoke screen interference is obtained.Through the analysis and verification of the data obtained from eight sets of field tests,it is concluded that the three features with better target representation ability under smoke screen interference are the directional gradient histogram feature,the Haar like rectangle feature and the depth feature.