In order to find the optimal feature subset for target recognition in foliage environment, a novel feature selection method based on quantum genetic algorithm (QGA) is proposed in this paper. The real data used to identify the targets were collected by ultra-wideband (UWB) radar system. Support vector machine (SVM) classifier is adopted to evaluate the proposed algorithm. The experimental results indicate that QGA can achieve higher accuracy of target recognition with fewer features when face with the feature selection problem. Also, the comparison between the proposed and the traditional genetic algorithm (GA)-based method will be discussed. According to the obtained results, the method presented in this paper is more effective for target identification.