Fingerprint somehow can be regarded as a relatively fullfledged application in biometrics. The use of this biometric modality is not limited to traditional public security area, but spread into the daily life, smart phone authentication control and e-payment, for instance. However, quality control of biometric sample is still a necessary task due in order to optimize the operational performance. Research works had shown that biometric systems performance could be greatly depressed for those uncontrolled conditions [2]. For fingerprint, the quality can be intuitively described as the clarity of its ridge and valley pattern, noise condition, and the feasibility of feature extraction such as minutiae points. However, intuitively bad quality fingerprint sample might generate high matching result for some cases and vice versa [10]. In this case, many previous research works contributed to validate fingerprint quality metrics in terms of the relation between biometric sample quality and system performance. Chen et al. [7] proposed quality metrics in both frequency domain and spatial domain, and discussed several criteria to evaluate quality metric, such as quality index predicting matching performance. Fernandez et al. [1] compared correlations between several previously proposed quality metrics. Grother et al. [10] discussed quality evaluation methods in terms of detection error trade-off characteristics (DET), the effect of low quality samples rejection rates on improving performance, Kolmogorov Smirnov (KS) statistic, and etc. Mohamad et al. [8] calculated the KS statistics for image-based quality metric of multimodal biometric data. The study of this paper makes an evaluation of a proposed quality metric generated by using a utility-based multi-features fusion approach [9]. The quality metric involves in several quality criteria, including no-reference image quality metric (NR-IQA) [24], measures derived from texture features, and measure based on the triplet representation of fingerprint minutiae. The paper is organized as follows. Section 2 presents a state of the art on fingerprint quality. In section 3, we describe the proposed quality metric called Q. Section 4 details experimental results. Conclusion and future works are given in section 5.