In the literature, several active learning techniques have been proposed for reducing the cost of data annotation. However, it is questionable whether the sample selection is fair with respect to sensitive attributes. Even when the active learning model considers fairness, it comes with a cost of reduced accuracy performance. Thus, it remains an open challenge to design an active learning algorithm that can maintain performance as well as fairness to underprivileged groups. This paper presents a novel active learning strategy called Fair Active Learning using fair Clustering, Uncertainty, and Representativeness (FAL-CUR) that provides a high accuracy while maintaining fairness during the sample acquisition phase. We introduce the FAL-CUR sample acquisition function that computes each sample's representative score based on the uncertainty and similarity score for sample selection. This acquisition function is added on top of the fair clustering method to add fairness constraints to the active learning method. We perform extensive experiments on four real-world datasets to compare the performance of the proposed methods. The experimental results show that the FAL-CUR algorithm maintains the performance accuracy while achieving high fairness measures and outperforms state-of-the-art methods on well-known fair active learning problems.