Content Based Image Retrieval technique is becoming increasingly important in various fields in order to store, manage and Retrieve images from database based on user query. Searching is done by image features such as texture, shape or different combinations of them. Texture feature plays an important role in image processing, computer vision and Pattern recognition. In this paper we propose a novel method of using dual tree complex wavelet transform for texture feature extraction followed by feature selection and similarity matching for retrieval of leaf images which matches the query image. Thus the performance is analyzed in terms of precision and recall values. This particular proposed method may find implementation in medical field of monitoring applications.