The explosive growth of image data leads to the research and development of Content-Based Image Retrieval (CBIR) systems. CBIR systems extract and retrieve images by their low-level features, such as color, texture, and shape. However, these visual contents do not allow users to query images by semantic meanings. Image annotation systems, a solution to solve the inadequacy of CBIR systems, aim at automatically annotating image with some controlled keywords. Machine learning techniques are used to develop the image annotation systems to map the low-level (visual) features to high-level concepts or semantics. This paper reviews 50 image annotation systems using supervised machine learning techniques to annotate images for image retrieval. Future research issues are provided.