Abstract: Increasing use of World Wide Web and communication channels like mobile networking has increased the number of images used throughout the world. Continuing advancements in both hardware and software coupled with higher image processing and image vision tools, have made it possible to store huge amount of images. This increase in number of images and image databases has necessitated the need for image mining. The extension of data mining into the image domain is known as image mining. It is an interdisciplinary endeavour that draws upon expertise in computer vision, image processing, image retrieval, data mining, machine learning, database, and artificial intelligence. Rather than the development of many algorithms and applications in various research fields, image mining is still rarely untouched area. It mainly comprises of faster image retrieval and quality of the retrieved image. The extraction of implicit knowledge, image data relationship, and similar type of patterns may be the possible candidature to speed up the process of image retrieval. The proper combination and parameterization of these attributes can help out to retrieve better images at short point in time. This paper incorporates such two algorithms namely- hierarchical and k-means to have a good quality image retrieval at efficient pace.