This article presents a Content Based Image Retrieval (CBIR) system based on SURF-ORB-AdaBoost methodology. Content-based image retrieval is a process that applies computer vision approaches for seeking and overseeing extensive image collections more efficiently. With the development of expansive digital image collections activated by fast advances in electronic capacity limit and processing power, there is a developing requirement for devices and computer systems to help productive browsing, searching, and retrieval for image collections. Hence, the aim of this article is to build up a content-based image retrieval system. In this article, the authors presented a combination of SURF and ORB features for image retrieval from a large image collection. Initially, SURF (Speeded Up Robust Features) and ORB (Oriented Fast Rotated and BRIEF) features are extracted from the given query image. Subsequently, K-Means clustering algorithm is used to analyze the data and the LPP dimensionality reduction method is used to reduce the space complexity of the system and increase the performance of the system. After that, classifier is applied to extract the relevant image. A precision rate of 97.8% has been reported using the proposed CBIR system for the Wang image dataset.