Image retrieval plays an important role in the Digital imaging and media such as image classification, photography, medical imaging etc., in which the obtained information is crucial for the analysis of images. Extraction of representative features is a challenge due to the variations in geometric, photometric image features. The feature fusion process affords compact discriminative features of an image; this crucial information requires in analysing images accurately to increase the accuracy. Hence, Image Retrieval using feature fusion and Glowworm Swarm Optimization (IR-FF-GSO) is proposed. Multiple features are extracted with Texture, Color, Statistical and Scale Invariant Feature Transform (SIFT) descriptors to perform retrieval process. Feature vector is fused using optimized weight value which is obtained from GSO algorithm. The proposed method yields 95.5% retrieval accuracy on ImageNet database and is accurate compared to the conventional image retrieval method by over 10% [1].