NOVEL CBIR System Using Spark MAP-Reduce with a Firefly Macqueen's K-Means Clustering Algorithm

Nowadays content-based image retrieval (CBIR) has been an active research area, demonstrating a feasible solution for recovering similar images from an image mine. The amount of digital images increases exponentially day by. Storage requirements for these images may increase from Gigabytes to Petabytes. Searching and retrieving the relevant images from such a large volume of image datasets, based on their contents, play a dynamic role in various applications of computer vision. The time taken to retrieve the images is more, and the accuracy of retrieved images is less in the existing systems. The limitation of Hadoop map-reduce is the lack of performing real-time tasks efficiently. A proficient content-based image retrieval framework based on Spark Map-Reduce with a Firefly MacQueen's k-means clustering (FMKC) algorithm and Bag of visual word (BoVW) is proposed to achieve high accuracy for big data. The Apache spark programming can be used to productively recover pictures with less retrieval time and retrieve the accurate images from the big database that resembles the query image. The experimental results demonstrate that the method proposed in our work outperforms the state-of-the-art methods in terms of accuracy of the retrieved images and average retrieval time. The proposed system is 93% accurate, and it is easier to retrieve the images from the large database.