Image Retrieval Through Free-Form Query using Intelligent Text Processing

Image Retrieval is the process of retrieving images from image/multimedia databases. are carried out with various types queries, free-form query a text-query that consists single or multiple keywords and/or concepts descriptions without inclusion wild-card characters punctuations. This work aims to handle image retrieval based on text queries. Simple & complex queries conceptual explored and an intelligent processing system Bag-of-Words model modified built for natural scene Diverse Social Images using Damerau-Levenshtein edit distance measure. The efficacy proposed evaluated by testing 1500 has resulted in recall accuracy 91.3% (of Wang/Corel database) 100% DIV400 dataset). These results show produced satisfactory performance compared published such as harmonic mean precision (i.e. F1-Score) 76.70% 63.32% at 20 etc reported works.