A Comprehensive Survey on Image Search Using Binary Hash Codes

Abstract: Effective content based Image Search based on hash codes is a very acceptable for efficient similarity search, due to its query time and storage efficiency. In binary hashing technique, first the high-dimensional visual features are extracted from images, then these extracted features are embedded into Hamming space and the distance or similarity of two points are approximately calculated by the Hamming distance between their binary codes. Though the Hamming distance calculation is efficient in practice, a query result is ambiguous as it returns multiple results sharing the same Hamming distance and poses a critical issue for similarity search where ranking is important. This paper presents a survey of various states-of-the-art- hashing techniques that allows faster visual similarity search in hamming space. And also discussed concept of weighted Hamming distance, which improve ranking performance of binary hash code so that result images can be ranked at fined grained level.

Keywords: Feature Extraction, Approximate Nearest Neighbour Search, Binary Hash Codes, Weighted Hamming Distance.