Content-based Image Retrieval Using Gabor-Zernike Features

Suitability of Shape and Texture Features in Retrieval of Medicinal Plants’ Images in Indian Context

“…Two dimensional gabor filter is proved to be very useful texture analysis and is widely used in literature.X.Fu and Y.Li et al [14] have demonstrated gabor filter(GF) and zernike moments(ZM) feature extraction technique on face, fingerprint and MPEG-7 database. ZM is effective for face and MPEG-7 database.GF is effective for face and fingerprint database but weak for MPEG-7 shape database.…”

“…Basically, Gabor filters are a group of wavelets, with each wavelet capturing energy at a specific frequency and a specific direction [14]. Expanding a signal using this basis provides a localized frequency description, therefore capturing local features/energy of the signal.…”

“…In terms of these, Zernike moments [14] are the one which gives a more robust to changes caused by image rotation, translation and scaling. Zernike moments have been used on MPEG-7 region shape DB CE-2 database which consists of 3621 binary shapes of trademarks.…”

“…Zernike moments (ZM) [14] are one of the most popular shape descriptors and have many desirable properties, such as rotation invariance, robustness to noise, expression efficiency. Hence in this paper we have used Zernike moments for Helpful Hints.extracting the plant shape with n = 4 (moment order) and m = 4 (repetition).The complex ZM are derived from Zernike polynomials which are a set of complex, orthogonal polynomials defined over the interior of a unit circle …”

“…In addition to shape representation, texture features are also used for content based medicinal plant image retrieval. We combined gabor texture features [7] and zernike features [14].…”