Content Based Image Retrieval based on Wavelet Transform coefficients distribution
Mathieu Lamard, Guy Cazuguel, Gwenolé Quellec et al. · Conference proceedings · 2007
In this paper we propose a content based image retrieval method for diagnosis aid in medical fields. We characterize images without extracting significant features by using distribution of coefficients obtained by building signatures from the distribution of wavelet transform. The research is carried out by computing signature distances between the query and database images. Several signatures are proposed; they use a model of wavelet coefficient distribution. To enhance results, a weighted distance between signatures is used and an adapted wavelet base is proposed. Retrieval efficiency is given for different databases including a diabetic retinopathy, a mammography and a face database. Results are promising: the retrieval efficiency is higher than 95% for some cases using an optimization process.
Cited by 55 works. doi:10.1109/iembs.2007.4353347
Authors
References
- A Wavelet Tour of Signal Processing
- Content-based image retrieval at the end of the early years
- Multifrequency channel decompositions of images and wavelet models
- The Lifting Scheme: A Custom-Design Construction of Biorthogonal Wavelets
- Parameterisation of a stochastic model for human face identification
- Proposed international clinical diabetic retinopathy and diabetic macular edema disease severity scales
- Texture classification by wavelet packet signatures
- Current Status of the Digital Database for Screening Mammography
- Two-channel perfect-reconstruction FIR QMF structures which yield linear-phase analysis and synthesis filters
- Statistical texture characterization from discrete wavelet representations