2. Objective
• The objective of image retrieval system is to retrieve the similar
images compared to a query image in a most effective and efficient
way.
3. Abstract
• Adaptive wavelet-based image characterizations have been proposed for content-
based image retrieval (CBIR) applications which can be used to characterize each
query image. In these applications, the same Adaptive wavelet basis was tuned to
maximize the retrieval performance in a training data set. I take it one step further
in this project, a different wavelet basis is used to characterize each query image.
A regression function, which is tuned to maximize the retrieval performance in the
training data set, is used to estimate the best wavelet filter.
• The proposed method to compute this image characterization almost instantly
for every possible separable or non-separable wavelet filter.Therefore, using a
different wavelet basis for each query image does not considerably increase
computation times. On the other hand, significant retrieval performance increases
were obtained in a medical image data set, a texture data set, a face recognition
data set, and an object picture data set.
4. Existing method
• Image shape, histogram orientation and edge analysis.
• Image retrieval based on text.
Disadvantages
• Bad results due to the semantic gap and the subjectivity of human
perception.
• Will not suitable for medical image applications.
5. Proposed method
• Content-based image retrieval,
• curvelet transformation.
Advantages
• This method suitable for medical image applications.
• Low time computation.
• Semantic gap provides efficient texture analysis.
7. Modules Description
• Texture Extraction
The main goal of using Haar wavelet is to achieve space frequency
localization. wavelet is a small wave which is used to analyze wavelet
transformation. It is a tool used for decomposition of an image and to
compute frequency domain by using the spatial domain of an image.
• Color Feature Extraction
The color feature using Haar discrete wavelet transform (DWT).It is a
tool used for decomposition of an image and to compute frequency
domain by using the spatial domain of an image.
18. Future Enhancement
• The work presented in this thesis can be extended in several directions.The denoising
algorithms can be extended for analyzing the images affected with salt and pepper noise ,
speckle noise and other noise models.
• The advantage of the DWT is in its flexibility caused by the choice of various wavelet functions
improves PSNR values and produce visually pleasing images.
• The algorithms can be extended in such a way that the way of reducing the noise from an image
without losing the actual and important information.
• Some image processing applications require an accurate determination of object boundary. In
future research, the present algorithms can be extended with the help of morphological
operations to extract the boundary of the medical objects which in turn useful in image
topology.
• Implementation of HWT can be improved by reducing the approximation errors obtained with
HilbertTransform. A thorough research can be done to extend HWT to Diversity Enhanced HWT
(DEHWT).
19. Conclusion
• The primary goal of the proposed system is to design a content
based image retrieval system that should be simple to use, easy to
handle very large image databases with different image category
models, and fastest to retrieve images using primitive features such
as color and texture, which are semantically related to the
image. The proposed system focused on the similarity between
query image and database images rather than the exact match.