An effective content-based visual image retrieval system
read more
Citations
A mobile butterfly-watching learning system for supporting independent learning
An Adaptive Color Image Segmentation
Realizing Outdoor Independent Learning with a Butterfly-Watching Mobile Learning System
The classification of style in fine-art painting
Seamless channel transition for the staircase video broadcasting scheme
References
Query by image and video content: the QBIC system
Image Retrieval
VisualSEEk: a fully automated content-based image query system
Photobook: content-based manipulation of image databases
Color image segmentation: advances and prospects
Related Papers (5)
Frequently Asked Questions (16)
Q2. What are the future works mentioned in the paper "An effective content-based visual image retrieval system" ?
In the future, the authors plan to add more images to their image database. Wavelet-based texture feature and the SPCPE algorithm with more classes will be integrated into their current system. Machine learning algorithms will also be considered to improve the retrieval precision.
Q3. What is the color feature extracted in the wavelet technique?
The color feature extracted in their system is a 13- dimension color label histogram vector, and the spatial feature is two 4-dimensional class parameter vectors.
Q4. What is the l2 distance between two color label histograms?
The L2-Distance (also called Euclidian Distance) is used to compare the class parameter vectors because the parameters in each class are assumed to be independent.
Q5. How many images pass the color filter?
The color filter effectively eliminates around eighty percent of the images in the database from the search range for the later stage.
Q6. What is the common method of color histogram?
Color histogram is the most widely used method since it is more robust to changes due to scaling, orientation, perspective, and occlusion of images [2].
Q7. Why is the color feature the widely used visual feature in image retrieval?
The color feature is the most widely used visual feature in image retrieval because it ismore robust to changes due to scaling, orientation, perspective and occlusion of images [2].
Q8. What is the main idea of the paper?
In this paper, the authors propose an effective content-based image retrieval system that consists of the visual content extraction and indexing component and the query engine component.
Q9. What is the idea of filtering in the query engine?
The query engine in their system employs the idea of filtering to reduce the search ranges at different stages so as to speed up the query processing.
Q10. Why is it the commonly used color space?
RGB is the most commonly used color space primarily because color image acquisition and recording hardware are designed for this space.
Q11. How many images are in the image database?
The image database in their current system contains 500 images that are downloaded from yahoo (www.yahoo.com) and corbis (www.corbis.com).
Q12. What are the images with objects in the sky?
There are images with objects in the sky, in the water or ocean, or on the grass, images with green trees or plants, images with mountains under different time periods (daytime, sunset and nighttime) or different weather situations (cloudy and sunny), etc.
Q13. What are the twelve categories of the color histogram?
These twelve categories are black, white, red, bright red, yellow, bright yellow, green, bright green, blue, bright blue, purple, and bright purple.
Q14. What is the definition of color coherence vector?
In the performance evaluation experiments in [2], it is shown that the color histogram runs much faster than the color coherence vector and color autocorrelogram, performs almost as good as the color coherence vector, and does not perform much worse than the best color autocorrelogram.
Q15. What is the a posterior probability of the partition variable and class parameter variable?
For each wavelet subband, a candidate initial partition is generated by labeling all pixels in the original image that correspond to the salient points in that subband as one class and the rest of the pixels as the other class.
Q16. What is the new category for the color histogram?
a new category “gray” is added to consider all possible value ranges since some images in their image database contain the gray color.