A self-organizing map with dynamic architecture for efficient color quantization

Kirk, J.S.; Chang, Dar-Jen; Żurada, Jacek M.

doi:10.1109/ijcnn.2001.938495

Cited by 6 publications

(3 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The inflexible structure of a fixed neuron number in the output layer has been redesigned recently [3,6]. Through a dynamically growing mechanism, the number of neurons can be adjusted correspondingly.…”

Section: Utilising the Self-organising Map For Colour Quantisationmentioning

confidence: 99%

See 1 more Smart Citation

Adaptive and efficient colour quantisation based on a growing self-organising map

2012

View full text Add to dashboard Cite

Studies on colour quantisation have indicated that its applications range from the relaxation of displaying hardware constraints in early years to a modern usage of facilitating content-based image retrieval tasks. Among many alternatives, approaches based on neural network models are generally accepted to be able to produce quality results in colour quantisation. However, these approaches using n quantised neurons require O(n) for a full search strategy, which is inefficient when n becomes large. In view of this, we propose to incorporate a growing quadtree structure into a self-organising map (GQSOM) which reaches a search time O(log n). Specifically, the strategy of inheriting from parent neurons hierarchically facilitates a much more efficient and flexible learning process. Both theoretical and empirical studies have shown that our approach is adaptive in determining an appropriate number of quantised colours, and the performance is significantly improved without compromise of the quantisation quality.

show abstract

Section: Utilising the Self-organising Map For Colour Quantisationmentioning

confidence: 99%

“…In previous approaches [1,[3][4][5][6][7][8], the number of colours to retain during the colour quantisation process is usually required to be predetermined. Nevertheless, it can be easily understood that the appropriate number of quantised colours varies with different image contents.…”

Section: Introductionmentioning

confidence: 99%

Adaptive and efficient colour quantisation based on a growing self-organising map

2012

View full text Add to dashboard Cite

show abstract

“…Unsupervised clustering methods like SOM are frequently used for image classification, and the SOM has proved to be especially convenient for this task due to its 2D mapping capabilities making the resulting clusters easy to visualize. In SOMs, the neuron's weight often acts as a representation of the color domain such as RGB in vector format, and is altered during the training period to more closely match randomly selected pixels from the input image, "mapping" the SOM's topology to form a 2D representation of a higher dimensional vector space [3]. As in image classification the input data is represented as a composite high dimensional feature vector with variable component numbers in various individual features.…”

Section: Related Workmentioning

confidence: 99%

Layered Self-Organizing Map for Image Classification in Unrestricted Domains

O’Connell

Kutics

Nakagawa

2013

Image Analysis and Processing – ICIAP 2013

View full text Add to dashboard Cite

The inherent difficultly in unrestricted image domain classification is due to the many different features exhibited by images. Efforts made toward classification of abstract features tend to focus on a single attribute. Without a method of unifying descriptors, it becomes very difficult to perform multifeature analysis. Extending the concept of the Self-Organizing Feature Map to include multiple competitive layers, it has been possible to create a new type of Artificial Neural Network capable of analyzing image and signal datasets with multiple feature descriptors concurrently in a powerful yet computationally light manner. Compared to standard CBIR retrieval approach, a marked increase in the precision of clustering of 13 points has been achieved, along with a reduction in computation time.

show abstract

Exploiting a Growing Self-organizing Map for Adaptive and Efficient Color Quantization

Chang

Teng

2009

Advances in Multimedia Information Processing - PCM 2009

View full text Add to dashboard Cite

A self-organizing map with dynamic architecture for efficient color quantization

Cited by 6 publications

References 4 publications

Adaptive and efficient colour quantisation based on a growing self-organising map

Adaptive and efficient colour quantisation based on a growing self-organising map

Layered Self-Organizing Map for Image Classification in Unrestricted Domains

Exploiting a Growing Self-organizing Map for Adaptive and Efficient Color Quantization

Contact Info

Product

Resources

About