Associating visual textures with human perceptions using genetic algorithms
Information Sciences: an International Journal
Content based retrieval and classification of cultural relic images
ISNN'05 Proceedings of the Second international conference on Advances in neural networks - Volume Part II
Content-based image retrieval of skin lesions by evolutionary feature synthesis
EvoApplicatons'10 Proceedings of the 2010 international conference on Applications of Evolutionary Computation - Volume Part I
A query-by-example content-based image retrieval system of non-melanoma skin lesions
MCBR-CDS'09 Proceedings of the First MICCAI international conference on Medical Content-Based Retrieval for Clinical Decision Support
Methodological review: Computerized analysis of pigmented skin lesions: A review
Artificial Intelligence in Medicine
| Hi-index | 0.00 |
In this work, we develop a system for retrieving medicalimages with focus objects incorporating models of humanperception. The approach is to guide the search for anoptimum similarity function using human perception. First,the images are segmented using an automatedsegmentation tool. Then, 20 shape features are computedfrom each image to obtain a feature matrix. Principalcomponent analysis is performed on this matrix to reducethe number of dimensions. Principal components obtainedfrom the analysis are used to select a subset of variablesthat best represents the data. A human perception ofsimilarity experiment is designed to obtain an aggregatedhuman response matrix. Finally, an optimum weightedManhattan distance function is designed using a geneticalgorithm utilizing the Mantel test as a fitness function. Thesystem is tested for content-based retrieval of skin lesionimages. The results show significant agreement betweenthe computer assessment and human perception ofsimilarity. Since the features extracted are not specific toskin lesion images, the system can be used to retrieve othertypes of images.