An autoassociator for automatic texture feature extraction
Siddhivinayak Kulkarni,Brijesh Verma +1 more
- pp 328-332
Reads0
Chats0
TLDR
The results and analysis showed that the autoassociator is capable of extracting texture features better than the other traditional techniques.Abstract:
This paper presents an autoassociator neural network for texture feature extraction. Texture features are extracted through the hidden layer of an autoassociator. The Resilient Propagation (RP) algorithm was employed to train the autoassociator with the texture input and output patterns. The performance of the feature extractor was evaluated on Brodatz benchmark database. A detail analysis of the results is included. The results and analysis showed that the autoassociator is capable of extracting texture features better than the other traditional techniques.read more
Citations
More filters
Patent
Cognitive memory and auto-associative neural network based search engine for computer and network located images and photographs
TL;DR: In this paper, a pre-processing query pattern is used to establish relationship between query pattern and sought stored pattern, to locate sought pattern, and to retrieve it and ancillary data.
Proceedings ArticleDOI
A texture extraction technique using 2D-DFT and Hamming distance
TL;DR: A novel technique for texture extraction and classification uses 2D-DFT transformation and a combination of this technique and a Hamming Distance based neural network for classification of extracted features is investigated.
Proceedings ArticleDOI
Unsupervised clustering of texture features using SOM and Fourier transform
TL;DR: This paper presents a novel technique for texture feature extraction that uses 2D-DFT transform and self-organizing map and compares its performance with a number of other existing techniques using a benchmark image database.
Journal Article
Generic Multimedia Database Architecture
TL;DR: In this article, a generic architecture for Multimedia Database is proposed, which includes Semantic Libraries to reduce semantic gap, levels of feature extraction for more specific and detailed feature extraction according to classes defined by prior level, and merging of two types of queries i.e. text and QBE for more accurate yet detailed results.
Book ChapterDOI
Risk-Based Neuro-Grid Architecture for Multimodal Biometrics
TL;DR: This research indicates that multimodal biometrics is the way forward for a highly reliable adoption of biometric identification systems in various applications, such as banks, businesses, government and even home environments, but such systems would require large distributed datasets with multiple computational realms spanning organisational boundaries and individual privacies.
References
More filters
Journal ArticleDOI
Statistical and structural approaches to texture
TL;DR: This survey reviews the image processing literature on the various approaches and models investigators have used for texture, including statistical approaches of autocorrelation function, optical transforms, digital transforms, textural edgeness, structural element, gray tone cooccurrence, run lengths, and autoregressive models.
Journal ArticleDOI
Texture features for browsing and retrieval of image data
B.S. Manjunath,Wei-Ying Ma +1 more
TL;DR: Comparisons with other multiresolution texture features using the Brodatz texture database indicate that the Gabor features provide the best pattern retrieval accuracy.
Journal ArticleDOI
Textural Features Corresponding to Visual Perception
TL;DR: The discrepancies between human vision and computerized techniques that are encountered in this study indicate fundamental problems in digital analysis of textures and could be overcome by analyzing their causes and using more sophisticated techniques.
Journal ArticleDOI
Filtering for texture classification: a comparative study
Trygve Randen,John Hakon Husoy +1 more
TL;DR: Most major filtering approaches to texture feature extraction are reviewed and a ranking of the tested approaches based on extensive experiments is presented, showing the effect of the filtering is highlighted, keeping the local energy function and the classification algorithm identical for most approaches.