A Computational Approach to Edge Detection
IEEE Transactions on Pattern Analysis and Machine Intelligence
Digital image processing (2nd ed.)
Digital image processing (2nd ed.)
Swarm intelligence: from natural to artificial systems
Swarm intelligence: from natural to artificial systems
Unsupervised Image Segmentation Using a Colony of Cooperating Ants
BMCV '02 Proceedings of the Second International Workshop on Biologically Motivated Computer Vision
Edge detection using ant algorithms
Soft Computing - A Fusion of Foundations, Methodologies and Applications
Image Thresholding Using Ant Colony Optimization
CRV '06 Proceedings of the The 3rd Canadian Conference on Computer and Robot Vision
Ant colony system: a cooperative learning approach to the traveling salesman problem
IEEE Transactions on Evolutionary Computation
Plant texture classification using gabor co-occurrences
ISVC'10 Proceedings of the 6th international conference on Advances in visual computing - Volume Part II
Review: Plant species identification using digital morphometrics: A review
Expert Systems with Applications: An International Journal
Ant algorithms for image feature extraction
Expert Systems with Applications: An International Journal
Hi-index | 0.00 |
This paper presents preliminary results on an investigation into using artificial swarms to extract and quantify features in digital images. An ant algorithm has been developed to automatically extract the outlines and primary venation patterns from digital images of living leaf specimens via an edge detection method. A qualitative and quantitative analysis of the results is carried out herein. The artificial swarms are shown to converge onto the edges within the leaf images and statistical accuracy, as measured against ground truth images, is shown to increase in accordance with the swarm convergence. Visual results are promising, however limitations due to background noise need to be addressed for the given application. The findings in this study present potential for increased robustness in using swarm based methods, by exploiting their stigmergic behaviour to reduce the need for parameter fine-tuning with respect to individual image characteristics.