ShapeCN Dataset

Description

The ShapeCN is a countour dataset of leaves and fishes. The leaves dataset is composed by leaves from different plants. Leaves classification is a difficult task, as the between-class similarity is considerable while the within-class similarity is not suitable. Besides, overlaps can occur between adjacent parts of leaves. As a result, there can be major differences among boundary contours of leaves in the same category. The fishes dataset objectives to show the performance of the method on large databases composed by shapes under rotation and scale transformations.

The fish contours database is composed of 10 different manifestations (5 rotations and 5 scales) of each one of the 1100 fish contours obtained from the dataset available at http://www.ee.surrey.ac.uk/cvssp/demos/css/demo.html . So, the database is composed of 11000 contours grouped into 1100 classes of 10 samples each.

Dataset name meaning:

c s - contour scaled

c r - contour rotated

c d - contour degrated

c n - contour noise

c - contour original

How to cite this dataset

Results

A complex network-based approach for boundary shape analysis

Leaves original --- 83.67%

Leaves rotated --- 83.89%

Leaves scaled --- 84.12%

Fishes rotate and scaled --- 99.38%

Papers using ShapeCN dataset

If you have a work that use this dataset, please send a message informing the reference of your paper and also the results obtained. (bruno at ifsc.usp.br)

Dataset snapshot example

Leaves contours dataset




Fishes contours rotated and scaled dataset