Identification of Orchid Species Using Content-Based Flower Image Retrieval

TL;DR

This paper presents a system for orchid species identification using content-based image retrieval, combining shape and color features with SVM, achieving up to 85.33% accuracy.

Contribution

It introduces a novel feature extraction approach focusing on both flower and lip regions, enhancing retrieval accuracy for orchid species.

Findings

System accuracy reaches 85.33% in validation.

Feature combination improves retrieval performance.

Lip region features contribute significantly to accuracy.

Abstract

In this paper, we developed the system for recognizing the orchid species by using the images of flower. We used MSRM (Maximal Similarity based on Region Merging) method for segmenting the flower object from the background and extracting the shape feature such as the distance from the edge to the centroid point of the flower, aspect ratio, roundness, moment invariant, fractal dimension and also extract color feature. We used HSV color feature with ignoring the V value. To retrieve the image, we used Support Vector Machine (SVM) method. Orchid is a unique flower. It has a part of flower called lip (labellum) that distinguishes it from other flowers even from other types of orchids. Thus, in this paper, we proposed to do feature extraction not only on flower region but also on lip (labellum) region. The result shows that our proposed method can increase the accuracy value of content based flower image retrieval for orchid species up to $\pm$ 14%. The most dominant feature is Centroid Contour Distance, Moment Invariant and HSV Color. The system accuracy is 85,33% in validation phase and 79,33% in testing phase.