STM Journals Peer-reviewed research from Consortium e-Learning Network

Journal of Image Processing & Pattern Recognition Progress

Performance Evaluation of Segmentation Algorithms for Apple Fruit Grading

  1. Aarti Abasaheb Sawant
  2. V. N. Kshirsagar

Abstract

For supplying high-quality food products within a short time, automated grading of fruits is getting special attention. Fruit grading technique will grade the fruits using their shapes, color and outer look. This emphasizes the necessity of image segmentation which extracts important features from an image. This paper consists of analysis of different segmentation algorithms for detecting damage part of fruit. Marker-based watershed algorithm uses internal and external markers to locate catchment basins which are regions of interest. Quadtree and merge algorithm divides an image within a complete tree representation and then merges the regions based on some criteria. Seed region growing algorithm selects the seed point and grows the region according to pre-defined criteria. The comparison of the present approach in terms of the measures like energy, discrete entropy, relative entropy, mutual information, normalized mutual information and redundancy is also carried out.

Keywords

Support