A Method of Fruits Recognition Based on SIFT Characteristics Matching

SIFT (scale invariant feature transform) used in fruits recognition presents a characteristics matching algorithm according to fruits images. The algorithm uses SIFT characteristics as matching feature, then introduces Euclidean distance as the similarity metrics of image matching, and uses a method of setting a threshold value to delete the false matching points. The experimental results prove that the algorithm can effectively solve the problem of fruits recognition and matching with translation, rotation, scaling, and partial occlusion.

1. Introduction

Currently, fruits sorting mostly depend on manual method which results in low efficiency, and annually lots of labor and finance is consumed. During the I970s machine vision has been widely used in the field of varietal recognition because of the specialization of image processing, lower cost and higher speed of computer and its many features. Machine vision system is mainly used for obtaining information of the working environment and performs detection and orientation of operation target. People have presented various methods for image recognition. Image matching is one of the most widely used methods.