A Research on the Fruit Recognition Algorithm Based on the Multi-Feature Fusion

In order to improve the accuracy of recognizing fruits, this paper proposes a fruit recognition algorithm based on color features and shape features. Firstly, the fruit image is preprocessed to obtain the target area of the image. Then, the HSV color model is chosen to analyze its histogram and extract its color features. After this, the contour method is used to extract the shape features of the fruit. Finally, the victor of color features and one of shape features are optimized and combined, and the BP neural network is used to train and classify the test samples. After testing, comparing and analyzing the different features of the input layer respectively, the research result shows that the classification and recognition rate of this algorithm is above 95%, which is higher than that of the single feature algorithm.