Detection of Red Grape Clusters Stands using Image Processing and Neural Network in Order to Harvest the Grapes in the Application Robot

Subject Areas :

Received: 2017-01-02 Accepted : 2017-01-30 Published : 2017-05-22

Keywords: image processing, Machine Vision, Artificial Intelligence, Grape, Red,

Abstract :

Detection of Red Grape Clusters Stands using Image Processing and Neural Network in Order to Harvest the Grapes in the Application Robot  Akbar Nazari In this study an automatic classification system based on machine vision and artificial neural networks based on color and its components were developed. about 300 grape gardens of the city of Urmia digit image in various conditions from early morning until evening in cloudy weather and sunny, of the prspetron neural network learning algorithm with multiple layers of error propagation as one of the tools of artificial intelligence for the detection and isolation of the image class. the main component of the input network, the average color (R,G,B) pixel images, and at the output of the network is the number one (1) red grape clusters or as the number zero (0) as the non-red grape (leaves, branches and trunk, the sky) was considered finally, after trial and error learning algorithm of neural network and the number of neurons, It became clear that the number of neurons in the hidden layer and 13 a hidden layer and the learning algorithm of the type Trainlm and sigmoid transfer function, neural network with 98% capable to detect and isolate the image class. The results of image analysis suggests the existence of a high accuracy in the classification section of the image. Key words: Grape, Red, Image Processing, Machine Vision, Artificial Intelligence