Coffee fruit maturity identification method based on deep learning

Abstract

The invention discloses a coffee fruit maturity identification method based on deep learning, and belongs to the field of intelligent agriculture, and the method comprises the steps: collecting original coffee fruit images with different maturity degrees, cutting the images into a uniform size, and processing the image into a form required by neural network model input; sending the training sampledata into a convolutional neural network model composed of a continuous convolutional layer and a sparse activation Maxout function layer to extract coffee fruit image features; and training a coffeefruit maturity classifier according to the extracted features to obtain a final recognition model. According to the invention, whether coffee fruits are ripe or not can be automatically identified without manual participation, so that the identification accuracy and efficiency are improved, and the labor cost can be reduced.

Description

technical field [0001] The invention belongs to the field of intelligent agriculture, and more specifically designs a method for identifying the ripeness of coffee fruit by opportunity deep learning. Background technique [0002] With the rapid development of artificial intelligence and its great effectiveness in the industrial field, agriculture has seen new opportunities for change, and the application of deep learning technology in the agricultural field has broad prospects for development. [0003] The traditional identification method of coffee fruit ripeness is mainly through visual observation, which is not only time-consuming and laborious, but also highly dependent on the experience accumulated by farmers in the agricultural production process, and it is also very likely to make mistakes, thus missing coffee The best time for fruit harvesting, to obtain high-quality coffee beans, has extremely high professional knowledge requirements for agricultural producers. The...

AI Technical Summary

Problems solved by technology

[0003] The traditional identification method of coffee fruit ripeness is mainly through visual observation, which is not only time-consuming and laborious, but also highly dependent on the experience accumulated by farmers in the agricultural production process, and it is also very likely to make mistakes, thus missing coffee. The best time to harvest the fruit, to obtain high-quality coffee beans, has extremely high professional knowledge requirements for agricultural producers

Images