一种水果分拣方法、装置、系统、终端及存储介质

本发明公开了一种水果分拣方法、装置、系统、终端及存储介质。该方法包括：获取待分拣水果的图像；将待分拣水果的图像输入水果分类模型，得到待分拣水果的目标分类标签，水果分类模型基于第一卷积神经网络训练生成；将待分拣水果的图像输入与待分拣水果的目标分类标签对应的水果分级模型，得到待分拣水果的目标等级标签，水果分级模型基于第二卷积神经网络训练生成；将待分拣水果的目标分类标签和目标等级标签发送给目标机器人，以使目标机器人根据待分拣水果的目标分类标签和目标等级标签对待分拣水果进行分拣。本发明实施例采用基于卷积神经网络训练生成的水果分类模型和水果分级模型，实现了自动提取图像特征，提高了水果分拣的效率和准确率。

The invention discloses a fruit sorting method, device and system, a terminal and a storage medium. The method comprises the following steps: acquiring an image of a to-be-sorted fruit; inputting theimage of the to-be-sorted fruit into a fruit classification model to obtain a target classification label of the to-be-sorted fruit, the fruit classification model being generated based on training ofa first convolutional neural network; inputting the image of the to-be-sorted fruit into a fruit classification model corresponding to the target classification label of the to-be-sorted fruit to obtain the target classification label of the to-be-sorted fruit, and training and generating the fruit classification model based on a second convolutional neural network; and sending the target classification label and the target grade label of the to-be-sorted fruit to the target robot, so that the target robot sorts the to-be-sorted fruit according to the target classification label and the target grade label of the to-be-sorted fruit. According to the embodiment of the invention, the fruit classification model and the fruit classification model generated by training based on the convolutional neural network are adopted, so that automatic image feature extraction is realized, and the fruit sorting efficiency and accuracy are improved.